Real‐time control design and implementation of ball balancer system based on machine learning and machine vision

Summary This article addresses a hybrid pseudo‐PD/machine learning controller algorithm to improve the stabilization and trajectory tracking of the ball on the plate system (BOPS). The proposed controller design depends on a machine learning algorithm that detect the angle of the servo motor required to correct the position of the ball on the plate also the parameters of the PD controller is changed online using the fuzzy logic to enhance the performance of the trajectory and point tracking of the system. This article offers three different machine learning techniques for predicting servo motor angle that obtain higher accuracies of 99.95%, 99.908%, and 99.998% for support vector regression, decision tree regression, and random forest regression, respectively. The proposed scheme has greatly improved the system's settling time and overshoot, according to simulation and practical results. The Lagrangian formulation may be used to obtain the mathematical formulation, and a practical identification experiment can be used to determine the servo motor parameter. Simulation and practical results on a BOPS dynamic model are both demonstrated to justify the validation of the proposed design scheme. The experimental practical results justify the feasibility and effectiveness of the proposed controller design strategy.