Mobile robot decision-making system based on deep machine learning
Апстракт
One of the major aspects of Industry 4.0 is enabling
the manufacturing entities to operate in the dynamical systems
autonomously. Therefore, to be autonomous, manufacturing
entities need to have sensors to perceive their environment and
utilize that information to make decisions regarding their
actions. Having that in mind, in this paper, the authors propose a
mobile robot decision-making system based on the integration of
visual data and mobile robot pose. Mobile robot pose (current
position and orientation) is integrated with two images gathered
by two cameras and utilized to predict the possibility of gripping
the part to be manufactured. A decision-making system is
created by utilizing the deep learning model Resnet18 with an
additional input for the mobile robot pose. The model is trained
end-to-end and experimental evaluation is performed by using
the mobile robot RACIO (Robot with Artificial Intelligence
based COgnition).
Кључне речи:
Decision-making system / Mobile robots / Deep learningИзвор:
Proceedings / 9th International Conference on Electrical, Electronics and Computing Engineering (IcETRAN 2022) Novi Pazar, Serbia, 6-9, June, 2022, 2022, 635-638Финансирање / пројекти:
- Министарство науке, технолошког развоја и иновација Републике Србије, институционално финансирање - 200105 (Универзитет у Београду, Машински факултет) (RS-200105)
- MISSION4.0 - Deep Machine Learning and Swarm Intelligence-Based Optimization Algorithms for Control and Scheduling of Cyber-Physical Systems in Industry 4.0 (RS-6523109)
Колекције
Институција/група
Mašinski fakultetTY - CONF AU - Jokić, Aleksandar AU - Petrović, Milica AU - Miljković, Zoran PY - 2022 UR - https://machinery.mas.bg.ac.rs/handle/123456789/4230 AB - One of the major aspects of Industry 4.0 is enabling the manufacturing entities to operate in the dynamical systems autonomously. Therefore, to be autonomous, manufacturing entities need to have sensors to perceive their environment and utilize that information to make decisions regarding their actions. Having that in mind, in this paper, the authors propose a mobile robot decision-making system based on the integration of visual data and mobile robot pose. Mobile robot pose (current position and orientation) is integrated with two images gathered by two cameras and utilized to predict the possibility of gripping the part to be manufactured. A decision-making system is created by utilizing the deep learning model Resnet18 with an additional input for the mobile robot pose. The model is trained end-to-end and experimental evaluation is performed by using the mobile robot RACIO (Robot with Artificial Intelligence based COgnition). C3 - Proceedings / 9th International Conference on Electrical, Electronics and Computing Engineering (IcETRAN 2022) Novi Pazar, Serbia, 6-9, June, 2022 T1 - Mobile robot decision-making system based on deep machine learning EP - 638 SP - 635 UR - https://hdl.handle.net/21.15107/rcub_machinery_4230 ER -
@conference{ author = "Jokić, Aleksandar and Petrović, Milica and Miljković, Zoran", year = "2022", abstract = "One of the major aspects of Industry 4.0 is enabling the manufacturing entities to operate in the dynamical systems autonomously. Therefore, to be autonomous, manufacturing entities need to have sensors to perceive their environment and utilize that information to make decisions regarding their actions. Having that in mind, in this paper, the authors propose a mobile robot decision-making system based on the integration of visual data and mobile robot pose. Mobile robot pose (current position and orientation) is integrated with two images gathered by two cameras and utilized to predict the possibility of gripping the part to be manufactured. A decision-making system is created by utilizing the deep learning model Resnet18 with an additional input for the mobile robot pose. The model is trained end-to-end and experimental evaluation is performed by using the mobile robot RACIO (Robot with Artificial Intelligence based COgnition).", journal = "Proceedings / 9th International Conference on Electrical, Electronics and Computing Engineering (IcETRAN 2022) Novi Pazar, Serbia, 6-9, June, 2022", title = "Mobile robot decision-making system based on deep machine learning", pages = "638-635", url = "https://hdl.handle.net/21.15107/rcub_machinery_4230" }
Jokić, A., Petrović, M.,& Miljković, Z.. (2022). Mobile robot decision-making system based on deep machine learning. in Proceedings / 9th International Conference on Electrical, Electronics and Computing Engineering (IcETRAN 2022) Novi Pazar, Serbia, 6-9, June, 2022, 635-638. https://hdl.handle.net/21.15107/rcub_machinery_4230
Jokić A, Petrović M, Miljković Z. Mobile robot decision-making system based on deep machine learning. in Proceedings / 9th International Conference on Electrical, Electronics and Computing Engineering (IcETRAN 2022) Novi Pazar, Serbia, 6-9, June, 2022. 2022;:635-638. https://hdl.handle.net/21.15107/rcub_machinery_4230 .
Jokić, Aleksandar, Petrović, Milica, Miljković, Zoran, "Mobile robot decision-making system based on deep machine learning" in Proceedings / 9th International Conference on Electrical, Electronics and Computing Engineering (IcETRAN 2022) Novi Pazar, Serbia, 6-9, June, 2022 (2022):635-638, https://hdl.handle.net/21.15107/rcub_machinery_4230 .