TY  - JOUR
AU  - Mitić, Marko
AU  - Vuković, Najdan
AU  - Petrović, Milica
AU  - Miljković, Zoran
PY  - 2018
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/2880
AB  - Most of today's mobile robots operate in controlled environments prone to various unpredictable conditions. Programming or reprogramming of such systems is time-consuming and requires significant efforts by number of experts. One of the solutions to this problem is to enable the robot to learn from human teacher through demonstrations or observations. This paper presents novel approach that integrates Learning from Demonstrations methodology and chaotic bioinspired optimization algorithms for reproduction of desired motion trajectories. Demonstrations of the different trajectories to reproduce are gathered by human teacher while teleoperating the mobile robot in working environment. The learning (optimization) goal is to produce such sequence of mobile robot actuator commands that generate minimal error in the final robot pose. Four different chaotic methods are implemented, namely chaotic Bat Algorithm, chaotic Firefly Algorithm, chaotic Accelerated Particle Swarm Optimization and newly developed chaotic Grey Wolf Optimizer (CGWO). In order to determine the best map for CGWO, this algorithm is tested on ten benchmark problems using ten well-known chaotic maps. Simulations compare aforementioned algorithms in reproduction of two complex motion trajectories with different length and shape. Moreover, these tests include variation of population in swarm and demonstration examples. Real-world experiment on a nonholonomic mobile robot in indoor environment proves the applicability of the proposed approach.
PB  - Springer London Ltd, London
T2  - Neural Computing & Applications
T1  - Chaotic metaheuristic algorithms for learning and reproduction of robot motion trajectories
EP  - 1083
IS  - 4
SP  - 1065
VL  - 30
DO  - 10.1007/s00521-016-2717-6
ER  - 

@article{
author = "Mitić, Marko and Vuković, Najdan and Petrović, Milica and Miljković, Zoran",
year = "2018",
abstract = "Most of today's mobile robots operate in controlled environments prone to various unpredictable conditions. Programming or reprogramming of such systems is time-consuming and requires significant efforts by number of experts. One of the solutions to this problem is to enable the robot to learn from human teacher through demonstrations or observations. This paper presents novel approach that integrates Learning from Demonstrations methodology and chaotic bioinspired optimization algorithms for reproduction of desired motion trajectories. Demonstrations of the different trajectories to reproduce are gathered by human teacher while teleoperating the mobile robot in working environment. The learning (optimization) goal is to produce such sequence of mobile robot actuator commands that generate minimal error in the final robot pose. Four different chaotic methods are implemented, namely chaotic Bat Algorithm, chaotic Firefly Algorithm, chaotic Accelerated Particle Swarm Optimization and newly developed chaotic Grey Wolf Optimizer (CGWO). In order to determine the best map for CGWO, this algorithm is tested on ten benchmark problems using ten well-known chaotic maps. Simulations compare aforementioned algorithms in reproduction of two complex motion trajectories with different length and shape. Moreover, these tests include variation of population in swarm and demonstration examples. Real-world experiment on a nonholonomic mobile robot in indoor environment proves the applicability of the proposed approach.",
publisher = "Springer London Ltd, London",
journal = "Neural Computing & Applications",
title = "Chaotic metaheuristic algorithms for learning and reproduction of robot motion trajectories",
pages = "1083-1065",
number = "4",
volume = "30",
doi = "10.1007/s00521-016-2717-6"
}

Mitić, M., Vuković, N., Petrović, M.,& Miljković, Z.. (2018). Chaotic metaheuristic algorithms for learning and reproduction of robot motion trajectories. in Neural Computing & Applications
Springer London Ltd, London., 30(4), 1065-1083.
https://doi.org/10.1007/s00521-016-2717-6

Mitić M, Vuković N, Petrović M, Miljković Z. Chaotic metaheuristic algorithms for learning and reproduction of robot motion trajectories. in Neural Computing & Applications. 2018;30(4):1065-1083.
doi:10.1007/s00521-016-2717-6 .

Mitić, Marko, Vuković, Najdan, Petrović, Milica, Miljković, Zoran, "Chaotic metaheuristic algorithms for learning and reproduction of robot motion trajectories" in Neural Computing & Applications, 30, no. 4 (2018):1065-1083,
https://doi.org/10.1007/s00521-016-2717-6 . .

TY  - JOUR
AU  - Vuković, Najdan
AU  - Petrović, Milica
AU  - Miljković, Zoran
PY  - 2018
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/2878
AB  - The Random Vector Functional Link Neural Network (RVFLNN) enables fast learning through a random selection of input weights while learning procedure determines only output weights. Unlike Extreme Learning Machines (ELM), RVFLNN exploits connection between the input layer and the output layer which means that RVFLNN are higher class of networks. Although RVFLNN has been proposed more than two decades ago (Pao, Park, Sobajic, 1994), the nonlinear expansion of the input vector into set of orthogonal functions has not been studied. The Orthogonal Polynomial Expanded Random Vector Functional Link Neural Network (OPE-RVFLNN) utilizes advantages from expansion of the input vector and random determination of the input weights. Through comprehensive experimental evaluation by using 30 UCI regression datasets, we tested four orthogonal polynomials (Chebyshev, Hermite, Laguerre and Legendre) and three activation functions (tansig, logsig, tribal). Rigorous non-parametric statistical hypotheses testing confirms two major conclusions made by Zhang and Suganthan for classification (Zhang and Suganthan, 2015) and Ren et al. for timeseries prediction (Ren, Suganthan, Srikanth, Amaratunga, 2016) in their RVFLNN papers: direct links between the input and output vectors are essential for improved network performance, and ridge regression generates significantly better network parameters than Moore-Penrose pseudoinversion. Our research shows a significant improvement of network performance when one uses tansig activation function and Chebyshev orthogonal polynomial for regression problems. Conclusions drawn from this study may be used as guidelines for OPE-RVFLNN development and implementation for regression problems.
PB  - Elsevier, Amsterdam
T2  - Applied Soft Computing
T1  - A comprehensive experimental evaluation of orthogonal polynomial expanded random vector functional link neural networks for regression
EP  - 1096
SP  - 1083
VL  - 70
DO  - 10.1016/j.asoc.2017.10.010
ER  - 

@article{
author = "Vuković, Najdan and Petrović, Milica and Miljković, Zoran",
year = "2018",
abstract = "The Random Vector Functional Link Neural Network (RVFLNN) enables fast learning through a random selection of input weights while learning procedure determines only output weights. Unlike Extreme Learning Machines (ELM), RVFLNN exploits connection between the input layer and the output layer which means that RVFLNN are higher class of networks. Although RVFLNN has been proposed more than two decades ago (Pao, Park, Sobajic, 1994), the nonlinear expansion of the input vector into set of orthogonal functions has not been studied. The Orthogonal Polynomial Expanded Random Vector Functional Link Neural Network (OPE-RVFLNN) utilizes advantages from expansion of the input vector and random determination of the input weights. Through comprehensive experimental evaluation by using 30 UCI regression datasets, we tested four orthogonal polynomials (Chebyshev, Hermite, Laguerre and Legendre) and three activation functions (tansig, logsig, tribal). Rigorous non-parametric statistical hypotheses testing confirms two major conclusions made by Zhang and Suganthan for classification (Zhang and Suganthan, 2015) and Ren et al. for timeseries prediction (Ren, Suganthan, Srikanth, Amaratunga, 2016) in their RVFLNN papers: direct links between the input and output vectors are essential for improved network performance, and ridge regression generates significantly better network parameters than Moore-Penrose pseudoinversion. Our research shows a significant improvement of network performance when one uses tansig activation function and Chebyshev orthogonal polynomial for regression problems. Conclusions drawn from this study may be used as guidelines for OPE-RVFLNN development and implementation for regression problems.",
publisher = "Elsevier, Amsterdam",
journal = "Applied Soft Computing",
title = "A comprehensive experimental evaluation of orthogonal polynomial expanded random vector functional link neural networks for regression",
pages = "1096-1083",
volume = "70",
doi = "10.1016/j.asoc.2017.10.010"
}

Vuković, N., Petrović, M.,& Miljković, Z.. (2018). A comprehensive experimental evaluation of orthogonal polynomial expanded random vector functional link neural networks for regression. in Applied Soft Computing
Elsevier, Amsterdam., 70, 1083-1096.
https://doi.org/10.1016/j.asoc.2017.10.010

Vuković N, Petrović M, Miljković Z. A comprehensive experimental evaluation of orthogonal polynomial expanded random vector functional link neural networks for regression. in Applied Soft Computing. 2018;70:1083-1096.
doi:10.1016/j.asoc.2017.10.010 .

Vuković, Najdan, Petrović, Milica, Miljković, Zoran, "A comprehensive experimental evaluation of orthogonal polynomial expanded random vector functional link neural networks for regression" in Applied Soft Computing, 70 (2018):1083-1096,
https://doi.org/10.1016/j.asoc.2017.10.010 . .

TY  - CONF
AU  - Petronijević, Jelena
AU  - Petrović, Milica
AU  - Vuković, Najdan
AU  - Mitić, Marko
AU  - Babić, Bojan
AU  - Miljković, Zoran
PY  - 2016
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/4561
AB  - Projektovanje tehnoloških procesa predstavlja određivanje postupka proizvodnje uz zadovoljenje prethodno definisanih ciljeva i ograničenja. Terminiranjem proizvodnje se na osnovu proizvodnog plana i prethodno određenih tehnoloških postupaka dodeljuju optimalni proizvodni resursi za odgovarajući vremenski period. Uvođenjem koncepta masovne kastomizacije, već ranije ključne funkcije, projektovanje i terminiranje proizvodnje, sada imaju krucijalnu ulogu u tehnološkom sistemu zbog sve većih zahteva koje se pred ove funkcije postavljaju. Rad se bavi uvođenjem koncepta multiagentnih i holon tehnoloških sistema uz pregled stanja u oblasti projektovanja tehnoloških procesa i terminiranja proizvodnje. Radom je obuhvaćen tradicionalni, sledstveni, pristup projektovanju i terminiranju, ali i integrisan prilaz problematici.
AB  - Process planning can be defined as determination of manufacturing processes by achieving its goals and constraints. Scheduling process assigns optimal manufacturing resources over time based on production plan and previously determined process plans. With the mass customization concept, previously key functions in the production, process planning and scheduling, now become crucial for satisfaction of more demanding requirements. The paper introduces the concepts of multi-agent and holonic manufacturing systems and presents state of the process planning and scheduling area of research. It gives an overview on both, sequential and integrated, process planning and scheduling.
PB  - JUPITER Asocijacija, Univerzitet u Beogradu - Mašinski fakultet
C3  - 40. JUPITER Konferencija, 36. simpozijum „NU-ROBOTI-FTS“ : Zbornik radova, Beograd, maj 2016
T1  - Мултиагентни и холон технолошки системи у пројектовању технолошких процеса и терминирању производње
T1  - Multi-agent and Holonic Manufacturing Systems for Process Plannong and Scheduling
EP  - 3.68
SP  - 3.63
UR  - https://hdl.handle.net/21.15107/rcub_machinery_4561
ER  - 

@conference{
author = "Petronijević, Jelena and Petrović, Milica and Vuković, Najdan and Mitić, Marko and Babić, Bojan and Miljković, Zoran",
year = "2016",
abstract = "Projektovanje tehnoloških procesa predstavlja određivanje postupka proizvodnje uz zadovoljenje prethodno definisanih ciljeva i ograničenja. Terminiranjem proizvodnje se na osnovu proizvodnog plana i prethodno određenih tehnoloških postupaka dodeljuju optimalni proizvodni resursi za odgovarajući vremenski period. Uvođenjem koncepta masovne kastomizacije, već ranije ključne funkcije, projektovanje i terminiranje proizvodnje, sada imaju krucijalnu ulogu u tehnološkom sistemu zbog sve većih zahteva koje se pred ove funkcije postavljaju. Rad se bavi uvođenjem koncepta multiagentnih i holon tehnoloških sistema uz pregled stanja u oblasti projektovanja tehnoloških procesa i terminiranja proizvodnje. Radom je obuhvaćen tradicionalni, sledstveni, pristup projektovanju i terminiranju, ali i integrisan prilaz problematici., Process planning can be defined as determination of manufacturing processes by achieving its goals and constraints. Scheduling process assigns optimal manufacturing resources over time based on production plan and previously determined process plans. With the mass customization concept, previously key functions in the production, process planning and scheduling, now become crucial for satisfaction of more demanding requirements. The paper introduces the concepts of multi-agent and holonic manufacturing systems and presents state of the process planning and scheduling area of research. It gives an overview on both, sequential and integrated, process planning and scheduling.",
publisher = "JUPITER Asocijacija, Univerzitet u Beogradu - Mašinski fakultet",
journal = "40. JUPITER Konferencija, 36. simpozijum „NU-ROBOTI-FTS“ : Zbornik radova, Beograd, maj 2016",
title = "Мултиагентни и холон технолошки системи у пројектовању технолошких процеса и терминирању производње, Multi-agent and Holonic Manufacturing Systems for Process Plannong and Scheduling",
pages = "3.68-3.63",
url = "https://hdl.handle.net/21.15107/rcub_machinery_4561"
}

Petronijević, J., Petrović, M., Vuković, N., Mitić, M., Babić, B.,& Miljković, Z.. (2016). Мултиагентни и холон технолошки системи у пројектовању технолошких процеса и терминирању производње. in 40. JUPITER Konferencija, 36. simpozijum „NU-ROBOTI-FTS“ : Zbornik radova, Beograd, maj 2016
JUPITER Asocijacija, Univerzitet u Beogradu - Mašinski fakultet., 3.63-3.68.
https://hdl.handle.net/21.15107/rcub_machinery_4561

Petronijević J, Petrović M, Vuković N, Mitić M, Babić B, Miljković Z. Мултиагентни и холон технолошки системи у пројектовању технолошких процеса и терминирању производње. in 40. JUPITER Konferencija, 36. simpozijum „NU-ROBOTI-FTS“ : Zbornik radova, Beograd, maj 2016. 2016;:3.63-3.68.
https://hdl.handle.net/21.15107/rcub_machinery_4561 .

