Decentralized Non-communicating Multiagent Collision Avoidance with Deep Reinforcement Learning

TL;DR

This paper introduces a decentralized deep reinforcement learning-based collision avoidance method for multiagent systems that predicts interaction patterns offline, enabling real-time, collision-free path planning even in non-communicating scenarios.

Contribution

It presents a novel value network approach that encodes time-to-goal estimates considering neighboring agents, improving path efficiency and computational feasibility.

Findings

Over 26% improvement in path quality over ORCA

Efficient real-time collision avoidance queries

Effective handling of uncertainty in agent motion

Abstract

Finding feasible, collision-free paths for multiagent systems can be challenging, particularly in non-communicating scenarios where each agent's intent (e.g. goal) is unobservable to the others. In particular, finding time efficient paths often requires anticipating interaction with neighboring agents, the process of which can be computationally prohibitive. This work presents a decentralized multiagent collision avoidance algorithm based on a novel application of deep reinforcement learning, which effectively offloads the online computation (for predicting interaction patterns) to an offline learning procedure. Specifically, the proposed approach develops a value network that encodes the estimated time to the goal given an agent's joint configuration (positions and velocities) with its neighbors. Use of the value network not only admits efficient (i.e., real-time implementable) queries for finding a collision-free velocity vector, but also considers the uncertainty in the other agents' motion. Simulation results show more than 26 percent improvement in paths quality (i.e., time to reach the goal) when compared with optimal reciprocal collision avoidance (ORCA), a state-of-the-art collision avoidance strategy.