Control Techniques Review for the Ball-on-Sphere (BOS) System

There are many important balancing systems in control theory. The Ball-on-Sphere (BOS) system is one of these systems. It finds applications in aerospace, transportation, robotics, etc. The system is a nonlinear multiple-input multiple-output (MIMO) system where the ball is controlled on the sphere. It is a testing ground, due to its naturally nonlinear, under-actuation, and unstable nature, in nonlinear systems control. Consequently, different techniques have been used to control the system. A review of these techniques of comprehensive nature is presented in this paper. Different reputable publishers were considered using different search engines utilizing different search terms. The techniques include Linear Quadratic Gaussian (LQG), Sliding Mode Control (SMC), Fuzzy Logic Control (FLC), Artificial Neural Network (ANN), Feedback Linearization (FL), Linear Quadratic Regulator (LQR) and their hybrids. It was established that the system is not covered much in the literature. The LQR, LQG, and SMC were found to be the most used controllers to control the system. Also, 93% of the techniques used no optimization. Additionally, it was established that more techniques need to be explored for the system control. Control techniques hybridization and exploration and optimization usage were found to be the major future directions.