Petronijević, Jelena, Petrović, Milica, Vuković, Najdan, Mitić, Marko, Babić, Bojan, Miljković, Zoran, "Мултиагентни и холон технолошки системи у пројектовању технолошких процеса и терминирању производње" in 40. JUPITER Konferencija, 36. simpozijum „NU-ROBOTI-FTS“ : Zbornik radova, Beograd, maj 2016 (2016):3.63-3.68,
https://hdl.handle.net/21.15107/rcub_machinery_4561 .

TY  - CONF
AU  - Petrović, Milica
AU  - Petronijević, Jelena
AU  - Mitić, Marko
AU  - Vuković, Najdan
AU  - Miljković, Zoran
AU  - Babić, Bojan
PY  - 2016
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/4560
AB  - U radu je prikazan pristup za integrisano projektovanje i terminiranje fleksibilnih teholoških procesa obrade delova primenom algoritma baziranog na inteligenciji roja čestica i teoriji haosa (cPSO algoritam). Pored metoda kodiranja/dekodiranja parametara planova terminiranja u jedinke cPSO algoritma, u radu je predložen matematički model za minimizaciju ukupnog vremena za obradu svih delova čije se terminiranje vrši, maksimizaciju uravnoteženog iskorišćenja mašina alatki i minimizaciju transportnih tokova materijala. Takođe, u cilju prevazilaženja nedostataka vezanih za brzu konvergenciju algoritma u ranim fazama optimizacije, predložena je implementacija haotičnih mapa u PSO algoritam. Predloženi pristup je eksperimentalno verifikovan na primeru dobijanja optimalnih planova terminiranja realnih delova.
AB  - This paper presents an approach for integration of process planning and scheduling based on the particle swarm optimization algorithm and chaos theory (cPSO). Besides scheduling plans representation and particle encoding/decoding scheme, mathematical model for the minimization of makespan, maximization of balanced level of machine utilization and minimization of mean flow time was presented. Also, we proposed implementation of chaotic maps in PSO algorithm in order to prevent algorithm from converging prematurely. Experimental verification of the proposed algorithm was done through the optimal scheduling of real parts.
PB  - JUPITER Asocijacija, Univerzitet u Beogradu - Mašinski fakultet
C3  - 40. JUPITER Konferencija, 36. simpozijum „NU-ROBOTI-FTS“ : Zbornik radova, Beograd, maj 2016
T1  - Интелигенција роја честица и теорија хаоса у интегрисаном пројектовању и терминирању флексибилних технолошких процеса
T1  - Particle Swarm Optimization Algorithm and Chaos Theory for Integration of Process Planning and Scheduling
EP  - 3.32
SP  - 3.22
UR  - https://hdl.handle.net/21.15107/rcub_machinery_4560
ER  - 

@conference{
author = "Petrović, Milica and Petronijević, Jelena and Mitić, Marko and Vuković, Najdan and Miljković, Zoran and Babić, Bojan",
year = "2016",
abstract = "U radu je prikazan pristup za integrisano projektovanje i terminiranje fleksibilnih teholoških procesa obrade delova primenom algoritma baziranog na inteligenciji roja čestica i teoriji haosa (cPSO algoritam). Pored metoda kodiranja/dekodiranja parametara planova terminiranja u jedinke cPSO algoritma, u radu je predložen matematički model za minimizaciju ukupnog vremena za obradu svih delova čije se terminiranje vrši, maksimizaciju uravnoteženog iskorišćenja mašina alatki i minimizaciju transportnih tokova materijala. Takođe, u cilju prevazilaženja nedostataka vezanih za brzu konvergenciju algoritma u ranim fazama optimizacije, predložena je implementacija haotičnih mapa u PSO algoritam. Predloženi pristup je eksperimentalno verifikovan na primeru dobijanja optimalnih planova terminiranja realnih delova., This paper presents an approach for integration of process planning and scheduling based on the particle swarm optimization algorithm and chaos theory (cPSO). Besides scheduling plans representation and particle encoding/decoding scheme, mathematical model for the minimization of makespan, maximization of balanced level of machine utilization and minimization of mean flow time was presented. Also, we proposed implementation of chaotic maps in PSO algorithm in order to prevent algorithm from converging prematurely. Experimental verification of the proposed algorithm was done through the optimal scheduling of real parts.",
publisher = "JUPITER Asocijacija, Univerzitet u Beogradu - Mašinski fakultet",
journal = "40. JUPITER Konferencija, 36. simpozijum „NU-ROBOTI-FTS“ : Zbornik radova, Beograd, maj 2016",
title = "Интелигенција роја честица и теорија хаоса у интегрисаном пројектовању и терминирању флексибилних технолошких процеса, Particle Swarm Optimization Algorithm and Chaos Theory for Integration of Process Planning and Scheduling",
pages = "3.32-3.22",
url = "https://hdl.handle.net/21.15107/rcub_machinery_4560"
}

Petrović, M., Petronijević, J., Mitić, M., Vuković, N., Miljković, Z.,& Babić, B.. (2016). Интелигенција роја честица и теорија хаоса у интегрисаном пројектовању и терминирању флексибилних технолошких процеса. in 40. JUPITER Konferencija, 36. simpozijum „NU-ROBOTI-FTS“ : Zbornik radova, Beograd, maj 2016
JUPITER Asocijacija, Univerzitet u Beogradu - Mašinski fakultet., 3.22-3.32.
https://hdl.handle.net/21.15107/rcub_machinery_4560

Petrović M, Petronijević J, Mitić M, Vuković N, Miljković Z, Babić B. Интелигенција роја честица и теорија хаоса у интегрисаном пројектовању и терминирању флексибилних технолошких процеса. in 40. JUPITER Konferencija, 36. simpozijum „NU-ROBOTI-FTS“ : Zbornik radova, Beograd, maj 2016. 2016;:3.22-3.32.
https://hdl.handle.net/21.15107/rcub_machinery_4560 .

Petrović, Milica, Petronijević, Jelena, Mitić, Marko, Vuković, Najdan, Miljković, Zoran, Babić, Bojan, "Интелигенција роја честица и теорија хаоса у интегрисаном пројектовању и терминирању флексибилних технолошких процеса" in 40. JUPITER Konferencija, 36. simpozijum „NU-ROBOTI-FTS“ : Zbornik radova, Beograd, maj 2016 (2016):3.22-3.32,
https://hdl.handle.net/21.15107/rcub_machinery_4560 .

TY  - JOUR
AU  - Petrović, Milica
AU  - Petronijević, Jelena
AU  - Mitić, Marko
AU  - Vuković, Najdan
AU  - Miljković, Zoran
AU  - Babić, Bojan
PY  - 2016
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/3964
AB  - Process planning and scheduling are two of the most important manufacturing functions which are usually performed sequentially in traditional approaches. Considering the fact that these functions are usually complementary, it is necessary to integrate them so as to improve performance of a manufacturing system. This paper presents implementation of novel nature-inspired Ant Lion Optimization (ALO) algorithm for solving this combinatorial optimization problem effectively. As the ALO algorithm mimics the intelligent behavior of antlions in hunting ants, the main steps of hunting prey, its mathematical modeling, and optimization procedure for integration of process planning and scheduling is proposed. The algorithm is implemented in Matlab environment and run on the 3.10 GHz processor with 2 GBs of RAM memory. Experimental results show applicability of the proposed approach in solving integrated process planning and scheduling problem.
T2  - Applied Mechanics and Materials
T1  - The Ant Lion Optimization Algorithm for Integrated Process Planning and Scheduling
SP  - 187-192
VL  - 834
DO  - 10.4028/www.scientific.net/AMM.834.187
ER  - 

@article{
author = "Petrović, Milica and Petronijević, Jelena and Mitić, Marko and Vuković, Najdan and Miljković, Zoran and Babić, Bojan",
year = "2016",
abstract = "Process planning and scheduling are two of the most important manufacturing functions which are usually performed sequentially in traditional approaches. Considering the fact that these functions are usually complementary, it is necessary to integrate them so as to improve performance of a manufacturing system. This paper presents implementation of novel nature-inspired Ant Lion Optimization (ALO) algorithm for solving this combinatorial optimization problem effectively. As the ALO algorithm mimics the intelligent behavior of antlions in hunting ants, the main steps of hunting prey, its mathematical modeling, and optimization procedure for integration of process planning and scheduling is proposed. The algorithm is implemented in Matlab environment and run on the 3.10 GHz processor with 2 GBs of RAM memory. Experimental results show applicability of the proposed approach in solving integrated process planning and scheduling problem.",
journal = "Applied Mechanics and Materials",
title = "The Ant Lion Optimization Algorithm for Integrated Process Planning and Scheduling",
pages = "187-192",
volume = "834",
doi = "10.4028/www.scientific.net/AMM.834.187"
}

Petrović, M., Petronijević, J., Mitić, M., Vuković, N., Miljković, Z.,& Babić, B.. (2016). The Ant Lion Optimization Algorithm for Integrated Process Planning and Scheduling. in Applied Mechanics and Materials, 834, 187-192.
https://doi.org/10.4028/www.scientific.net/AMM.834.187

Petrović M, Petronijević J, Mitić M, Vuković N, Miljković Z, Babić B. The Ant Lion Optimization Algorithm for Integrated Process Planning and Scheduling. in Applied Mechanics and Materials. 2016;834:187-192.
doi:10.4028/www.scientific.net/AMM.834.187 .

Petrović, Milica, Petronijević, Jelena, Mitić, Marko, Vuković, Najdan, Miljković, Zoran, Babić, Bojan, "The Ant Lion Optimization Algorithm for Integrated Process Planning and Scheduling" in Applied Mechanics and Materials, 834 (2016):187-192,
https://doi.org/10.4028/www.scientific.net/AMM.834.187 . .

TY  - JOUR
AU  - Petronijević, J
AU  - Petrović, Milica
AU  - Vuković, Najdan
AU  - Mitić, Marko
AU  - Babić, Bojan
AU  - Miljković, Zoran
PY  - 2016
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/3965
AB  - Market growth and mass customization cause a need for a change in traditional manufacturing. Decentralized decision making and integration of process planning is necessary in order to become concurrent in the market. The paper presents decentralized decision making methodology using multi-agent systems. The model is used for integrated process planning and scheduling based on the minimum processing time under dynamic change of the environment. Two types of disturbance are used to represent the change: part arrival and machine breakdown. The proposed model comprises part agent, job agent, machine agent and optimization agent. Comparative analysis is conducted using simulation in AnyLogic software in order to verify the proposed approach.
T2  - Applied Mechanics and Materials
T1  - Integrated process planning and scheduling using multi-agent methodology
SP  - 192-198
VL  - 834
DO  - 10.4028/www.scientific.net/AMM.834.193
ER  - 

@article{
author = "Petronijević, J and Petrović, Milica and Vuković, Najdan and Mitić, Marko and Babić, Bojan and Miljković, Zoran",
year = "2016",
abstract = "Market growth and mass customization cause a need for a change in traditional manufacturing. Decentralized decision making and integration of process planning is necessary in order to become concurrent in the market. The paper presents decentralized decision making methodology using multi-agent systems. The model is used for integrated process planning and scheduling based on the minimum processing time under dynamic change of the environment. Two types of disturbance are used to represent the change: part arrival and machine breakdown. The proposed model comprises part agent, job agent, machine agent and optimization agent. Comparative analysis is conducted using simulation in AnyLogic software in order to verify the proposed approach.",
journal = "Applied Mechanics and Materials",
title = "Integrated process planning and scheduling using multi-agent methodology",
pages = "192-198",
volume = "834",
doi = "10.4028/www.scientific.net/AMM.834.193"
}

Petronijević, J., Petrović, M., Vuković, N., Mitić, M., Babić, B.,& Miljković, Z.. (2016). Integrated process planning and scheduling using multi-agent methodology. in Applied Mechanics and Materials, 834, 192-198.
https://doi.org/10.4028/www.scientific.net/AMM.834.193

Petronijević J, Petrović M, Vuković N, Mitić M, Babić B, Miljković Z. Integrated process planning and scheduling using multi-agent methodology. in Applied Mechanics and Materials. 2016;834:192-198.
doi:10.4028/www.scientific.net/AMM.834.193 .

