Monocular Vision Measurement for Drones Obstacle Avoidance and Autonomous Path Planning With Circle RRT*

Abstract

Drones have garnered increasing attention for applications, such as disaster response, surveillance, and smart city monitoring, necessitating effective collision avoidance systems (CASs) to ensure safe navigation in complex environments. Traditional CAS often relies on multiple sensors, which bring cost, weight, and computational concerns, posing challenges for lightweight civilian drones with limited payload capacity. This article proposes a novel, cost-effective obstacle avoidance system that utilizes only a monocular camera, integrating traditional image processing techniques, background subtraction, and a deep learning (DL) method for robust obstacle sensing. Enhanced by a depth predictor for distance estimation, the system is paired with a new path planning algorithm, Circle Rapidly Exploring Random Tree (RRT*), which improves upon the RRT* algorithm by focusing computational efforts on targeted sectors, thereby enhancing efficiency and enabling real-time navigation. The experimental evaluations in both 2-D and 3-D environments demonstrate the system’s high accuracy in obstacle sensing and its ability to generate optimized, collision-free paths with reduced computational overhead. This lightweight, single-camera solution offers a practical and efficient approach for autonomous drones operations, addressing the limitations of multisensor systems while maintaining reliability and performance in dynamic environments.