APPLE: Adaptive Planner Parameter Learning from Evaluative Feedback
arXiv:2108.09801 · doi:10.1109/LRA.2021.3100940
Abstract
Classical autonomous navigation systems can control robots in a collision-free manner, oftentimes with verifiable safety and explainability. When facing new environments, however, fine-tuning of the system parameters by an expert is typically required before the system can navigate as expected. To alleviate this requirement, the recently-proposed Adaptive Planner Parameter Learning paradigm allows robots to \emph{learn} how to dynamically adjust planner parameters using a teleoperated demonstration or corrective interventions from non-expert users. However, these interaction modalities require users to take full control of the moving robot, which requires the users to be familiar with robot teleoperation. As an alternative, we introduce \textsc{apple}, Adaptive Planner Parameter Learning from \emph{Evaluative Feedback} (real-time, scalar-valued assessments of behavior), which represents a less-demanding modality of interaction. Simulated and physical experiments show \textsc{apple} can achieve better performance compared to the planner with static default parameters and even yield improvement over learned parameters from richer interaction modalities.
arXiv comment6 pages, 4 figures, accepted in IROS 2021. arXiv admin note: substantial text overlap with arXiv:2105.07620
References in corpus (6)
- Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
- End to End Learning for Self-Driving Cars
- From Perception to Decision: A Data-driven Approach to End-to-end Motion Planning for Autonomous Ground Robots
- Interactive Learning from Policy-Dependent Human Feedback
- Intention-Net: Integrating Planning and Deep Learning for Goal-Directed Autonomous Navigation
- ROS Navigation Tuning Guide