Petronijević, J, Petrović, Milica, Vuković, Najdan, Mitić, Marko, Babić, Bojan, Miljković, Zoran, "Integrated process planning and scheduling using multi-agent methodology" in Applied Mechanics and Materials, 834 (2016):192-198,
https://doi.org/10.4028/www.scientific.net/AMM.834.193 . .

TY  - GEN
AU  - Petrović, Milica
AU  - Miljković, Zoran
AU  - Vuković, Najdan
PY  - 2016
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/4816
AB  - Tehničko rešenje - nova metoda (M85), odnosi se na rešavanje problema generisanja optimalnih tehnoloških procesa mašinske obrade delova primenom biološki inspirisanog algoritma na bazi inteligencije mravolovaca (engl. Ant Lion Optimization – ALO), čime se ostvaruju i optimalni tehnološki procesi sa minimalnim proizvodnim vremenom i minimalnim proizvodnim troškovima. Ova metoda je razvijana u okviru naučnog projekta TR-35004 MPNiTR Vlade Republike Srbije.
T2  - Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер
T1  - Optimizacija fleksibilnih tehnoloških procesa primenom biološki inspirisanog "Ant Lion Optimization" algoritma
UR  - https://hdl.handle.net/21.15107/rcub_machinery_4816
ER  - 

@misc{
author = "Petrović, Milica and Miljković, Zoran and Vuković, Najdan",
year = "2016",
abstract = "Tehničko rešenje - nova metoda (M85), odnosi se na rešavanje problema generisanja optimalnih tehnoloških procesa mašinske obrade delova primenom biološki inspirisanog algoritma na bazi inteligencije mravolovaca (engl. Ant Lion Optimization – ALO), čime se ostvaruju i optimalni tehnološki procesi sa minimalnim proizvodnim vremenom i minimalnim proizvodnim troškovima. Ova metoda je razvijana u okviru naučnog projekta TR-35004 MPNiTR Vlade Republike Srbije.",
journal = "Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер",
title = "Optimizacija fleksibilnih tehnoloških procesa primenom biološki inspirisanog "Ant Lion Optimization" algoritma",
url = "https://hdl.handle.net/21.15107/rcub_machinery_4816"
}

Petrović, M., Miljković, Z.,& Vuković, N.. (2016). Optimizacija fleksibilnih tehnoloških procesa primenom biološki inspirisanog "Ant Lion Optimization" algoritma. in Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер.
https://hdl.handle.net/21.15107/rcub_machinery_4816

Petrović M, Miljković Z, Vuković N. Optimizacija fleksibilnih tehnoloških procesa primenom biološki inspirisanog "Ant Lion Optimization" algoritma. in Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер. 2016;.
https://hdl.handle.net/21.15107/rcub_machinery_4816 .

Petrović, Milica, Miljković, Zoran, Vuković, Najdan, "Optimizacija fleksibilnih tehnoloških procesa primenom biološki inspirisanog "Ant Lion Optimization" algoritma" in Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер (2016),
https://hdl.handle.net/21.15107/rcub_machinery_4816 .

TY  - JOUR
AU  - Petrović, Milica
AU  - Vuković, Najdan
AU  - Mitić, Marko
AU  - Miljković, Zoran
PY  - 2016
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/2462
AB  - Process planning and scheduling are two of the most important manufacturing functions traditionally performed separately and sequentially. These functions being complementary and interrelated, their integration is essential for the optimal utilization of manufacturing resources. Such integration is also significant for improving the performance of the modern manufacturing system. A variety of alternative manufacturing resources (machine tools, cutting tools, tool access directions, etc.) causes integrated process planning and scheduling (IPPS) problem to be strongly NP-hard (non deterministic polynomial) in terms of combinatorial optimization. Therefore, an optimal solution for the problem is searched in a vast search space. In order to explore the search space comprehensively and avoid being trapped into local optima, this paper focuses on using the method based on the particle swarm optimization algorithm and chaos theory (cPSO). The initial solutions for the IPPS problem are presented in the form of the particles of cPSO algorithm. The particle encoding/decoding scheme is also proposed in this paper. Flexible process and scheduling plans are presented using AND/OR network and five flexibility types: machine, tool, tool access direction (TAD), process, and sequence flexibility. Optimal process plans are obtained by multi objective optimization of production time and production cost. On the other hand, optimal scheduling plans are generated based on three objective functions: makespan, balanced level of machine utilization, and mean flow time. The proposed cPSO algorithm is implemented in Matlab environment and verified extensively using five experimental studies. The experimental results show that the proposed algorithm outperforms genetic algorithm (GA), simulated annealing (SA) based approach, and hybrid algorithm. Moreover, the scheduling plans obtained by the proposed methodology are additionally tested by Khepera II mobile robot using a laboratory model of manufacturing environment.
PB  - Pergamon-Elsevier Science Ltd, Oxford
T2  - Expert Systems With Applications
T1  - Integration of process planning and scheduling using chaotic particle swarm optimization algorithm
EP  - 588
SP  - 569
VL  - 64
DO  - 10.1016/j.eswa.2016.08.019
ER  - 

@article{
author = "Petrović, Milica and Vuković, Najdan and Mitić, Marko and Miljković, Zoran",
year = "2016",
abstract = "Process planning and scheduling are two of the most important manufacturing functions traditionally performed separately and sequentially. These functions being complementary and interrelated, their integration is essential for the optimal utilization of manufacturing resources. Such integration is also significant for improving the performance of the modern manufacturing system. A variety of alternative manufacturing resources (machine tools, cutting tools, tool access directions, etc.) causes integrated process planning and scheduling (IPPS) problem to be strongly NP-hard (non deterministic polynomial) in terms of combinatorial optimization. Therefore, an optimal solution for the problem is searched in a vast search space. In order to explore the search space comprehensively and avoid being trapped into local optima, this paper focuses on using the method based on the particle swarm optimization algorithm and chaos theory (cPSO). The initial solutions for the IPPS problem are presented in the form of the particles of cPSO algorithm. The particle encoding/decoding scheme is also proposed in this paper. Flexible process and scheduling plans are presented using AND/OR network and five flexibility types: machine, tool, tool access direction (TAD), process, and sequence flexibility. Optimal process plans are obtained by multi objective optimization of production time and production cost. On the other hand, optimal scheduling plans are generated based on three objective functions: makespan, balanced level of machine utilization, and mean flow time. The proposed cPSO algorithm is implemented in Matlab environment and verified extensively using five experimental studies. The experimental results show that the proposed algorithm outperforms genetic algorithm (GA), simulated annealing (SA) based approach, and hybrid algorithm. Moreover, the scheduling plans obtained by the proposed methodology are additionally tested by Khepera II mobile robot using a laboratory model of manufacturing environment.",
publisher = "Pergamon-Elsevier Science Ltd, Oxford",
journal = "Expert Systems With Applications",
title = "Integration of process planning and scheduling using chaotic particle swarm optimization algorithm",
pages = "588-569",
volume = "64",
doi = "10.1016/j.eswa.2016.08.019"
}

Petrović, M., Vuković, N., Mitić, M.,& Miljković, Z.. (2016). Integration of process planning and scheduling using chaotic particle swarm optimization algorithm. in Expert Systems With Applications
Pergamon-Elsevier Science Ltd, Oxford., 64, 569-588.
https://doi.org/10.1016/j.eswa.2016.08.019

Petrović M, Vuković N, Mitić M, Miljković Z. Integration of process planning and scheduling using chaotic particle swarm optimization algorithm. in Expert Systems With Applications. 2016;64:569-588.
doi:10.1016/j.eswa.2016.08.019 .

Petrović, Milica, Vuković, Najdan, Mitić, Marko, Miljković, Zoran, "Integration of process planning and scheduling using chaotic particle swarm optimization algorithm" in Expert Systems With Applications, 64 (2016):569-588,
https://doi.org/10.1016/j.eswa.2016.08.019 . .

TY  - JOUR
AU  - Petrović, Milica
AU  - Mitić, Marko
AU  - Vuković, Najdan
AU  - Miljković, Zoran
PY  - 2016
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/2475
AB  - A variety of manufacturing operations together with a variety of alternative manufacturing resources provide that most jobs in the modern manufacturing systems may have a large number of alternative process plans. For that reason, obtaining an optimal process plan according to all alternative manufacturing resources (machine tools, cutting tools, tool access directions, etc.) as well as alternative operations has become a very important task in flexible process planning problem research. In this paper, we present and evaluate a new algorithm for optimization of flexible process plans based on utilization of particle swarm optimization (PSO) algorithm and chaos theory. The main idea is to prevent the convergence of PSO in early stages of optimization process by implementing ten different chaotic maps which enlarge search space and provide its diversity. The flexible process plans are represented by using AND/OR network, and machine flexibility, tool flexibility, tool access direction (TAD) flexibility, process flexibility and sequence flexibility are considered. Further, mathematical models for minimization of production time and total production cost are derived. The newly developed algorithm is extensively experimentally verified by using four experimental studies, which show that the developed method outperforms genetic algorithm (GA), simulated annealing (SA), hybrid GA-SA and generic PSO based approach.
PB  - Springer London Ltd, London
T2  - International Journal of Advanced Manufacturing Technology
T1  - Chaotic particle swarm optimization algorithm for flexible process planning
EP  - 2555
IS  - 9-12
SP  - 2535
VL  - 85
DO  - 10.1007/s00170-015-7991-4
ER  - 

@article{
author = "Petrović, Milica and Mitić, Marko and Vuković, Najdan and Miljković, Zoran",
year = "2016",
abstract = "A variety of manufacturing operations together with a variety of alternative manufacturing resources provide that most jobs in the modern manufacturing systems may have a large number of alternative process plans. For that reason, obtaining an optimal process plan according to all alternative manufacturing resources (machine tools, cutting tools, tool access directions, etc.) as well as alternative operations has become a very important task in flexible process planning problem research. In this paper, we present and evaluate a new algorithm for optimization of flexible process plans based on utilization of particle swarm optimization (PSO) algorithm and chaos theory. The main idea is to prevent the convergence of PSO in early stages of optimization process by implementing ten different chaotic maps which enlarge search space and provide its diversity. The flexible process plans are represented by using AND/OR network, and machine flexibility, tool flexibility, tool access direction (TAD) flexibility, process flexibility and sequence flexibility are considered. Further, mathematical models for minimization of production time and total production cost are derived. The newly developed algorithm is extensively experimentally verified by using four experimental studies, which show that the developed method outperforms genetic algorithm (GA), simulated annealing (SA), hybrid GA-SA and generic PSO based approach.",
publisher = "Springer London Ltd, London",
journal = "International Journal of Advanced Manufacturing Technology",
title = "Chaotic particle swarm optimization algorithm for flexible process planning",
pages = "2555-2535",
number = "9-12",
volume = "85",
doi = "10.1007/s00170-015-7991-4"
}

Petrović, M., Mitić, M., Vuković, N.,& Miljković, Z.. (2016). Chaotic particle swarm optimization algorithm for flexible process planning. in International Journal of Advanced Manufacturing Technology
Springer London Ltd, London., 85(9-12), 2535-2555.
https://doi.org/10.1007/s00170-015-7991-4

Petrović M, Mitić M, Vuković N, Miljković Z. Chaotic particle swarm optimization algorithm for flexible process planning. in International Journal of Advanced Manufacturing Technology. 2016;85(9-12):2535-2555.
doi:10.1007/s00170-015-7991-4 .

Petrović, Milica, Mitić, Marko, Vuković, Najdan, Miljković, Zoran, "Chaotic particle swarm optimization algorithm for flexible process planning" in International Journal of Advanced Manufacturing Technology, 85, no. 9-12 (2016):2535-2555,
https://doi.org/10.1007/s00170-015-7991-4 . .

