Distributed multi-agent target search and tracking with Gaussian process and reinforcement learning

TL;DR

This paper introduces a multi-agent reinforcement learning approach combined with distributed Gaussian processes for efficient target search and tracking, demonstrated on UAV swarms with both simulation and hardware tests.

Contribution

It presents a novel integration of distributed Gaussian processes with reinforcement learning for multi-agent target tracking, enabling effective planning over unknown targets.

Findings

Effective target localization in simulation

Successful transfer to hardware UAVs

Improved planning over unknown targets

Abstract

Deploying multiple robots for target search and tracking has many practical applications, yet the challenge of planning over unknown or partially known targets remains difficult to address. With recent advances in deep learning, intelligent control techniques such as reinforcement learning have enabled agents to learn autonomously from environment interactions with little to no prior knowledge. Such methods can address the exploration-exploitation tradeoff of planning over unknown targets in a data-driven manner, eliminating the reliance on heuristics typical of traditional approaches and streamlining the decision-making pipeline with end-to-end training. In this paper, we propose a multi-agent reinforcement learning technique with target map building based on distributed Gaussian process. We leverage the distributed Gaussian process to encode belief over the target locations and efficiently plan over unknown targets. We evaluate the performance and transferability of the trained policy in simulation and demonstrate the method on a swarm of micro unmanned aerial vehicles with hardware experiments.