Genetic Algorithm-Optimized LQR for Enhanced Stability in Self-Balancing Wheelchair Systems

Balancing systems, exemplified by electric wheelchairs, require accurate and effective functioning to maintain equilibrium across many situations. This research looks at how well a standard linear quadratic regulator (LQR) and its genetic algorithm (GA)-optimized version keep an electric wheelchair stable when it stands on its own. The aim of the optimization was to improve energy economy, robustness, and responsiveness through the refinement of control settings. Simulations were performed under two scenarios: stabilizing the system from a tilt and recovering from an external force. Both controllers stabilized the system; however, the GA-optimized LQR demonstrated considerable improvements in control efficiency, decreased stabilization time, and enhanced response fluidity. It exhibited improved resilience to external disturbances, as indicated by a decrease in oscillations and an increase in fluid displacement recovery. These enhancements highlight the LQR's versatility, resilience, and appropriateness for real-world applications, including Segways, balancing robots, and patient wheelchairs. This study highlights the ability of evolutionary algorithms to enhance the effectiveness of traditional control systems in dynamic and unpredictable settings.