TY  - JOUR
AU  - Miljković, Zoran
AU  - Vuković, Najdan
AU  - Mitić, Marko
PY  - 2016
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/2472
AB  - The extended Kalman filter (EKF) has become a popular solution for the simultaneous localization and mapping (SLAM). This paper presents the implementation of the EKF coupled with a feedforward neural network for the monocular SLAM. The neural extended Kalman filter (NEKF) is applied online to approximate an error between the motion model of the mobile robot and the real system performance. Inadequate modeling of the robot motion can jeopardize the quality of estimation. The paper shows integration of EKF with feedforward neural network and simulation analysis of its consistency and implementation of the NEKF with a mobile robot, laboratory experimental environment, and a simple USB camera. The simulation and experimental results show that integration of neural network into EKF prediction-correction cycle results in improved consistency and accuracy.
PB  - Sage Publications Ltd, London
T2  - Proceedings of The Institution of Mechanical Engineers Part C-Journal of Mechanical Engineering Scie
T1  - Neural extended Kalman filter for monocular SLAM in indoor environment
EP  - 866
IS  - 5
SP  - 856
VL  - 230
DO  - 10.1177/0954406215586589
ER  - 

@article{
author = "Miljković, Zoran and Vuković, Najdan and Mitić, Marko",
year = "2016",
abstract = "The extended Kalman filter (EKF) has become a popular solution for the simultaneous localization and mapping (SLAM). This paper presents the implementation of the EKF coupled with a feedforward neural network for the monocular SLAM. The neural extended Kalman filter (NEKF) is applied online to approximate an error between the motion model of the mobile robot and the real system performance. Inadequate modeling of the robot motion can jeopardize the quality of estimation. The paper shows integration of EKF with feedforward neural network and simulation analysis of its consistency and implementation of the NEKF with a mobile robot, laboratory experimental environment, and a simple USB camera. The simulation and experimental results show that integration of neural network into EKF prediction-correction cycle results in improved consistency and accuracy.",
publisher = "Sage Publications Ltd, London",
journal = "Proceedings of The Institution of Mechanical Engineers Part C-Journal of Mechanical Engineering Scie",
title = "Neural extended Kalman filter for monocular SLAM in indoor environment",
pages = "866-856",
number = "5",
volume = "230",
doi = "10.1177/0954406215586589"
}

Miljković, Z., Vuković, N.,& Mitić, M.. (2016). Neural extended Kalman filter for monocular SLAM in indoor environment. in Proceedings of The Institution of Mechanical Engineers Part C-Journal of Mechanical Engineering Scie
Sage Publications Ltd, London., 230(5), 856-866.
https://doi.org/10.1177/0954406215586589

Miljković Z, Vuković N, Mitić M. Neural extended Kalman filter for monocular SLAM in indoor environment. in Proceedings of The Institution of Mechanical Engineers Part C-Journal of Mechanical Engineering Scie. 2016;230(5):856-866.
doi:10.1177/0954406215586589 .

Miljković, Zoran, Vuković, Najdan, Mitić, Marko, "Neural extended Kalman filter for monocular SLAM in indoor environment" in Proceedings of The Institution of Mechanical Engineers Part C-Journal of Mechanical Engineering Scie, 230, no. 5 (2016):856-866,
https://doi.org/10.1177/0954406215586589 . .

TY  - CONF
AU  - Petronijević, Jelena
AU  - Petrović, Milica
AU  - Vuković, Najdan
AU  - Mitić, Marko
AU  - Babić, Bojan
AU  - Miljković, Zoran
PY  - 2015
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/4623
AB  - Multi-agent systems have been used for modelling various problems in the social, biological and technical domain. When comes to technical systems, especially manufacturing systems, agents are most often applied in optimization and scheduling problems. Traditionally, scheduling is done after creation of process plans. In this paper, agent methodology is used for integration of these two functions. The proposed multi-agent architecture provides simultaneous performance of process planning and scheduling and it consists of four intelligent agents: part and job agents, machine agent, and optimization agent. Verification and feasibility of a proposed approach is
conducted using agent based simulation in AnyLogic software.
PB  - Novi Sad : Faculty of Technical Sciences
C3  - Proceedings of the 12th International Scientific Conference MMA 2015 – Flexible Technologies, Novi Sad, 25-26 September 2015
T1  - Multi-Agent Modeling for Integrated Process Planning and Scheduling
EP  - 124
SP  - 121
UR  - https://hdl.handle.net/21.15107/rcub_machinery_4623
ER  - 

@conference{
author = "Petronijević, Jelena and Petrović, Milica and Vuković, Najdan and Mitić, Marko and Babić, Bojan and Miljković, Zoran",
year = "2015",
abstract = "Multi-agent systems have been used for modelling various problems in the social, biological and technical domain. When comes to technical systems, especially manufacturing systems, agents are most often applied in optimization and scheduling problems. Traditionally, scheduling is done after creation of process plans. In this paper, agent methodology is used for integration of these two functions. The proposed multi-agent architecture provides simultaneous performance of process planning and scheduling and it consists of four intelligent agents: part and job agents, machine agent, and optimization agent. Verification and feasibility of a proposed approach is
conducted using agent based simulation in AnyLogic software.",
publisher = "Novi Sad : Faculty of Technical Sciences",
journal = "Proceedings of the 12th International Scientific Conference MMA 2015 – Flexible Technologies, Novi Sad, 25-26 September 2015",
title = "Multi-Agent Modeling for Integrated Process Planning and Scheduling",
pages = "124-121",
url = "https://hdl.handle.net/21.15107/rcub_machinery_4623"
}

Petronijević, J., Petrović, M., Vuković, N., Mitić, M., Babić, B.,& Miljković, Z.. (2015). Multi-Agent Modeling for Integrated Process Planning and Scheduling. in Proceedings of the 12th International Scientific Conference MMA 2015 – Flexible Technologies, Novi Sad, 25-26 September 2015
Novi Sad : Faculty of Technical Sciences., 121-124.
https://hdl.handle.net/21.15107/rcub_machinery_4623

Petronijević J, Petrović M, Vuković N, Mitić M, Babić B, Miljković Z. Multi-Agent Modeling for Integrated Process Planning and Scheduling. in Proceedings of the 12th International Scientific Conference MMA 2015 – Flexible Technologies, Novi Sad, 25-26 September 2015. 2015;:121-124.
https://hdl.handle.net/21.15107/rcub_machinery_4623 .

Petronijević, Jelena, Petrović, Milica, Vuković, Najdan, Mitić, Marko, Babić, Bojan, Miljković, Zoran, "Multi-Agent Modeling for Integrated Process Planning and Scheduling" in Proceedings of the 12th International Scientific Conference MMA 2015 – Flexible Technologies, Novi Sad, 25-26 September 2015 (2015):121-124,
https://hdl.handle.net/21.15107/rcub_machinery_4623 .

TY  - CONF
AU  - Mitić, Marko
AU  - Vuković, Najdan
AU  - Petrović, Milica
AU  - Petronijević, Jelena
AU  - Diryag, Ali
AU  - Miljković, Zoran
PY  - 2015
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/4469
AB  - This paper presents concise comparison study of newly developed bioinspired algorithms for global optimization problems. Three different metaheuristic techniques, namely Accelerated Particle Swarm Optimization (APSO), Firefly Algorithm (FA), and Grey Wolf Optimizer (GWO) are investigated and implemented in Matlab environment. These methods are compared on four unimodal and multimodal nonlinear functions in order to find global optimum values. Computational results indicate that GWO outperforms other intelligent techniques, and that all aforementioned algorithms can be successfully used for optimization of continuous functions.
PB  - Society for Information Systems and Computer Networks
C3  - Proceedings of the 5th International Conference on Information Society and Technology (ICIST 2015), Kopaonik 8-11. March 2015
T1  - Bioinspired metaheuristic algorithms for global optimization
EP  - 42
SP  - 38
UR  - https://hdl.handle.net/21.15107/rcub_machinery_4469
ER  - 

@conference{
author = "Mitić, Marko and Vuković, Najdan and Petrović, Milica and Petronijević, Jelena and Diryag, Ali and Miljković, Zoran",
year = "2015",
abstract = "This paper presents concise comparison study of newly developed bioinspired algorithms for global optimization problems. Three different metaheuristic techniques, namely Accelerated Particle Swarm Optimization (APSO), Firefly Algorithm (FA), and Grey Wolf Optimizer (GWO) are investigated and implemented in Matlab environment. These methods are compared on four unimodal and multimodal nonlinear functions in order to find global optimum values. Computational results indicate that GWO outperforms other intelligent techniques, and that all aforementioned algorithms can be successfully used for optimization of continuous functions.",
publisher = "Society for Information Systems and Computer Networks",
journal = "Proceedings of the 5th International Conference on Information Society and Technology (ICIST 2015), Kopaonik 8-11. March 2015",
title = "Bioinspired metaheuristic algorithms for global optimization",
pages = "42-38",
url = "https://hdl.handle.net/21.15107/rcub_machinery_4469"
}

Mitić, M., Vuković, N., Petrović, M., Petronijević, J., Diryag, A.,& Miljković, Z.. (2015). Bioinspired metaheuristic algorithms for global optimization. in Proceedings of the 5th International Conference on Information Society and Technology (ICIST 2015), Kopaonik 8-11. March 2015
Society for Information Systems and Computer Networks., 38-42.
https://hdl.handle.net/21.15107/rcub_machinery_4469

Mitić M, Vuković N, Petrović M, Petronijević J, Diryag A, Miljković Z. Bioinspired metaheuristic algorithms for global optimization. in Proceedings of the 5th International Conference on Information Society and Technology (ICIST 2015), Kopaonik 8-11. March 2015. 2015;:38-42.
https://hdl.handle.net/21.15107/rcub_machinery_4469 .

Mitić, Marko, Vuković, Najdan, Petrović, Milica, Petronijević, Jelena, Diryag, Ali, Miljković, Zoran, "Bioinspired metaheuristic algorithms for global optimization" in Proceedings of the 5th International Conference on Information Society and Technology (ICIST 2015), Kopaonik 8-11. March 2015 (2015):38-42,
https://hdl.handle.net/21.15107/rcub_machinery_4469 .

TY  - CONF
AU  - Vuković, Najdan
AU  - Mitić, Marko
AU  - Petrović, Milica
AU  - Petronijević, Jelena
AU  - Miljković, Zoran
PY  - 2015
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/4468
AB  - In this paper we test Extended Information Filter (EIF) for sequential training of Hyper Basis Function Neural Networks with growing and pruning ability (HBF-GP). The HBF neuron allows different scaling of input dimensions to provide better generalization property when dealing with complex nonlinear problems in engineering practice. The main intuition behind HBF is in generalization of Gaussian type of neuron that applies Mahalanobis-like distance as a distance metrics between input training sample and prototype vector. We exploit concept of neuron’s significance and allow growing and pruning of HBF neurons during sequential learning process. From engineer’s perspective, EIF is attractive for training of neural networks because it allows a designer to have scarce initial knowledge of the system/problem. Extensive experimental study shows that HBF neural network trained with EIF achieves same prediction error and compactness of network topology when compared to EKF, but without the need to know initial state uncertainty, which is its main advantage over EKF.
PB  - Society for Information Systems and Computer Networks
C3  - Proceedings of the 5th International Conference on Information Society and Technology (ICIST 2015), Kopaonik 8-11. March 2015
T1  - Experimental Evaluation of Growing and Pruning Hyper Basis Function Neural Networks Trained with Extended Information Filter
EP  - 94
SP  - 89
UR  - https://hdl.handle.net/21.15107/rcub_machinery_4468
ER  - 

@conference{
author = "Vuković, Najdan and Mitić, Marko and Petrović, Milica and Petronijević, Jelena and Miljković, Zoran",
year = "2015",
abstract = "In this paper we test Extended Information Filter (EIF) for sequential training of Hyper Basis Function Neural Networks with growing and pruning ability (HBF-GP). The HBF neuron allows different scaling of input dimensions to provide better generalization property when dealing with complex nonlinear problems in engineering practice. The main intuition behind HBF is in generalization of Gaussian type of neuron that applies Mahalanobis-like distance as a distance metrics between input training sample and prototype vector. We exploit concept of neuron’s significance and allow growing and pruning of HBF neurons during sequential learning process. From engineer’s perspective, EIF is attractive for training of neural networks because it allows a designer to have scarce initial knowledge of the system/problem. Extensive experimental study shows that HBF neural network trained with EIF achieves same prediction error and compactness of network topology when compared to EKF, but without the need to know initial state uncertainty, which is its main advantage over EKF.",
publisher = "Society for Information Systems and Computer Networks",
journal = "Proceedings of the 5th International Conference on Information Society and Technology (ICIST 2015), Kopaonik 8-11. March 2015",
title = "Experimental Evaluation of Growing and Pruning Hyper Basis Function Neural Networks Trained with Extended Information Filter",
pages = "94-89",
url = "https://hdl.handle.net/21.15107/rcub_machinery_4468"
}

Vuković, N., Mitić, M., Petrović, M., Petronijević, J.,& Miljković, Z.. (2015). Experimental Evaluation of Growing and Pruning Hyper Basis Function Neural Networks Trained with Extended Information Filter. in Proceedings of the 5th International Conference on Information Society and Technology (ICIST 2015), Kopaonik 8-11. March 2015
Society for Information Systems and Computer Networks., 89-94.
https://hdl.handle.net/21.15107/rcub_machinery_4468

Vuković N, Mitić M, Petrović M, Petronijević J, Miljković Z. Experimental Evaluation of Growing and Pruning Hyper Basis Function Neural Networks Trained with Extended Information Filter. in Proceedings of the 5th International Conference on Information Society and Technology (ICIST 2015), Kopaonik 8-11. March 2015. 2015;:89-94.
https://hdl.handle.net/21.15107/rcub_machinery_4468 .

