Immersive Virtual Reality-Integrated Multi-Agent Deep Q-Learning Framework for Scalable, Adaptive, and Collision of Autonomous Uav Swarms in Highly Dynamic Environments

View at publisher Summarise with VIRAT
Authors: Akash Sharma P, Arun M V, Jeffin Gracewell J
Cited by: 0
References: 20

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The fast growth of autonomous unmanned aerial vehicle swarms underlines critical challenges in gaining efficient coordination and collision-free navigation within complex, dynamic environments. Most of the traditional control strategies are not scalable and cannot make real-time decisions, especially in dense or uncertain airspaces. In this context, a VRassisted multi-agent deep Q-learning framework is proposed for UAVs to learn cooperative behaviors through immersive virtual simulations. The system employs virtual reality to provide rich environmental feedback, which enhances state representation and decision-making accuracy for every participating agent. The idea is to achieve an optimized swarm-level coordination, while ensuring trajectories that are safe and free from collisions, even in highly dynamic scenarios. The contribution of this work differs from state-of-the-art methods by incorporating multiagent reinforcement learning with VR-enabled environmental perception to allow UAVs to dynamically adapt to evolving conditions. This work proposes a new approach for combining VR-based environmental immersion with multi-agent deep Qlearning to arrive at a scalable, adaptive, safety-oriented navigation strategy for autonomous UAV swarms that outperforms conventional approaches in complex scenarios.

Fields of Study & Subjects

Reinforcement Learning in Robotics Distributed Control Multi-Agent Systems Robotic Path Planning Algorithms

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
1207–1213
Publication date
21 January 2026
Language
EN