D2CoPlan: A Differentiable Decentralized Planner for Multi-Robot Coverage

TL;DR

D2CoPlan is a scalable, learning-based decentralized coverage planner for multi-robot systems that outperforms classical algorithms in efficiency and can be integrated with other learning methods for improved performance.

Contribution

The paper introduces D2CoPlan, a differentiable, distributed coverage planning algorithm that scales efficiently and can be combined with other learning approaches for enhanced multi-robot coverage.

Findings

D2CoPlan scales better than expert algorithms in runtime and number of agents.

D2CoPlan performs on par with classical distributed algorithms.

Combining D2CoPlan with other learning methods yields superior solutions.

Abstract

Centralized approaches for multi-robot coverage planning problems suffer from the lack of scalability. Learning-based distributed algorithms provide a scalable avenue in addition to bringing data-oriented feature generation capabilities to the table, allowing integration with other learning-based approaches. To this end, we present a learning-based, differentiable distributed coverage planner (D2COPL A N) which scales efficiently in runtime and number of agents compared to the expert algorithm, and performs on par with the classical distributed algorithm. In addition, we show that D2COPlan can be seamlessly combined with other learning methods to learn end-to-end, resulting in a better solution than the individually trained modules, opening doors to further research for tasks that remain elusive with classical methods.