Optimizing Path Planning and Human Detection Using YOLO and an RRT Algorithm in Autonomous Robots

This research paper explores the integration of advanced technologies to improve the capabilities of autonomous robots. Specifically, the optimization of path planning and human detection using the RRT (Rapidly Exploring Random Tree) algorithm and the YOLO (You Only Look Once) object detection model. By combining these powerful tools, robots can navigate controlled environments and interact with humans. The RRT algorithm generates collision-free paths for efficient robot navigation. This algorithm efficiently explores the search space and identifies optimal trajectories to achieve specific goals. YOLO, a state-of-the-art object detection model, is utilized to accurately detect and localize humans within the robot's field of view. This information is then fed into the RRT algorithm to dynamically adjust the robot's path and prioritize human interaction. To implement this system, the computational power of a Jetson Nano and a Raspberry Pi 4 is combined. The Jetson Nano handles the RRT path planning algorithm, while the Raspberry Pi 4 processes the camera feed and runs the YOLO model for human detection. Effective communication between the two devices is crucial for seamless integration and real-time response. Through rigorous testing, this research paper demonstrates the successful simulation of this integrated system. The robot is able to accurately detect humans, plan efficient paths, and simulate the navigation towards them while avoiding obstacles. This research contributes to the advancement of autonomous UGV (Unmanned Ground Vehicles) and has the potential to revolutionize various industries, including manufacturing, logistics, and healthcare.