Adaptive sliding mode control for two-wheeled self-balancing vehicle with input delay

An adaptive sliding mode control algorithm was designed for two-wheeled self-balancing vehicle with input delay. The Lagrange equation was used to establish the dynamic mathematical model of two-wheeled self-balancing vehicle system. In the system model, the input delay in practice environment and the unknown disturbance in dealing with the input delay were considered. After the singular value decomposition of transformed input matrix, an adaptive sliding mode controller with adaptive estimation ability for the disturbance parameters was designed. Based on the Lyapunov stability theory, the robust asymptotic stability of closed-loop system was guaranteed. In the experiment, the gyroscope MPU-6050 and acceleration sensor were used to construct the vehicle attitude detection device.Analysis result shows that when the control parameters are small, the overshoot of the system is small, while the regulation time of the system is long. When the control parameters are large, the system has a more obvious overshoot, while the regulation time of the system is shortened. The velocity range is less than 0.08 m·s^-1 and the angular velocity range is less than 0.6°·s^-1 when the vehicle body is subjected to a small disturbance. The velocity range is less than 0.1 m·s^-1 and the angular velocity range is less than 0.8°·s^-1 when the vehicle body is subjected to a large disturbance. From the initial inclination of 5°, the velocity of vehicle is within 0.005 m·s^-1 and the angular velocity of vehicle is within 0.022°·s^-1. While from the initial inclination of 10°, the velocity of vehicle is within 0.007 m·s^-1 and the angular velocity of vehicle is within 0.031°·s^-1. So the adaptive sliding mode control algorithm can make the vehicle adjust itself and quickly return to a stable state under an appropriate interference and different initial vehicle inclinations.