Vuković, Najdan, Mitić, Marko, Petrović, Milica, Petronijević, Jelena, Miljković, Zoran, "Experimental Evaluation of Growing and Pruning Hyper Basis Function Neural Networks Trained with Extended Information Filter" in Proceedings of the 5th International Conference on Information Society and Technology (ICIST 2015), Kopaonik 8-11. March 2015 (2015):89-94,
https://hdl.handle.net/21.15107/rcub_machinery_4468 .

TY  - GEN
AU  - Petronijević, Jelena
AU  - Petrović, Milica
AU  - Vuković, Najdan
AU  - Mitić, Marko
AU  - Babić, Bojan
AU  - Miljković, Zoran
PY  - 2015
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/4815
AB  - Tehničko rešenje - nova metoda (M85), pripada oblasti mašinstva i odnosi se na domen dinamičkog integrisanog planiranja i terminiranja proizvodnje. Rešavanje problema izvodi se multiagentnom metodologijom. Predložena multiagentna arhitektura se sastoji iz pet agenata: agent za delove, agent za operacije, agent za mašine, agent za alate i agent za sinhronizaciju. Sinhronizovanim dejstvom svih agenata uz posedovanje informacije o alternativnim tehnološkim postupcima, a u zavisnosti od stanja okruženja vrši se dinamičko planiranje i terminiranje proizvodnje. Verifikacija predloženog rešenja izvedena je u AnyLogic softverskom paketu. Rezultati simulacije pokazuju da predložena arhitektura omogućuje promenu i prilagođavanje tehnoloških postupaka, kao i planova terminiranja, u zavisnosti od stanja simuliranog modela tehnološkog okruženja. Razvijana je kroz aktivnosti u okviru naučnog projekta pod oznakom TR-35004 MPNiTR Vlade Republike Srbije.
T2  - Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер
T1  - Multiagentni sistem za dinamičko integrisano planiranje i terminiranje proizvodnje
UR  - https://hdl.handle.net/21.15107/rcub_machinery_4815
ER  - 

@misc{
author = "Petronijević, Jelena and Petrović, Milica and Vuković, Najdan and Mitić, Marko and Babić, Bojan and Miljković, Zoran",
year = "2015",
abstract = "Tehničko rešenje - nova metoda (M85), pripada oblasti mašinstva i odnosi se na domen dinamičkog integrisanog planiranja i terminiranja proizvodnje. Rešavanje problema izvodi se multiagentnom metodologijom. Predložena multiagentna arhitektura se sastoji iz pet agenata: agent za delove, agent za operacije, agent za mašine, agent za alate i agent za sinhronizaciju. Sinhronizovanim dejstvom svih agenata uz posedovanje informacije o alternativnim tehnološkim postupcima, a u zavisnosti od stanja okruženja vrši se dinamičko planiranje i terminiranje proizvodnje. Verifikacija predloženog rešenja izvedena je u AnyLogic softverskom paketu. Rezultati simulacije pokazuju da predložena arhitektura omogućuje promenu i prilagođavanje tehnoloških postupaka, kao i planova terminiranja, u zavisnosti od stanja simuliranog modela tehnološkog okruženja. Razvijana je kroz aktivnosti u okviru naučnog projekta pod oznakom TR-35004 MPNiTR Vlade Republike Srbije.",
journal = "Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер",
title = "Multiagentni sistem za dinamičko integrisano planiranje i terminiranje proizvodnje",
url = "https://hdl.handle.net/21.15107/rcub_machinery_4815"
}

Petronijević, J., Petrović, M., Vuković, N., Mitić, M., Babić, B.,& Miljković, Z.. (2015). Multiagentni sistem za dinamičko integrisano planiranje i terminiranje proizvodnje. in Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер.
https://hdl.handle.net/21.15107/rcub_machinery_4815

Petronijević J, Petrović M, Vuković N, Mitić M, Babić B, Miljković Z. Multiagentni sistem za dinamičko integrisano planiranje i terminiranje proizvodnje. in Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер. 2015;.
https://hdl.handle.net/21.15107/rcub_machinery_4815 .

Petronijević, Jelena, Petrović, Milica, Vuković, Najdan, Mitić, Marko, Babić, Bojan, Miljković, Zoran, "Multiagentni sistem za dinamičko integrisano planiranje i terminiranje proizvodnje" in Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер (2015),
https://hdl.handle.net/21.15107/rcub_machinery_4815 .

TY  - GEN
AU  - Petrović, Milica
AU  - Petronijević, Jelena
AU  - Mitić, Marko
AU  - Vuković, Najdan
AU  - Miljković, Zoran
AU  - Babić, Bojan
PY  - 2015
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/4813
AB  - Tehničko rešenje - nova metoda (M85), pripada oblasti mašinstva i direktno se odnosi na domen integrisanog projektovanja i terminiranja fleksibilnih tehnoloških procesa. Shodno tome, metoda rešava problem generisanja optimalnih planova terminiranja primenom biološki inspirisanog algoritma na bazi inteligencije roja čestica (engl. PSO – Particle Swarm Optimization) i teorije haosa (engl. Chaos theory). Jedan od nedostataka tradicionalnog PSO algoritma je i konvergencija ka lokalnom optimalnom rešenju u ranim fazama optimizacije. U cilju prevazilaženja nedostatka vezanih za brzu konvergenciju algoritma i povećavanje prostora alternativnih rešenja, haotične mape su implementirane u PSO algoritam. Rezultati optimizacije planova terminiranja za odabrane „benchmark“ delove iz literature pokazuju opravdanost primene predloženog koncepta. Razvijana je u okviru aktivnosti naučnog projekta pod oznakomТР-35004 MPNiTR Vlade Republike Srbije.
T2  - Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер
T1  - Integrisano projektovanje i terminiranje tehnoloških procesa primenom inteligencije roja čestica i teorije haosa
UR  - https://hdl.handle.net/21.15107/rcub_machinery_4813
ER  - 

@misc{
author = "Petrović, Milica and Petronijević, Jelena and Mitić, Marko and Vuković, Najdan and Miljković, Zoran and Babić, Bojan",
year = "2015",
abstract = "Tehničko rešenje - nova metoda (M85), pripada oblasti mašinstva i direktno se odnosi na domen integrisanog projektovanja i terminiranja fleksibilnih tehnoloških procesa. Shodno tome, metoda rešava problem generisanja optimalnih planova terminiranja primenom biološki inspirisanog algoritma na bazi inteligencije roja čestica (engl. PSO – Particle Swarm Optimization) i teorije haosa (engl. Chaos theory). Jedan od nedostataka tradicionalnog PSO algoritma je i konvergencija ka lokalnom optimalnom rešenju u ranim fazama optimizacije. U cilju prevazilaženja nedostatka vezanih za brzu konvergenciju algoritma i povećavanje prostora alternativnih rešenja, haotične mape su implementirane u PSO algoritam. Rezultati optimizacije planova terminiranja za odabrane „benchmark“ delove iz literature pokazuju opravdanost primene predloženog koncepta. Razvijana je u okviru aktivnosti naučnog projekta pod oznakomТР-35004 MPNiTR Vlade Republike Srbije.",
journal = "Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер",
title = "Integrisano projektovanje i terminiranje tehnoloških procesa primenom inteligencije roja čestica i teorije haosa",
url = "https://hdl.handle.net/21.15107/rcub_machinery_4813"
}

Petrović, M., Petronijević, J., Mitić, M., Vuković, N., Miljković, Z.,& Babić, B.. (2015). Integrisano projektovanje i terminiranje tehnoloških procesa primenom inteligencije roja čestica i teorije haosa. in Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер.
https://hdl.handle.net/21.15107/rcub_machinery_4813

Petrović M, Petronijević J, Mitić M, Vuković N, Miljković Z, Babić B. Integrisano projektovanje i terminiranje tehnoloških procesa primenom inteligencije roja čestica i teorije haosa. in Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер. 2015;.
https://hdl.handle.net/21.15107/rcub_machinery_4813 .

Petrović, Milica, Petronijević, Jelena, Mitić, Marko, Vuković, Najdan, Miljković, Zoran, Babić, Bojan, "Integrisano projektovanje i terminiranje tehnoloških procesa primenom inteligencije roja čestica i teorije haosa" in Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер (2015),
https://hdl.handle.net/21.15107/rcub_machinery_4813 .

TY  - JOUR
AU  - Mitić, Marko
AU  - Vuković, Najdan
AU  - Petrović, Milica
AU  - Miljković, Zoran
PY  - 2015
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/3929
AB  - Fruit fly optimization algorithm (FOA) is recently presented metaheuristic technique that is inspired by the behavior of fruit flies. This paper improves the standard FOA by introducing the novel parameter integrated with chaos. The performance of developed chaotic fruit fly algorithm (CFOA) is investigated in details on ten well known benchmark problems using fourteen different chaotic maps. Moreover, we performed comparison studies with basic FOA, FOA with Levy flight distribution, and other recently published chaotic algorithms. Statistical results on every optimization task indicate that the chaotic fruit fly algorithm (CFOA) has a very fast convergence rate. In addition, CFOA is compared with recently developed chaos enhanced algorithms such as chaotic bat algorithm, chaotic accelerated particle swarm optimization, chaotic firefly algorithm, chaotic artificial bee colony algorithm, and chaotic cuckoo search. Overall research findings show that FOA with Chebyshev map show superiority in terms of reliability of global optimality and algorithm success rate.
PB  - Elsevier, Amsterdam
T2  - Knowledge-Based Systems
T1  - Chaotic fruit fly optimization algorithm
EP  - 458
SP  - 446
VL  - 89
DO  - 10.1016/j.knosys.2015.08.010
ER  - 

@article{
author = "Mitić, Marko and Vuković, Najdan and Petrović, Milica and Miljković, Zoran",
year = "2015",
abstract = "Fruit fly optimization algorithm (FOA) is recently presented metaheuristic technique that is inspired by the behavior of fruit flies. This paper improves the standard FOA by introducing the novel parameter integrated with chaos. The performance of developed chaotic fruit fly algorithm (CFOA) is investigated in details on ten well known benchmark problems using fourteen different chaotic maps. Moreover, we performed comparison studies with basic FOA, FOA with Levy flight distribution, and other recently published chaotic algorithms. Statistical results on every optimization task indicate that the chaotic fruit fly algorithm (CFOA) has a very fast convergence rate. In addition, CFOA is compared with recently developed chaos enhanced algorithms such as chaotic bat algorithm, chaotic accelerated particle swarm optimization, chaotic firefly algorithm, chaotic artificial bee colony algorithm, and chaotic cuckoo search. Overall research findings show that FOA with Chebyshev map show superiority in terms of reliability of global optimality and algorithm success rate.",
publisher = "Elsevier, Amsterdam",
journal = "Knowledge-Based Systems",
title = "Chaotic fruit fly optimization algorithm",
pages = "458-446",
volume = "89",
doi = "10.1016/j.knosys.2015.08.010"
}

Mitić, M., Vuković, N., Petrović, M.,& Miljković, Z.. (2015). Chaotic fruit fly optimization algorithm. in Knowledge-Based Systems
Elsevier, Amsterdam., 89, 446-458.
https://doi.org/10.1016/j.knosys.2015.08.010

Mitić M, Vuković N, Petrović M, Miljković Z. Chaotic fruit fly optimization algorithm. in Knowledge-Based Systems. 2015;89:446-458.
doi:10.1016/j.knosys.2015.08.010 .

Mitić, Marko, Vuković, Najdan, Petrović, Milica, Miljković, Zoran, "Chaotic fruit fly optimization algorithm" in Knowledge-Based Systems, 89 (2015):446-458,
https://doi.org/10.1016/j.knosys.2015.08.010 . .

