Visual Swarm Navigation via Deep Reinforcement Learning and Evolutionary Hybrid Design

原文：Visual Swarm Navigation via Deep Reinforcement Learning and Evolutionary Hybrid Design

作者：Álvaro Díez, Fidel Aznar

来源：arXiv cs.RO（机器人）

正文

Computer Science > Robotics

arXiv:2610.06400v1 (cs)

[Submitted on 5 Oct 2026]

Title:Visual Swarm Navigation via Deep Reinforcement Learning and Evolutionary Hybrid Design

Authors:Álvaro Díez (Department of Computer Science and Artificial Intelligence, University of Alicante), Fidel Aznar (Department of Computer Science and Artificial Intelligence, University of Alicante)

View a PDF of the paper titled Visual Swarm Navigation via Deep Reinforcement Learning and Evolutionary Hybrid Design, by '{A}lvaro D'{i}ez (Department of Computer Science and Artificial Intelligence and 3 other authors

View PDF

HTML (experimental)

Abstract:Swarm robotics presents a robust and cost-effective paradigm for advanced automation in complex, dynamic environments, such as those encountered in search and rescue or environmental monitoring. A fundamental challenge for this field is the data-driven design of decentralized controllers capable of generating emergent collective behaviors. This paper proposes a novel, AI-driven hybrid methodology for the automatic synthesis of swarm robotic controllers for autonomous visual navigation. This approach synergistically combines multi-agent reinforcement learning with neuro-evolutionary strategies, specifically leveraging implementations of the cross-entropy method and the covariance matrix adaptation evolution strategy to optimize a pre-trained individual navigation policy. The underlying deep architecture is engineered for low-cost, resource-constrained platforms, utilizing a compact neural network that relies exclusively on monocular camera imagery. This vision-based design emphasizes computational and energy efficiency, a critical requirement for practical swarm deployments. Experiments, performed in a high-fidelity physics simulator, demonstrate that the resulting controllers enable robust and scalable collective exploration of diverse indoor environments. The controller trained using our cross-entropy method achieves superior exploration coverage, visiting 36.20% more regions compared to the covariance matrix adaptation evolution strategy. Critically, our best vision-based policy achieves exploration performance statistically comparable to traditional methods relying on more expensive distance sensors, while delivering a significant 31.40% average reduction in energy consumption. These findings validate an effective and economically viable autonomous control system, establishing a path for deploying highly efficient collective intelligence in real-world engineering applications.

Comments:

46 pages, 21 figures. Published in Engineering Applications of Artificial Intelligence under a CC BY 4.0 license

Subjects:

Robotics (cs.RO); Multiagent Systems (cs.MA); Neural and Evolutionary Computing (cs.NE)

ACM classes:

I.2.9; I.2.11; I.2.6; I.2.8

Cite as:

arXiv:2610.06400 [cs.RO]

(or

arXiv:2610.06400v1 [cs.RO] for this version)

https://doi.org/10.48550/arXiv.2610.06400

Focus to learn more

arXiv-issued DOI via DataCite (pending registration)

主题

机器人 · 多智能体 · 神经计算

由「前沿雷达」于 2026-10-05 采集。正文取自原文页面，已保留出处链接。标题与正文版权归原作者所有。