Efficient, Decentralized, and Collaborative Multi-Robot Exploration using Optimal Transport Theory

TL;DR

This paper introduces a decentralized multi-robot exploration strategy based on optimal transport theory, enabling efficient, collaborative coverage of prioritized areas without central control, validated through simulations.

Contribution

It presents a novel OT-based framework for decentralized multi-robot exploration that accounts for priority maps and enables collaboration without supervision.

Findings

Efficient coverage of priority areas demonstrated in simulations

Decentralized approach reduces reliance on central coordination

Quantitative performance measurement method developed

Abstract

An Optimal Transport (OT)-based decentralized collaborative multi-robot exploration strategy is proposed in this paper. This method is to achieve an efficient exploration with a predefined priority in the given domain. In this context, the efficiency indicates how a team of robots (agents) cover the domain reflecting the corresponding priority map (or degrees of importance) in the domain. The decentralized exploration implies that each agent carries out their exploration task independently in the absence of any supervisory agent/computer. When an agent encounters another agent within a communication range, each agent receives the information about which areas are already covered by other agents, yielding a collaborative exploration. The OT theory is employed to quantify the difference between the distribution formed by the robot trajectories and the given reference spatial distribution indicating the priority. A computationally feasible way is developed to measure the performance of the proposed exploration scheme. Further, the formal algorithm is provided for the efficient, decentralized, and collaborative exploration plan. Simulation results are presented to validate the proposed methods.