Kinect camera sensor-based object tracking and following of four wheel independent steering automatic guided vehicle using Kalman filter

摘要

This paper presents a Kinect camera sensor-based object tracking and following system for the four wheel independent steering automatic guided vehicle (4WIS-AGV) using Kalman filter and backstepping control method. To accomplish this task, the following steps are executed: Firstly, a blue colored object is detected by the Kinect camera sensor using a color-based object detection method and its center coordinates inside the RGB image frames are obtained. The global position coordinates of the detected object and the 4WIS-AGV are obtained using Kinect depth data and NAV200 navigation system. Secondly, a Kalman filter is used to estimate the global position and velocity coordinates of the detected moving object. Thirdly, a backstepping control method using Lyapunov function is adopted for the 4WIS-AGV to perform the object following task. Finally, simulation and experiment are performed to verify the effectiveness of the proposed system. The results show that the 4WIS-AGV with Kinect camera sensor can follow the moving object well with the designed controller.