TY  - JOUR
AU  - Petrović, Milica
AU  - Mitić, Marko
AU  - Vuković, Najdan
AU  - Petronijević, Jelena
AU  - Miljković, Zoran
AU  - Babi, Bojan
PY  - 2015
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/4557
AB  - The paper presents an approach based on the application of the Particle Swarm Intelligence algorithm for solving the combinatorial optimization problem of determining the order of execution of operations when processing parts on machines. The proposed approach considers the following types of flexibility: machine flexibility, tool flexibility, process flexibility, and operation sequence flexibility. To represent the flexibility of the machining process of part processing, the method of representing the manufacturing process through networks was chosen, while for the described mathematical model, the criteria for optimization are minimum production time and minimum costs. Experimental results show that the presented algorithm is more efficient, i.e. to give optimal orders of operations in less time and fewer iterations compared to single GA, SA and hybrid GA-SA algorithm.
PB  - Beograd : JUQS
T2  - International Journal Advanced Quality
T1  - Modified Chaotic Particle Swarm Optimization Algorithm for Flexible Process Planning
EP  - 32
IS  - 3
SP  - 25
VL  - 43
UR  - https://hdl.handle.net/21.15107/rcub_machinery_4557
ER  - 

@article{
author = "Petrović, Milica and Mitić, Marko and Vuković, Najdan and Petronijević, Jelena and Miljković, Zoran and Babi, Bojan",
year = "2015",
abstract = "The paper presents an approach based on the application of the Particle Swarm Intelligence algorithm for solving the combinatorial optimization problem of determining the order of execution of operations when processing parts on machines. The proposed approach considers the following types of flexibility: machine flexibility, tool flexibility, process flexibility, and operation sequence flexibility. To represent the flexibility of the machining process of part processing, the method of representing the manufacturing process through networks was chosen, while for the described mathematical model, the criteria for optimization are minimum production time and minimum costs. Experimental results show that the presented algorithm is more efficient, i.e. to give optimal orders of operations in less time and fewer iterations compared to single GA, SA and hybrid GA-SA algorithm.",
publisher = "Beograd : JUQS",
journal = "International Journal Advanced Quality",
title = "Modified Chaotic Particle Swarm Optimization Algorithm for Flexible Process Planning",
pages = "32-25",
number = "3",
volume = "43",
url = "https://hdl.handle.net/21.15107/rcub_machinery_4557"
}

Petrović, M., Mitić, M., Vuković, N., Petronijević, J., Miljković, Z.,& Babi, B.. (2015). Modified Chaotic Particle Swarm Optimization Algorithm for Flexible Process Planning. in International Journal Advanced Quality
Beograd : JUQS., 43(3), 25-32.
https://hdl.handle.net/21.15107/rcub_machinery_4557

Petrović M, Mitić M, Vuković N, Petronijević J, Miljković Z, Babi B. Modified Chaotic Particle Swarm Optimization Algorithm for Flexible Process Planning. in International Journal Advanced Quality. 2015;43(3):25-32.
https://hdl.handle.net/21.15107/rcub_machinery_4557 .

Petrović, Milica, Mitić, Marko, Vuković, Najdan, Petronijević, Jelena, Miljković, Zoran, Babi, Bojan, "Modified Chaotic Particle Swarm Optimization Algorithm for Flexible Process Planning" in International Journal Advanced Quality, 43, no. 3 (2015):25-32,
https://hdl.handle.net/21.15107/rcub_machinery_4557 .

TY  - JOUR
AU  - Petrović, Milica
AU  - Petronijević, Jelena
AU  - Mitić, Marko
AU  - Vuković, Najdan
AU  - Plemić, Aleksandar
AU  - Miljković, Zoran
AU  - Babić, Bojan
PY  - 2015
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/4419
AB  - Obtaining an optimal process plan according to all alternative manufacturing resources has become very important task in flexible process planning problem research. In this paper, we use a novel nature-inspired algorithm called Ant Lion Optimizer (ALO) to solve this NP-hard combinatorial optimization problem. The network representation is adopted to describe flexibilities in process planning and mathematical model for the minimization of the total production time and cost is presented. The algorithm is implemented in Matlab environment and run on the 3.10 GHz processor with 2 GBs of RAM memory. The presented experimental results show that the proposed algorithm performs better in comparison with other bio-inspired optimization algorithms.
PB  - University of Novi Sad - Faculty of Technical Sciences
T2  - Journal of Production Engineering
T1  - The Ant Lion Optimization Algorithm for Flexible Process Planning
EP  - 68
IS  - 2
SP  - 65
VL  - 18
UR  - https://hdl.handle.net/21.15107/rcub_machinery_4419
ER  - 

@article{
author = "Petrović, Milica and Petronijević, Jelena and Mitić, Marko and Vuković, Najdan and Plemić, Aleksandar and Miljković, Zoran and Babić, Bojan",
year = "2015",
abstract = "Obtaining an optimal process plan according to all alternative manufacturing resources has become very important task in flexible process planning problem research. In this paper, we use a novel nature-inspired algorithm called Ant Lion Optimizer (ALO) to solve this NP-hard combinatorial optimization problem. The network representation is adopted to describe flexibilities in process planning and mathematical model for the minimization of the total production time and cost is presented. The algorithm is implemented in Matlab environment and run on the 3.10 GHz processor with 2 GBs of RAM memory. The presented experimental results show that the proposed algorithm performs better in comparison with other bio-inspired optimization algorithms.",
publisher = "University of Novi Sad - Faculty of Technical Sciences",
journal = "Journal of Production Engineering",
title = "The Ant Lion Optimization Algorithm for Flexible Process Planning",
pages = "68-65",
number = "2",
volume = "18",
url = "https://hdl.handle.net/21.15107/rcub_machinery_4419"
}

Petrović, M., Petronijević, J., Mitić, M., Vuković, N., Plemić, A., Miljković, Z.,& Babić, B.. (2015). The Ant Lion Optimization Algorithm for Flexible Process Planning. in Journal of Production Engineering
University of Novi Sad - Faculty of Technical Sciences., 18(2), 65-68.
https://hdl.handle.net/21.15107/rcub_machinery_4419

Petrović M, Petronijević J, Mitić M, Vuković N, Plemić A, Miljković Z, Babić B. The Ant Lion Optimization Algorithm for Flexible Process Planning. in Journal of Production Engineering. 2015;18(2):65-68.
https://hdl.handle.net/21.15107/rcub_machinery_4419 .

Petrović, Milica, Petronijević, Jelena, Mitić, Marko, Vuković, Najdan, Plemić, Aleksandar, Miljković, Zoran, Babić, Bojan, "The Ant Lion Optimization Algorithm for Flexible Process Planning" in Journal of Production Engineering, 18, no. 2 (2015):65-68,
https://hdl.handle.net/21.15107/rcub_machinery_4419 .

TY  - JOUR
AU  - Vuković, Najdan
AU  - Miljković, Zoran
PY  - 2015
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/2212
AB  - Feedforward neural networks (FFNN) are among the most used neural networks for modeling of various nonlinear problems in engineering. In sequential and especially real time processing all neural networks models fail when faced with outliers. Outliers are found across a wide range of engineering problems. Recent research results in the field have shown that to avoid overfitting or divergence of the model, new approach is needed especially if FFNN is to run sequentially or in real time. To accommodate limitations of FFNN when training data contains a certain number of outliers, this paper presents new learning algorithm based on improvement of conventional extended Kalman filter (EKF). Extended Kalman filter robust to outliers (EKF-OR) is probabilistic generative model in which measurement noise covariance is not constant; the sequence of noise measurement covariance is modeled as stochastic process over the set of symmetric positive-definite matrices in which prior is modeled as inverse Wishart distribution. In each iteration EKF-OR simultaneously estimates noise estimates and current best estimate of FFNN parameters. Bayesian framework enables one to mathematically derive expressions, while analytical intractability of the Bayes' update step is solved by using structured variational approximation. All mathematical expressions in the paper are derived using the first principles. Extensive experimental study shows that FFNN trained with developed learning algorithm, achieves low prediction error and good generalization quality regardless of outliers' presence in training data.
PB  - Pergamon-Elsevier Science Ltd, Oxford
T2  - Neural Networks
T1  - Robust sequential learning of feedforward neural networks in the presence of heavy-tailed noise
EP  - 47
SP  - 31
VL  - 63
DO  - 10.1016/j.neunet.2014.11.001
ER  - 

@article{
author = "Vuković, Najdan and Miljković, Zoran",
year = "2015",
abstract = "Feedforward neural networks (FFNN) are among the most used neural networks for modeling of various nonlinear problems in engineering. In sequential and especially real time processing all neural networks models fail when faced with outliers. Outliers are found across a wide range of engineering problems. Recent research results in the field have shown that to avoid overfitting or divergence of the model, new approach is needed especially if FFNN is to run sequentially or in real time. To accommodate limitations of FFNN when training data contains a certain number of outliers, this paper presents new learning algorithm based on improvement of conventional extended Kalman filter (EKF). Extended Kalman filter robust to outliers (EKF-OR) is probabilistic generative model in which measurement noise covariance is not constant; the sequence of noise measurement covariance is modeled as stochastic process over the set of symmetric positive-definite matrices in which prior is modeled as inverse Wishart distribution. In each iteration EKF-OR simultaneously estimates noise estimates and current best estimate of FFNN parameters. Bayesian framework enables one to mathematically derive expressions, while analytical intractability of the Bayes' update step is solved by using structured variational approximation. All mathematical expressions in the paper are derived using the first principles. Extensive experimental study shows that FFNN trained with developed learning algorithm, achieves low prediction error and good generalization quality regardless of outliers' presence in training data.",
publisher = "Pergamon-Elsevier Science Ltd, Oxford",
journal = "Neural Networks",
title = "Robust sequential learning of feedforward neural networks in the presence of heavy-tailed noise",
pages = "47-31",
volume = "63",
doi = "10.1016/j.neunet.2014.11.001"
}

Vuković, N.,& Miljković, Z.. (2015). Robust sequential learning of feedforward neural networks in the presence of heavy-tailed noise. in Neural Networks
Pergamon-Elsevier Science Ltd, Oxford., 63, 31-47.
https://doi.org/10.1016/j.neunet.2014.11.001

Vuković N, Miljković Z. Robust sequential learning of feedforward neural networks in the presence of heavy-tailed noise. in Neural Networks. 2015;63:31-47.
doi:10.1016/j.neunet.2014.11.001 .

Vuković, Najdan, Miljković, Zoran, "Robust sequential learning of feedforward neural networks in the presence of heavy-tailed noise" in Neural Networks, 63 (2015):31-47,
https://doi.org/10.1016/j.neunet.2014.11.001 . .

TY  - JOUR
AU  - Vuković, Najdan
AU  - Mitić, Marko
AU  - Miljković, Zoran
PY  - 2015
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/2256
AB  - In this paper, we present new Learning from Demonstration-based algorithm that generalizes and extracts relevant features of desired motion trajectories for differential drive mobile robots. The algorithm is tested through series of simulations and real world experiments in which desired task is demonstrated by the human teacher while teleoperating the mobile robot in the working environment. In the first step of the developed method, Gaussian Mixture Model (GMM) is built for incremental motions of the mobile robot between two consecutive poses. After this, the hidden Markov model is used to capture transitions between states (temporal variations of the data between clusters) which are missing from static GMM representation. Generalization of the motion is achieved by using the concept of keyframes, defined as points in which significant changes between GMM/HMM states occur. In the second step, the resulting GMM/HMM representation is used to generate optimal state sequences for each demonstration and to temporally align them, using 1D dynamic time warping, with respect to the one most consistent with the GMM/HMM model. This phase implies extraction of keyframes along all state sequences and projecting them into control space, in which controls are aligned in time as well. Finally, the generalized controls are obtained by averaging over all controls at the keyframes; simple piecewise cubic spline method is used for interpolation between generated control values. The main advantage of the developed algorithm is its ability to learn and generalize from all demonstrated examples which results in high quality reproductions of the motion. The proposed approach is verified both in simulated environment and using real mobile robot.
PB  - Pergamon-Elsevier Science Ltd, Oxford
T2  - Engineering Applications of Artificial Intelligence
T1  - Trajectory learning and reproduction for differential drive mobile robots based on GMM/HMM and dynamic time warping using learning from demonstration framework
EP  - 404
SP  - 388
VL  - 45
DO  - 10.1016/j.engappai.2015.07.002
ER  - 

@article{
author = "Vuković, Najdan and Mitić, Marko and Miljković, Zoran",
year = "2015",
abstract = "In this paper, we present new Learning from Demonstration-based algorithm that generalizes and extracts relevant features of desired motion trajectories for differential drive mobile robots. The algorithm is tested through series of simulations and real world experiments in which desired task is demonstrated by the human teacher while teleoperating the mobile robot in the working environment. In the first step of the developed method, Gaussian Mixture Model (GMM) is built for incremental motions of the mobile robot between two consecutive poses. After this, the hidden Markov model is used to capture transitions between states (temporal variations of the data between clusters) which are missing from static GMM representation. Generalization of the motion is achieved by using the concept of keyframes, defined as points in which significant changes between GMM/HMM states occur. In the second step, the resulting GMM/HMM representation is used to generate optimal state sequences for each demonstration and to temporally align them, using 1D dynamic time warping, with respect to the one most consistent with the GMM/HMM model. This phase implies extraction of keyframes along all state sequences and projecting them into control space, in which controls are aligned in time as well. Finally, the generalized controls are obtained by averaging over all controls at the keyframes; simple piecewise cubic spline method is used for interpolation between generated control values. The main advantage of the developed algorithm is its ability to learn and generalize from all demonstrated examples which results in high quality reproductions of the motion. The proposed approach is verified both in simulated environment and using real mobile robot.",
publisher = "Pergamon-Elsevier Science Ltd, Oxford",
journal = "Engineering Applications of Artificial Intelligence",
title = "Trajectory learning and reproduction for differential drive mobile robots based on GMM/HMM and dynamic time warping using learning from demonstration framework",
pages = "404-388",
volume = "45",
doi = "10.1016/j.engappai.2015.07.002"
}

Vuković, N., Mitić, M.,& Miljković, Z.. (2015). Trajectory learning and reproduction for differential drive mobile robots based on GMM/HMM and dynamic time warping using learning from demonstration framework. in Engineering Applications of Artificial Intelligence
Pergamon-Elsevier Science Ltd, Oxford., 45, 388-404.
https://doi.org/10.1016/j.engappai.2015.07.002

Vuković N, Mitić M, Miljković Z. Trajectory learning and reproduction for differential drive mobile robots based on GMM/HMM and dynamic time warping using learning from demonstration framework. in Engineering Applications of Artificial Intelligence. 2015;45:388-404.
doi:10.1016/j.engappai.2015.07.002 .

