D2M2N: Decentralized Differentiable Memory-Enabled Mapping and Navigation for Multiple Robots

TL;DR

D2M2N introduces a decentralized, memory-enabled neural architecture for multi-robot navigation that leverages past observations for improved path planning in complex environments.

Contribution

The paper presents D2M2N, a novel fully differentiable, decentralized architecture that incorporates memory and value iteration for enhanced multi-robot mapping and navigation.

Findings

D2M2N outperforms existing models in complex mapping tasks.

The architecture effectively leverages past observations for better navigation.

Experimental results demonstrate significant improvements over state-of-the-art methods.

Abstract

Recently, a number of learning-based models have been proposed for multi-robot navigation. However, these models lack memory and only rely on the current observations of the robot to plan their actions. They are unable to leverage past observations to plan better paths, especially in complex environments. In this work, we propose a fully differentiable and decentralized memory-enabled architecture for multi-robot navigation and mapping called D2M2N. D2M2N maintains a compact representation of the environment to remember past observations and uses Value Iteration Network for complex navigation. We conduct extensive experiments to show that D2M2N significantly outperforms the state-of-the-art model in complex mapping and navigation task.