The article examines current approaches to training robotic manipulators for executing complex tasks in dynamic and changing environments. It provides a comparative analysis of modern training methods, highlighting their advantages and disadvantages. Additionally, the paper outlines the typical areas in which these methods are applied. Particular attention is given to approaches that involve human instructors, self-learning, and reinforcement learning. Special emphasis is placed on training efficiency, robot adaptability to new conditions, human-robot interaction, and the transfer of skills from virtual training environments to the real world. Based on the analysis, the authors recommend imitation learning — specifically, the learning from demonstration approach — as it enables the rapid and safe transfer of skills from humans to robots without the need for task formalization. The article also highlights the challenges of adapting trained models to real-world conditions and ensuring effective human-robot collaboration. It identifies key challenges faced by modern robot training systems. Based on these challenges, the article offers recommendations for selecting optimal training strategies according to the specific task type and available resources.
- Argall, B. D., Chernova, S., Veloso, M., & Browning, B. (2009). A survey of robot learning from demonstration. Robotics and Autonomous Systems, 57(5), 469–483. https://doi.org/10.1016/j.robot.2008.10.024
- Billard, A., Calinon, S., Dillmann, R., & Schaal, S. (2008). Robot programming by demonstration. In Springer handbook of Robotics (pp. 1371-1394). Springer. https://doi.org/10.1007/978-3-540-30301-5_60
- Barekatain, A., Habibi, H., & Voos, H. (2024). A practical roadmap to learning from demonstration for robotic manipulators in manufacturing. Robotics, 13(3), 100. https://doi.org/10.3390/robotics13070100
- Underactuated Robotics. (n.d.). Ch. 21 – Imitation Learning. Retrieved from https://underactuated.mit.edu/imitation.html.
- Ross, S., Gordon, G., & Bagnell, D. A reduction of imitation learning and structured prediction to no-regret online learning. AISTATS. 2011. – P. 627–635.
- Sutton, R. S., & Barto, A. G. (2018). Reinforcement Learning: An Introduction (2nd Edition). MIT Press.
- Kober, J., Bagnell, J. A., & Peters, J. (2013). Reinforcement learning in robotics: A survey. International Journal of Robotics Research, 32(11), 1238–1274. https://doi.org/10.1177/0278364913495721
- Lillicrap, T. P., Hunt, J. J., Pritzel, A., et al. (2015). Continuous control with deep reinforcement learning.
- Schulman, J., Wolski, F., Dhariwal, P., Radford, A., & Klimov, O. (2017). Proximal policy optimization algorithms.
- Haarnoja, T., Zhou, A., Abbeel, P., & Levine, S. (2018). Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor.
- Ng, A., & Russell, S. (2000). Algorithms for inverse reinforcement learning. Proceedings of the 17th International Conference on Machine Learning (ICML), 663–670.
- Panait, L., & Luke, S. (2005). Cooperative multi-agent learning: The state of the art. Autonomous Agents and Multi-Agent Systems, 11(3), 387–434. https://doi.org/10.1007/s10458-005-2631-2
- Nikolaidis, S., & Shah, J. A. (2013). Human-robot cross-training: computational formulation, modeling and evaluation of a human team training strategy. ACM/IEEE International Conference on Human-Robot Interaction. https://doi.org/10.1109/HRI.2013.6483499
- DEEPCOBOT – Collective Efficient Deep Learning and Networked Control for Multiple Collaborative Robot Systems. UiA WISENET Lab. 2024. URL: https://deepcobot.uia.no/about
- Zhu, Y., Mottaghi, R., Kolve, E., et al. (2018). Reinforcement and imitation learning for diverse visuomotor skills. Proceedings of Robotics: Science and Systems (RSS), 33–40. https://doi.org/10.15607/RSS.2018.XIV.009
- Levine, S., Finn, C., Darrell, T., & Abbeel, P. (2016). End-to-End Training of Deep Visuomotor Policies. Journal of Machine Learning Research, 17(39), 1–40.
- Yaqing Wang, Quanming Yao, James Kwok, Lionel M. Ni (2019). Generalizing from a Few Examples: A Survey on Few-Shot Learning. Science, 53(3), 1–34. https://doi.org/10.1145/3386252
- Finn, C., Abbeel, P., & Levine, S. (2017). Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks. Proceedings of the 34th International Conference on Machine Learning (ICML), 1126–1135.
- Naeem, S., Ali, A., & Anam, S. (2023). An unsupervised machine learning algorithms: Comprehensive review. International Journal of Computing and Digital Systems, 12(1), 1–10. https://doi.org/10.12785/ijcds/130172
- Hjelm, R. D., Fedorov, A., Lavoie, E., et al. (2019). Learning deep representations by mutual information estimation and maximization. International Conference on Learning Representations (ICLR).
- James, S., Davison, A. J., & Johns, E. (2019). "Sim-to-Real via Sim-to-Sim: Data-efficient Robotic Grasping via Randomized-to-Canonical Adaptation Networks." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019. https://doi.org/10.1109/CVPR.2019.01291
- Pan, S. J., & Yang, Q. (2010). "A Survey on Transfer Learning." IEEE Transactions on Knowledge and Data Engineering, 22(10), 1345–1359. https://doi.org/10.1109/TKDE.2009.191