Object tracking and following six-legged robot system using Kinect camera based on Kalman filter and backstepping controller

Abstract

This paper proposes an object tracking and following six-legged robot (6LR) system that uses a Kinect camera based on Kalman filter and backstepping control method. To achieve this task, the following steps are executed. First, the 6LR is developed with several interconnected devices, such as servomotors, a microcontroller, Bluetooth, and so on. The Kinect camera is installed on the 6LR to perform image processing. Second, the kinematic modeling of the 6LR is presented. Third, a blue-colored candidate object is detected by the Kinect camera through a color-based object detection method, and the position coordinate of the detected object inside the RGB image frames is obtained. The real position coordinate of the detected object (in mm) is obtained by using simple trigonometry and Kinect depth data. Fourth, Kalman filter algorithm is used to estimate the real position coordinate and velocity coordinate of the moving candidate object. Fifth, backstepping method using Lyapunov function is adopted to design a controller for the 6LR to perform the object-following task. Finally, the experimental results are presented to verify the effectiveness and performance of the proposed control method. The results show that the 6LR can successfully follow the moving candidate object with the designed controller.