Attitude Adjustment of a Two-Wheel Balance Vehicle Based on an Enhanced Model-Free Adaptive Control Algorithm

In this work, an enhanced model-free adaptive control with tracking differentiator (TD-EMFAC) method is proposed for the attitude control of a two-wheel balance vehicle. First, a dynamic linearization technique based on a pseudo partial derivative (PPD) with time-varying parameter factors is introduced to dynamically linearize the dynamic process of the two-wheel balance vehicle. Second, a model free adaptive controller is designed. At the same time, the controller considers the time delay of the system in the design process. By proposing an adaptive factor coefficient and combining it with the Smith prediction method, the designed controller can manage systems with time delays. The advantage of the proposed method is that the entire controller design process is based on the input and output data of the controlled object without requiring an accurate mathematical model of the controlled object. Third, a series of experimental results based on an actual two-wheel balance vehicle verify the effectiveness of the proposed method. A strict mathematical proof verifies the stability of the proposed method.