Vuković, Najdan, Mitić, Marko, Miljković, Zoran, "Trajectory learning and reproduction for differential drive mobile robots based on GMM/HMM and dynamic time warping using learning from demonstration framework" in Engineering Applications of Artificial Intelligence, 45 (2015):388-404,
https://doi.org/10.1016/j.engappai.2015.07.002 . .

TY  - JOUR
AU  - Vuković, Najdan
AU  - Mitić, Marko
AU  - Miljković, Zoran
PY  - 2015
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/2173
AB  - U radu je prikazan i izveden novi sekvencijalni algoritam za obučavanje višeslojnog perceptrona u prisustvu autlajera. Autlajeri predstavljaju značajan problem, posebno ukoliko sprovodimo sekvencijalno obučavanje ili obučavanje u realnom vremenu. Linearizovani Kalmanov filtar robustan na autlajere (LKF-RA), je statistički generativni model u kome je matrica kovarijansi šuma merenja modelovana kao stohastički proces, a apriorna informacija usvojena kao inverzna Višartova raspodela. Izvođenje svih jednakosti je bazirano na prvim principima Bajesovske metodologije. Da bi se rešio korak modifikacije primenjen je varijacioni metod, u kome rešenje problema tražimo u familiji raspodela odgovarajuće funkcionalne forme. Eksperimentalni rezultati primene LKF-RA, dobijeni korišćenjem stvarnih vremenskih serija, pokazuju da je LKF-RA bolji od konvencionalnog linearizovanog Kalmanovog filtra u smislu generisanja niže greške na test skupu podataka. Prosečna vrednost poboljšanja određena u eksperimentalnom procesu je 7%.
AB  - We derive a new sequential learning algorithm for Multilayered Perceptron (MLP) neural network robust to outliers. Presence of outliers in data results in failure of the model especially if data processing is performed on-line or in real time. Extended Kalman filter robust to outliers (EKF-OR) is probabilistic generative model in which measurement noise covariance is modeled as stochastic process over the set of symmetric positive-definite matrices in which prior is given as inverse Wishart distribution. Derivation of expressions comes straight form first principles, within Bayesian framework. Analytical intractability of Bayes' update step is solved using Variational Inference (VI). Experimental results obtained using real world stochastic data show that MLP network trained with proposed algorithm achieves low error and average improvement rate of 7% when compared directly to conventional EKF learning algorithm.
PB  - Univerzitet u Beogradu - Mašinski fakultet, Beograd
T2  - FME Transactions
T1  - Varijacioni pristup robustnom obučavanju višeslojnog perceptrona na bazi Bajesovske metodologije
T1  - Variational inference for robust sequential learning of multilayered perceptron neural network
EP  - 130
IS  - 2
SP  - 123
VL  - 43
DO  - 10.5937/fmet1502123V
ER  - 

@article{
author = "Vuković, Najdan and Mitić, Marko and Miljković, Zoran",
year = "2015",
abstract = "U radu je prikazan i izveden novi sekvencijalni algoritam za obučavanje višeslojnog perceptrona u prisustvu autlajera. Autlajeri predstavljaju značajan problem, posebno ukoliko sprovodimo sekvencijalno obučavanje ili obučavanje u realnom vremenu. Linearizovani Kalmanov filtar robustan na autlajere (LKF-RA), je statistički generativni model u kome je matrica kovarijansi šuma merenja modelovana kao stohastički proces, a apriorna informacija usvojena kao inverzna Višartova raspodela. Izvođenje svih jednakosti je bazirano na prvim principima Bajesovske metodologije. Da bi se rešio korak modifikacije primenjen je varijacioni metod, u kome rešenje problema tražimo u familiji raspodela odgovarajuće funkcionalne forme. Eksperimentalni rezultati primene LKF-RA, dobijeni korišćenjem stvarnih vremenskih serija, pokazuju da je LKF-RA bolji od konvencionalnog linearizovanog Kalmanovog filtra u smislu generisanja niže greške na test skupu podataka. Prosečna vrednost poboljšanja određena u eksperimentalnom procesu je 7%., We derive a new sequential learning algorithm for Multilayered Perceptron (MLP) neural network robust to outliers. Presence of outliers in data results in failure of the model especially if data processing is performed on-line or in real time. Extended Kalman filter robust to outliers (EKF-OR) is probabilistic generative model in which measurement noise covariance is modeled as stochastic process over the set of symmetric positive-definite matrices in which prior is given as inverse Wishart distribution. Derivation of expressions comes straight form first principles, within Bayesian framework. Analytical intractability of Bayes' update step is solved using Variational Inference (VI). Experimental results obtained using real world stochastic data show that MLP network trained with proposed algorithm achieves low error and average improvement rate of 7% when compared directly to conventional EKF learning algorithm.",
publisher = "Univerzitet u Beogradu - Mašinski fakultet, Beograd",
journal = "FME Transactions",
title = "Varijacioni pristup robustnom obučavanju višeslojnog perceptrona na bazi Bajesovske metodologije, Variational inference for robust sequential learning of multilayered perceptron neural network",
pages = "130-123",
number = "2",
volume = "43",
doi = "10.5937/fmet1502123V"
}

Vuković, N., Mitić, M.,& Miljković, Z.. (2015). Varijacioni pristup robustnom obučavanju višeslojnog perceptrona na bazi Bajesovske metodologije. in FME Transactions
Univerzitet u Beogradu - Mašinski fakultet, Beograd., 43(2), 123-130.
https://doi.org/10.5937/fmet1502123V

Vuković N, Mitić M, Miljković Z. Varijacioni pristup robustnom obučavanju višeslojnog perceptrona na bazi Bajesovske metodologije. in FME Transactions. 2015;43(2):123-130.
doi:10.5937/fmet1502123V .

Vuković, Najdan, Mitić, Marko, Miljković, Zoran, "Varijacioni pristup robustnom obučavanju višeslojnog perceptrona na bazi Bajesovske metodologije" in FME Transactions, 43, no. 2 (2015):123-130,
https://doi.org/10.5937/fmet1502123V . .

TY  - JOUR
AU  - Vuković, Najdan
AU  - Miljković, Zoran
PY  - 2014
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/1837
AB  - U ovom radu se analizira problem mašinskog učenja veštačke neuronske mreže sa radijalnim aktivacionim funkcijama Gausovog tipa na bazi Kalmanovog filtra. Prikazana su tri nova sekvencijalna algoritma mašinskog učenja: prvi algoritam direktno primenjuje linearizovani Kalmanov filtar kao algoritam mašinskog učenja, drugi algoritam primenjuje dual Kalmanovom filtru pod nazivom linearizovani informacioni filtar, dok treći algoritam na poseban način aproksimira prvi i drugi moment Gausove raspodele. U radu se naglašavaju osnovne prednosti koje pomenuti algoritmi imaju u poređenju sa konvencionalnim vidovima mašinskog učenja. Za sva tri algoritma razvijen je odgovarajući matematički model veštačke neuronske mreže sa radijalnim aktivacionim funkcijama Gausovog tipa. Analizirane su osnovne postavke izvedenih algoritama u cilju njihove primene na složene probleme u inženjerskoj praksi.
AB  - This paper analyzes machine learning of radial basis function neural network based on Kalman filtering. Three algorithms are derived: linearized Kalman filter, linearized information filter and unscented Kalman filter. We emphasize basic properties of these estimation algorithms, demonstrate how their advantages can be used for optimization of network parameters, derive mathematical models and show how they can be applied to model problems in engineering practice.
PB  - Savez inženjera i tehničara Srbije, Beograd
T2  - Tehnika
T1  - Mašinsko učenje veštačke neuronske mreže sa radijalnim aktivacionim funkcijama gausovog tipa na bazi Kalmanovog filtra - teorijske osnove
T1  - Machine learning of radial basis function neural network based on Kalman filter: Introduction
EP  - 620
IS  - 4
SP  - 613
VL  - 69
DO  - 10.5937/tehnika1404613V
ER  - 

@article{
author = "Vuković, Najdan and Miljković, Zoran",
year = "2014",
abstract = "U ovom radu se analizira problem mašinskog učenja veštačke neuronske mreže sa radijalnim aktivacionim funkcijama Gausovog tipa na bazi Kalmanovog filtra. Prikazana su tri nova sekvencijalna algoritma mašinskog učenja: prvi algoritam direktno primenjuje linearizovani Kalmanov filtar kao algoritam mašinskog učenja, drugi algoritam primenjuje dual Kalmanovom filtru pod nazivom linearizovani informacioni filtar, dok treći algoritam na poseban način aproksimira prvi i drugi moment Gausove raspodele. U radu se naglašavaju osnovne prednosti koje pomenuti algoritmi imaju u poređenju sa konvencionalnim vidovima mašinskog učenja. Za sva tri algoritma razvijen je odgovarajući matematički model veštačke neuronske mreže sa radijalnim aktivacionim funkcijama Gausovog tipa. Analizirane su osnovne postavke izvedenih algoritama u cilju njihove primene na složene probleme u inženjerskoj praksi., This paper analyzes machine learning of radial basis function neural network based on Kalman filtering. Three algorithms are derived: linearized Kalman filter, linearized information filter and unscented Kalman filter. We emphasize basic properties of these estimation algorithms, demonstrate how their advantages can be used for optimization of network parameters, derive mathematical models and show how they can be applied to model problems in engineering practice.",
publisher = "Savez inženjera i tehničara Srbije, Beograd",
journal = "Tehnika",
title = "Mašinsko učenje veštačke neuronske mreže sa radijalnim aktivacionim funkcijama gausovog tipa na bazi Kalmanovog filtra - teorijske osnove, Machine learning of radial basis function neural network based on Kalman filter: Introduction",
pages = "620-613",
number = "4",
volume = "69",
doi = "10.5937/tehnika1404613V"
}

Vuković, N.,& Miljković, Z.. (2014). Mašinsko učenje veštačke neuronske mreže sa radijalnim aktivacionim funkcijama gausovog tipa na bazi Kalmanovog filtra - teorijske osnove. in Tehnika
Savez inženjera i tehničara Srbije, Beograd., 69(4), 613-620.
https://doi.org/10.5937/tehnika1404613V

Vuković N, Miljković Z. Mašinsko učenje veštačke neuronske mreže sa radijalnim aktivacionim funkcijama gausovog tipa na bazi Kalmanovog filtra - teorijske osnove. in Tehnika. 2014;69(4):613-620.
doi:10.5937/tehnika1404613V .

Vuković, Najdan, Miljković, Zoran, "Mašinsko učenje veštačke neuronske mreže sa radijalnim aktivacionim funkcijama gausovog tipa na bazi Kalmanovog filtra - teorijske osnove" in Tehnika, 69, no. 4 (2014):613-620,
https://doi.org/10.5937/tehnika1404613V . .

