RaCIL: Ray Tracing based Multi-UAV Obstacle Avoidance through Composite Imitation Learning

TL;DR

This paper introduces RaCIL, a novel obstacle avoidance method for UAVs that combines ray-tracing with composite imitation learning techniques like PPO, BC, and GAIL, demonstrating improved collision avoidance in multi-UAV scenarios.

Contribution

The paper presents a new integrated framework using ray-tracing and composite imitation learning for scalable and reliable UAV obstacle avoidance, especially in multi-UAV environments.

Findings

Enhanced obstacle detection with ray-tracing

Improved collision avoidance in multi-UAV scenarios

Scalability to four UAVs demonstrated

Abstract

In this study, we address the challenge of obstacle avoidance for Unmanned Aerial Vehicles (UAVs) through an innovative composite imitation learning approach that combines Proximal Policy Optimization (PPO) with Behavior Cloning (BC) and Generative Adversarial Imitation Learning (GAIL), enriched by the integration of ray-tracing techniques. Our research underscores the significant role of ray-tracing in enhancing obstacle detection and avoidance capabilities. Moreover, we demonstrate the effectiveness of incorporating GAIL in coordinating the flight paths of two UAVs, showcasing improved collision avoidance capabilities. Extending our methodology, we apply our combined PPO, BC, GAIL, and ray-tracing framework to scenarios involving four UAVs, illustrating its scalability and adaptability to more complex scenarios. The findings indicate that our approach not only improves the reliability of basic PPO based obstacle avoidance but also paves the way for advanced autonomous UAV operations in crowded or dynamic environments.