Adaptive backstepping control of tracked robot running trajectory based on real-time slip parameter estimation

To ensure the stable driving of tracked robots in a complex farmland environment, an adaptive backstepping control method for tracked robots was proposed based on real-time slip parameter estimation. According to the kinematics analysis method, the kinematic model of the tracked robot was established, and then, its pose error differential equation was further obtained. On this basis, the trajectory tracking controller of the tracked robot was designed based on the backstepping control theory. Subsequently, according to the trajectory tracking error of the tracked robot, the back propagation neural network (BPNN) was used to adaptively adjust the control parameters in the backstepping controller, and the inputs of the BPNN are the trajectory tracking error xe, ye, θe. After that, the soft-switching sliding mode observer (SSMO) was designed to identify the slip parameters during the running of the tracked robot. And then the parameters were compensated into the adaptive backstepping controller to reduce the trajectory tracking error. The simulation results show that the proposed adaptive backstepping control method with SSMO can improve the accuracy of the trajectory tracking control of the tracked robot. Additionally, the designed SSMO can accurately estimate the slip parameters. Keywords: tracked robot, trajectory control, adaptive backstepping control, neural networks, slip parameter, sliding mode observer DOI: 10.25165/j.ijabe.20201304.5739 Citation: Lu E, Ma Z, Li Y M, Xu L Z, Tang Z. Adaptive backstepping control of tracked robot running trajectory based on real-time slip parameter estimation. Int J Agric & Biol Eng, 2020; 13(4): 178–187.