Implementing distributed collaboration and applying the YOLO algorithm to robots

Abstract

Recently, the world has witnessed the development of artificial intelligence and robot programming, which have enabled robots to work together to perform specific assigned tasks while overcoming obstacles in the environment. Robots can now operate independently of each other. In this research, four-wheeled robots were created in webots and placed in different environments in Gazebo. These robots are associated with LIDAR and Kinect cameras. Due to the distributed collaboration between robots, a robot cannot traverse a path that it has previously traversed. The three robots are given the same goal, and the first robot to reach the goal signals the end of the mission to the remaining robots and learns about the surrounding objects on the way to the goal. Therefore, the YOLO algorithm was used (You Only Look Once). This is one of the best algorithms for detecting objects in their environment. Regarding the results obtained in the simulation, the robot performed all the assigned tasks.