TY  - JOUR
AU  - Vuković, Najdan
AU  - Miljković, Zoran
PY  - 2014
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/1831
AB  - U ovom radu su prikazani eksperimentalni rezultati primene tri nova sekvencijalna algoritma mašinskog učenja u cilju optimizacije parametara veštačke neuronske mreže sa radijalnim aktivacionim funkcijama Gausovog tipa na bazi Kalmanovog filtra. Uvedena su tri nova sekvencijalna algoritma mašinskog učenja: linearizovani Kalmanov filtar, linearizovani informacioni filtar, algoritam specifične aproksimacije momenata Gausove raspodele. Nakon prikaza odgovarajućih matematičkih modela datih u prvom delu ovog rada, u ovom delu razvijeni algoritmi su testirani u MATLAB® programskom okruženju razvojem odgovarajućeg softverskog koda i primenom test skupova podataka. Iako svi izabrani test skupovi podataka predstavljaju realne probleme, razvijeni algoritmi su testirani i na realnom inženjerskom problemu modeliranja izgleda segmenta obrađene površine. Sva tri algoritma su prilikom modeliranja ovih problema pokazala visok stepen tačnosti.
AB  - In this paper we test three new sequential machine learning algorithms for radial basis function (RBF) neural network based on Kalman filter theory. Three new algorithms are derived: linearized Kalman filter, linearized information filter and unscented Kalman filter. Having introduced and derived mathematical model of each algorithm in the previous part of the paper, in this part we test and assess their performance using standard test sets from machine learning community. RBF neural network and three developed algorithms are implemented in MATLAB® programming environment. Experimental results obtained on real data sets as well as on real engineering problem show that developed algorithms result in more accurate models of the problem being investigated based on radial basis function neural network.
PB  - Savez inženjera i tehničara Srbije, Beograd
T2  - Tehnika
T1  - Mašinsko učenje veštačke neuronske mreže sa radijalnim aktivacionim funkcijama Gausovog tipa na bazi Kalmanovog filtra - rezultati primene
T1  - Machine learning of radial basis function neural network based on Kalman filter: Implementation
EP  - 628
IS  - 4
SP  - 621
VL  - 69
DO  - 10.5937/tehnika1404621V
ER  - 

@article{
author = "Vuković, Najdan and Miljković, Zoran",
year = "2014",
abstract = "U ovom radu su prikazani eksperimentalni rezultati primene tri nova sekvencijalna algoritma mašinskog učenja u cilju optimizacije parametara veštačke neuronske mreže sa radijalnim aktivacionim funkcijama Gausovog tipa na bazi Kalmanovog filtra. Uvedena su tri nova sekvencijalna algoritma mašinskog učenja: linearizovani Kalmanov filtar, linearizovani informacioni filtar, algoritam specifične aproksimacije momenata Gausove raspodele. Nakon prikaza odgovarajućih matematičkih modela datih u prvom delu ovog rada, u ovom delu razvijeni algoritmi su testirani u MATLAB® programskom okruženju razvojem odgovarajućeg softverskog koda i primenom test skupova podataka. Iako svi izabrani test skupovi podataka predstavljaju realne probleme, razvijeni algoritmi su testirani i na realnom inženjerskom problemu modeliranja izgleda segmenta obrađene površine. Sva tri algoritma su prilikom modeliranja ovih problema pokazala visok stepen tačnosti., In this paper we test three new sequential machine learning algorithms for radial basis function (RBF) neural network based on Kalman filter theory. Three new algorithms are derived: linearized Kalman filter, linearized information filter and unscented Kalman filter. Having introduced and derived mathematical model of each algorithm in the previous part of the paper, in this part we test and assess their performance using standard test sets from machine learning community. RBF neural network and three developed algorithms are implemented in MATLAB® programming environment. Experimental results obtained on real data sets as well as on real engineering problem show that developed algorithms result in more accurate models of the problem being investigated based on radial basis function neural network.",
publisher = "Savez inženjera i tehničara Srbije, Beograd",
journal = "Tehnika",
title = "Mašinsko učenje veštačke neuronske mreže sa radijalnim aktivacionim funkcijama Gausovog tipa na bazi Kalmanovog filtra - rezultati primene, Machine learning of radial basis function neural network based on Kalman filter: Implementation",
pages = "628-621",
number = "4",
volume = "69",
doi = "10.5937/tehnika1404621V"
}

Vuković, N.,& Miljković, Z.. (2014). Mašinsko učenje veštačke neuronske mreže sa radijalnim aktivacionim funkcijama Gausovog tipa na bazi Kalmanovog filtra - rezultati primene. in Tehnika
Savez inženjera i tehničara Srbije, Beograd., 69(4), 621-628.
https://doi.org/10.5937/tehnika1404621V

Vuković N, Miljković Z. Mašinsko učenje veštačke neuronske mreže sa radijalnim aktivacionim funkcijama Gausovog tipa na bazi Kalmanovog filtra - rezultati primene. in Tehnika. 2014;69(4):621-628.
doi:10.5937/tehnika1404621V .

Vuković, Najdan, Miljković, Zoran, "Mašinsko učenje veštačke neuronske mreže sa radijalnim aktivacionim funkcijama Gausovog tipa na bazi Kalmanovog filtra - rezultati primene" in Tehnika, 69, no. 4 (2014):621-628,
https://doi.org/10.5937/tehnika1404621V . .

TY  - GEN
AU  - Petrović, Milica
AU  - Petronijević, Jelena
AU  - Vuković, Najdan
AU  - Mitić, Marko
AU  - Miljković, Zoran
AU  - Babić, Bojan
PY  - 2014
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/4756
AB  - Tehničko rešenje - nova metoda (M85), odnosi se na problem generisanja optimalnih planova terminiranja primenom multiagentnih sistema i tehnika veštačke inteligencije, konkretno biološki inspirisanog algoritma na bazi inteligencije roja čestica (engl. PSO – Particle Swarm Optimization) i veštačkih neuronskih mreža (engl. ANN – Artificial Neural Networks). Ova nova metoda pripada oblasti mašinstva i odnosi se na domen integrisanog projektovanja i terminiranja optimalnih fleksibilnih tehnoloških procesa. Predložena multiagentna arhitehtura se sastoji od šest agenata: agent za optimizaciju, agent za učenje, agent za delove, agent za mašine, agent za alate i agent za transport. Agent za učenje zajedno sa agentom za optimizaciju vrši generisanje optimalnih fleksibilnih tehnoloških procesa, dok preostala četiri agenta učestvuju u njihovom terminiranju. Dakle, nakon generisanja optimalnih i približno optimalnih alternativnih tehnoloških procesa obrade delova, u AnyLogic softverskom paketu je izvršeno terminiranje primenom razvijenih agenata. Simulacioni rezultati optimizacije planova terminiranja za odabrane „benchmark“ delove iz literature pokazuju opravdanost primene predložene metodologije u simuliranom modelu tehnološkog okruženja. Razvijana je kroz opsežne aktivnosti u okviru naučnog projekta pod oznakom TR-35004 MPNiTR Vlade Republike Srbije.
T2  - Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер
T1  - Integrisano projektovanje i teriminiranje otimalnih fleksibilnih tehnoloških procesa bazirano na multiagentnim sistemima i tehnikama veštačke inteligencije
UR  - https://hdl.handle.net/21.15107/rcub_machinery_4756
ER  - 

@misc{
author = "Petrović, Milica and Petronijević, Jelena and Vuković, Najdan and Mitić, Marko and Miljković, Zoran and Babić, Bojan",
year = "2014",
abstract = "Tehničko rešenje - nova metoda (M85), odnosi se na problem generisanja optimalnih planova terminiranja primenom multiagentnih sistema i tehnika veštačke inteligencije, konkretno biološki inspirisanog algoritma na bazi inteligencije roja čestica (engl. PSO – Particle Swarm Optimization) i veštačkih neuronskih mreža (engl. ANN – Artificial Neural Networks). Ova nova metoda pripada oblasti mašinstva i odnosi se na domen integrisanog projektovanja i terminiranja optimalnih fleksibilnih tehnoloških procesa. Predložena multiagentna arhitehtura se sastoji od šest agenata: agent za optimizaciju, agent za učenje, agent za delove, agent za mašine, agent za alate i agent za transport. Agent za učenje zajedno sa agentom za optimizaciju vrši generisanje optimalnih fleksibilnih tehnoloških procesa, dok preostala četiri agenta učestvuju u njihovom terminiranju. Dakle, nakon generisanja optimalnih i približno optimalnih alternativnih tehnoloških procesa obrade delova, u AnyLogic softverskom paketu je izvršeno terminiranje primenom razvijenih agenata. Simulacioni rezultati optimizacije planova terminiranja za odabrane „benchmark“ delove iz literature pokazuju opravdanost primene predložene metodologije u simuliranom modelu tehnološkog okruženja. Razvijana je kroz opsežne aktivnosti u okviru naučnog projekta pod oznakom TR-35004 MPNiTR Vlade Republike Srbije.",
journal = "Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер",
title = "Integrisano projektovanje i teriminiranje otimalnih fleksibilnih tehnoloških procesa bazirano na multiagentnim sistemima i tehnikama veštačke inteligencije",
url = "https://hdl.handle.net/21.15107/rcub_machinery_4756"
}

Petrović, M., Petronijević, J., Vuković, N., Mitić, M., Miljković, Z.,& Babić, B.. (2014). Integrisano projektovanje i teriminiranje otimalnih fleksibilnih tehnoloških procesa bazirano na multiagentnim sistemima i tehnikama veštačke inteligencije. in Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер.
https://hdl.handle.net/21.15107/rcub_machinery_4756

Petrović M, Petronijević J, Vuković N, Mitić M, Miljković Z, Babić B. Integrisano projektovanje i teriminiranje otimalnih fleksibilnih tehnoloških procesa bazirano na multiagentnim sistemima i tehnikama veštačke inteligencije. in Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер. 2014;.
https://hdl.handle.net/21.15107/rcub_machinery_4756 .

Petrović, Milica, Petronijević, Jelena, Vuković, Najdan, Mitić, Marko, Miljković, Zoran, Babić, Bojan, "Integrisano projektovanje i teriminiranje otimalnih fleksibilnih tehnoloških procesa bazirano na multiagentnim sistemima i tehnikama veštačke inteligencije" in Техничко решење (M85) је прихваћено од стране Матичног научног одбора за машинство и индустријски софтвер (2014),
https://hdl.handle.net/21.15107/rcub_machinery_4756 .

TY  - CONF
AU  - Vuković, Najdan
AU  - Mitić, Marko
AU  - Miljković, Zoran
PY  - 2014
UR  - https://machinery.mas.bg.ac.rs/handle/123456789/4633
AB  - In this paper, we present new Learning from Demonstration ((LfD) - based algorithm that generalizes and extracts relevant features of desired motion trajectories for differential drive mobile robots. The algorithm is tested through series of simulations and real world experiments in which desired task is demonstrated by the human teacher while teleoperating the mobile robot in the working environment. The human teacher shows the specifics of the new skill or motion behavior that robot needs to learn; the robot extracts information from demonstrations and builds the internal model based on the learning algorithm and in this paper we will refer to it as the Learning from Demonstration.
PB  - The Aristotle University of Thessaloniki
C3  - Proceedings of the 5th International Conference on Manufacturing Engineering (ICMEN 2014)
T1  - Learning Motion from Demonstration for Differential Drive Mobile Robot
EP  - 108
SP  - 99
UR  - https://hdl.handle.net/21.15107/rcub_machinery_4633
ER  - 

@conference{
author = "Vuković, Najdan and Mitić, Marko and Miljković, Zoran",
year = "2014",
abstract = "In this paper, we present new Learning from Demonstration ((LfD) - based algorithm that generalizes and extracts relevant features of desired motion trajectories for differential drive mobile robots. The algorithm is tested through series of simulations and real world experiments in which desired task is demonstrated by the human teacher while teleoperating the mobile robot in the working environment. The human teacher shows the specifics of the new skill or motion behavior that robot needs to learn; the robot extracts information from demonstrations and builds the internal model based on the learning algorithm and in this paper we will refer to it as the Learning from Demonstration.",
publisher = "The Aristotle University of Thessaloniki",
journal = "Proceedings of the 5th International Conference on Manufacturing Engineering (ICMEN 2014)",
title = "Learning Motion from Demonstration for Differential Drive Mobile Robot",
pages = "108-99",
url = "https://hdl.handle.net/21.15107/rcub_machinery_4633"
}

Vuković, N., Mitić, M.,& Miljković, Z.. (2014). Learning Motion from Demonstration for Differential Drive Mobile Robot. in Proceedings of the 5th International Conference on Manufacturing Engineering (ICMEN 2014)
The Aristotle University of Thessaloniki., 99-108.
https://hdl.handle.net/21.15107/rcub_machinery_4633

Vuković N, Mitić M, Miljković Z. Learning Motion from Demonstration for Differential Drive Mobile Robot. in Proceedings of the 5th International Conference on Manufacturing Engineering (ICMEN 2014). 2014;:99-108.
https://hdl.handle.net/21.15107/rcub_machinery_4633 .

Vuković, Najdan, Mitić, Marko, Miljković, Zoran, "Learning Motion from Demonstration for Differential Drive Mobile Robot" in Proceedings of the 5th International Conference on Manufacturing Engineering (ICMEN 2014) (2014):99-108,
https://hdl.handle.net/21.15107/rcub_machinery_4633 .