SURVEY: crowd robot navigation
https://gyazo.com/8e64c9835b21926de6c48dd11c3b5207
Sathyamoorthy20icra_DenseCAvoid: Real-time Navigation in Dense Crowds using Anticipatory Behaviors
policyを学習する手法の例
歩行者の計測に基づき,将来の時間における歩行者の位置を外挿するバウンディングボックスを計算
3Dシミュレータでロボットのpolicyを学習
A recent trend is to use learning methods for sensor-based robot navigation in crowds. These include techniques based on end-to-end deep learning 14 , 21 , generative adversarial imitation learning 24 , and deep reinforcement learning 10 , 17 . (14) Y. Kim, J. Jang, and S. Yun. 2018. End-to-end deep learning for autonomous navigation of mobile robot. In 2018 IEEE International Conference on Consumer Electronics (ICCE). 1–6. DOI:http://dx.doi.org/10.1109/ICCE.2018.8326229 (24) L. Tai, J. Zhang, M. Liu, and W. Burgard. 2018. Socially Compliant Navigation Through Raw Depth Inputs with Generative Adversarial Imitation Learning. In ICRA. 1111–1117. DOI:http://dx.doi.org/10.1109/ICRA.2018.8460968 EPFLのChenらの研究
ICRA2020, RAL
ICRA2019
van den Berg, Reciprocal n-body Collision Avoidance (ORCA)
In this paper, we present the principle of optimal reciprocal collision avoidance (ORCA) that overcomes this limitation; ORCA provides a sufficient condition for multiple robots to avoid collisions among one another, and thus can guarantee collision-free navigation.
CADRL
M. Everett, Y. F. Chen, and J. P. How, “Motion planning among
dynamic, decision-making agents with deep reinforcement learning,”
2018.