Data-Efficient Learning from Human Interventions for Mobile Robots
This addresses data efficiency and safety issues for mobile robots in applications like autonomous delivery, though it appears incremental as it builds on existing IL and RL approaches.
The paper tackles the problem of data inefficiency and safety in learning-based methods for mobile robots by proposing PVP4Real, an online human-in-the-loop method that combines imitation and reinforcement learning, achieving training completion within 15 minutes for tasks including using raw RGBD images.
Mobile robots are essential in applications such as autonomous delivery and hospitality services. Applying learning-based methods to address mobile robot tasks has gained popularity due to its robustness and generalizability. Traditional methods such as Imitation Learning (IL) and Reinforcement Learning (RL) offer adaptability but require large datasets, carefully crafted reward functions, and face sim-to-real gaps, making them challenging for efficient and safe real-world deployment. We propose an online human-in-the-loop learning method PVP4Real that combines IL and RL to address these issues. PVP4Real enables efficient real-time policy learning from online human intervention and demonstration, without reward or any pretraining, significantly improving data efficiency and training safety. We validate our method by training two different robots -- a legged quadruped, and a wheeled delivery robot -- in two mobile robot tasks, one of which even uses raw RGBD image as observation. The training finishes within 15 minutes. Our experiments show the promising future of human-in-the-loop learning in addressing the data efficiency issue in real-world robotic tasks. More information is available at: https://metadriverse.github.io/pvp4real/