Learning Vision-based Pursuit-Evasion Robot Policies
This work addresses the problem of enabling robots to perform complex strategic tasks like pursuit-evasion in real-world environments, which is incremental as it builds on existing supervised learning approaches but applies them to a challenging robotics domain.
The paper tackled the challenge of learning strategic robot behaviors for pursuit-evasion interactions under real-world constraints by transforming it into a supervised learning problem, where a fully-observable policy generates supervision for a partially-observable one, and deployed the policy on a physical quadruped robot with an RGB-D camera, achieving successful pursuit-evasion in the wild.
Learning strategic robot behavior -- like that required in pursuit-evasion interactions -- under real-world constraints is extremely challenging. It requires exploiting the dynamics of the interaction, and planning through both physical state and latent intent uncertainty. In this paper, we transform this intractable problem into a supervised learning problem, where a fully-observable robot policy generates supervision for a partially-observable one. We find that the quality of the supervision signal for the partially-observable pursuer policy depends on two key factors: the balance of diversity and optimality of the evader's behavior and the strength of the modeling assumptions in the fully-observable policy. We deploy our policy on a physical quadruped robot with an RGB-D camera on pursuit-evasion interactions in the wild. Despite all the challenges, the sensing constraints bring about creativity: the robot is pushed to gather information when uncertain, predict intent from noisy measurements, and anticipate in order to intercept. Project webpage: https://abajcsy.github.io/vision-based-pursuit/