This paper proposes an optimal controller design to enhance the upright balancing and driving performance of a two-wheeled robot. The control system for the two-wheeled robot is based on the Linear Quadratic Regulator (LQR) technique. To determine the optimal control gain by tuning the weighting matrices, the Bacterial Foraging Optimization Algorithm (BFOA) and the hybrid BFOA-PSO (Particle Swarm Optimization) algorithm were implemented. Control performance was quantitatively evaluated using a fitness function that integrates the Integral of Time-weighted Absolute Error (ITAE) index with control input weighting. To verify the practical effectiveness of the proposed controller, a 3D model was established by interfacing SolidWorks CAD data with the Simscape Multibody environment. The results demonstrate that the BFOA-based optimization technique is effective in deriving robust control parameters for the robot. While the hybrid algorithm exhibited slightly superior performance under specific operating conditions, it was observed that the relative superiority among the algorithms varies depending on the initial parameter configurations and environmental factors.
Dong-Sang Yoo (Thu,) studied this question.