Learning Motion Trajectories of Differential Drive Mobile Robot Using Gaussian Mixtures and Hidden Markov Model
Nema prikaza
2013
Konferencijski prilog (Objavljena verzija)
Metapodaci
Prikaz svih podataka o dokumentuApstrakt
In this paper we focus on learning of motion trajectories for differential drive mobile robot. Learning from Demonstration (LfD) is applied to enable mobile robot to learn and reproduce trajectories which are hard to model due to unknown dynamics and uncertainty. To solve this problem, we build the probabilistic model of the mobile robot motion in two steps. Firstly, Gaussian Mixture Model (GMM) of incremental robot motions is built, while in the second step the hidden Markov Model (HMM) is applied to extract transition matrix and most likely sequence of GMMs for each training trajectory. The final step assumes estimating the optimal sequence of incremental motions for each of N different training trajectories. To summarize the main idea: each of N hidden Markov models model the desired robot motion given as a sequence of M low level robot motions. In contrast to conventional mixture models, hidden Markov model captures temporal dependencies between mixtures, which is i a significant a...dvantage of HMM over conventional mixture models. Experimental results demonstrate applicability and optimal performance of the proposed learning algorithm.
Ključne reči:
Mobile robot navigation / Learning from Demonstration (LfD) / Unknown dynamics and uncertainty / The probabilistic model / Gaussian Mixture Model (GMM) / Hidden Markov Model (HMM) / Learning of motion trajectories / Differential drive mobile robot.Izvor:
Proceedings of the 4th Serbian Congress on Theoretical and Applied Mechanics, Vrnjačka Banja, Serbia, 4-7 June 2013, 2013, A-13:1-A-13:6Izdavač:
- Belgrade : Serbian Society of Mechanics
Finansiranje / projekti:
- Inovativni pristup u primeni inteligentnih tehnoloških sistema za proizvodnju delova od lima zasnovan na ekološkim principima (RS-MESTD-Technological Development (TD or TR)-35004)
Kolekcije
Institucija/grupa
Mašinski fakultetTY - CONF AU - Vuković, Najdan AU - Miljković, Zoran AU - Mitić, Marko AU - Petrović, Milica PY - 2013 UR - https://machinery.mas.bg.ac.rs/handle/123456789/4467 AB - In this paper we focus on learning of motion trajectories for differential drive mobile robot. Learning from Demonstration (LfD) is applied to enable mobile robot to learn and reproduce trajectories which are hard to model due to unknown dynamics and uncertainty. To solve this problem, we build the probabilistic model of the mobile robot motion in two steps. Firstly, Gaussian Mixture Model (GMM) of incremental robot motions is built, while in the second step the hidden Markov Model (HMM) is applied to extract transition matrix and most likely sequence of GMMs for each training trajectory. The final step assumes estimating the optimal sequence of incremental motions for each of N different training trajectories. To summarize the main idea: each of N hidden Markov models model the desired robot motion given as a sequence of M low level robot motions. In contrast to conventional mixture models, hidden Markov model captures temporal dependencies between mixtures, which is i a significant advantage of HMM over conventional mixture models. Experimental results demonstrate applicability and optimal performance of the proposed learning algorithm. PB - Belgrade : Serbian Society of Mechanics C3 - Proceedings of the 4th Serbian Congress on Theoretical and Applied Mechanics, Vrnjačka Banja, Serbia, 4-7 June 2013 T1 - Learning Motion Trajectories of Differential Drive Mobile Robot Using Gaussian Mixtures and Hidden Markov Model EP - A-13:6 SP - A-13:1 UR - https://hdl.handle.net/21.15107/rcub_machinery_4467 ER -
@conference{ author = "Vuković, Najdan and Miljković, Zoran and Mitić, Marko and Petrović, Milica", year = "2013", abstract = "In this paper we focus on learning of motion trajectories for differential drive mobile robot. Learning from Demonstration (LfD) is applied to enable mobile robot to learn and reproduce trajectories which are hard to model due to unknown dynamics and uncertainty. To solve this problem, we build the probabilistic model of the mobile robot motion in two steps. Firstly, Gaussian Mixture Model (GMM) of incremental robot motions is built, while in the second step the hidden Markov Model (HMM) is applied to extract transition matrix and most likely sequence of GMMs for each training trajectory. The final step assumes estimating the optimal sequence of incremental motions for each of N different training trajectories. To summarize the main idea: each of N hidden Markov models model the desired robot motion given as a sequence of M low level robot motions. In contrast to conventional mixture models, hidden Markov model captures temporal dependencies between mixtures, which is i a significant advantage of HMM over conventional mixture models. Experimental results demonstrate applicability and optimal performance of the proposed learning algorithm.", publisher = "Belgrade : Serbian Society of Mechanics", journal = "Proceedings of the 4th Serbian Congress on Theoretical and Applied Mechanics, Vrnjačka Banja, Serbia, 4-7 June 2013", title = "Learning Motion Trajectories of Differential Drive Mobile Robot Using Gaussian Mixtures and Hidden Markov Model", pages = "A-13:6-A-13:1", url = "https://hdl.handle.net/21.15107/rcub_machinery_4467" }
Vuković, N., Miljković, Z., Mitić, M.,& Petrović, M.. (2013). Learning Motion Trajectories of Differential Drive Mobile Robot Using Gaussian Mixtures and Hidden Markov Model. in Proceedings of the 4th Serbian Congress on Theoretical and Applied Mechanics, Vrnjačka Banja, Serbia, 4-7 June 2013 Belgrade : Serbian Society of Mechanics., A-13:1-A-13:6. https://hdl.handle.net/21.15107/rcub_machinery_4467
Vuković N, Miljković Z, Mitić M, Petrović M. Learning Motion Trajectories of Differential Drive Mobile Robot Using Gaussian Mixtures and Hidden Markov Model. in Proceedings of the 4th Serbian Congress on Theoretical and Applied Mechanics, Vrnjačka Banja, Serbia, 4-7 June 2013. 2013;:A-13:1-A-13:6. https://hdl.handle.net/21.15107/rcub_machinery_4467 .
Vuković, Najdan, Miljković, Zoran, Mitić, Marko, Petrović, Milica, "Learning Motion Trajectories of Differential Drive Mobile Robot Using Gaussian Mixtures and Hidden Markov Model" in Proceedings of the 4th Serbian Congress on Theoretical and Applied Mechanics, Vrnjačka Banja, Serbia, 4-7 June 2013 (2013):A-13:1-A-13:6, https://hdl.handle.net/21.15107/rcub_machinery_4467 .