Human-Inspired Multi-Agent Navigation using Knowledge Distillation

Abstract

Despite significant advancements in the field of multi-agent navigation, agents still lack the sophistication and intelligence that humans exhibit in multi-agent settings. In this paper, we propose a framework for learning a human-like general collision avoidance policy for agent-agent interactions in fully decentralized, multi-agent environments. Our approach uses knowledge distillation with reinforcement learning to shape the reward function based on expert policies extracted from human trajectory demonstrations through behavior cloning. We show that agents trained with our approach can take human-like trajectories in collision avoidance and goal-directed steering tasks not provided by the demonstrations, outperforming the experts as well as learning-based agents trained without knowledge distillation.

arXiv commentIEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021

References in corpus (11)

• Distilling the Knowledge in a Neural Network
• Simulating Dynamical Features of Escape Panic
• End to End Learning for Self-Driving Cars
• High-Dimensional Continuous Control Using Generalized Advantage Estimation
• How simple rules determine pedestrian behavior and crowd disasters
• Emergence of Locomotion Behaviours in Rich Environments
• From Perception to Decision: A Data-driven Approach to End-to-end Motion Planning for Autonomous Ground Robots
• Universal power law governing pedestrian interactions
• Anomaly Detection with Score functions based on Nearest Neighbor Graphs
• Fully Distributed Multi-Robot Collision Avoidance via Deep Reinforcement Learning for Safe and Efficient Navigation in Complex Scenarios
• Distilling Policy Distillation

Cited by in corpus (2)

• DRL-VO: Learning to Navigate Through Crowded Dynamic Scenes Using Velocity Obstacles
• Reward Function Design for Crowd Simulation via Reinforcement Learning