Classical navigation systems typically operate using a fixed set of hand-picked parameters (e.g. maximum speed, sampling rate, inflation radius, etc.) and require heavy expert re-tuning in order to work in new environments. To mitigate this requirement, it has been proposed to learn parameters for different contexts in a new environment using human demonstrations collected via teleoperation. However, learning from human demonstration limits deployment to the training environment, and limits overall performance to that of a potentially-suboptimal demonstrator. In this paper, we introduce APPLR, Adaptive Planner Parameter Learning from Reinforcement, which allows existing navigation systems to adapt to new scenarios by using a parameter selection scheme discovered via reinforcement learning (RL) in a wide variety of simulation environments. We evaluate APPLR on a robot in both simulated and physical experiments, and show that it can outperform both a fixed set of hand-tuned parameters and also a dynamic parameter tuning scheme learned from human demonstration.

PDF

In Proceedings of the 2021 IEEE International Conference on Robotics and Automation (ICRA 2021), Xi'an, China, June 2021.

@inproceedings{icra21-xu, title={APPLR: Adaptive Planner Parameter Learning from Reinforcement}, author={Zifan Xu and Gauraang Dhamankar and Anirudh Nair and Xuesu Xiao and Garrett Warnell and Bo Liu and Zizhao Wang and Peter Stone}, booktitle={Proceedings of the 2021 IEEE International Conference on Robotics and Automation (ICRA 2021)}, month={June}, address={Xi'an, China}, url="http://nn.cs.utexas.edu/?icra21-xu", year={2021} }