Research and Development of Omnidirectional Mobile Robot Tracking Control Based on Artificial Intelligence

This paper presents the research and construction of a motion tracing control system for omnidirectional mobile robots based on reinforcement learning techniques in automatic control. The process controlling robot flat environment with definite unknown obstacles, taking into account nonlinear factor interference. Research application programming tools are operating systems (Robot Operating System - ROS). From updated information maps, environment, position, obstacle identification (SLAM) to calculate movement trajectory three-wheeled robot. positioning calculates orbital tracking Q-learning algorithm. results simulation Gazebo running tests real Turtlebot have shown practical effectiveness problem intelligent navigation robots.