Keywords: offline reinforcement learning, visual navigation, motion planning
TL;DR: Coupling the offline learned value function with a topological graph, where the values provide distance estimates, is a great way to make things scale. And it works with diverse rewards, like "stay in the sun", or "stay off my lawn"."
Abstract: Reinforcement learning can enable robots to navigate to distant goals while optimizing user-specified reward functions, including preferences for following lanes, staying on paved paths, or avoiding freshly mowed grass. However, online learning from trial-and-error for real-world robots is logistically challenging, and methods that instead can utilize existing datasets of robotic navigation data could be significantly more scalable and enable broader generalization. In this paper, we present ReViND, the first offline RL system for robotic navigation that can leverage previously collected data to optimize user-specified reward functions in the real-world. We evaluate our system for off-road navigation without any additional data collection or fine-tuning, and show that it can navigate to distant goals using only offline training from this dataset, and exhibit behaviors that qualitatively differ based on the user-specified reward function.
Student First Author: yes
Supplementary Material: zip