Hybrid Fuzzy-PID Controller for Enhanced Trajectory Accuracy in Differential Drive Mobile Robots

Mobile robot navigation requires precise base controllers to achieve accurate trajectory tracking. Differential drives commonly suffer from odometry errors caused by wheel slippage, encoder noise, and motor control inaccuracies, leading to cumulative positioning errors. This paper presents a hybrid Fuzzy-PID controller that dynamically optimizes PID gains based on trajectory error and error rate to improve tracking accuracy. The research contribution is the development of an adaptive gain-scheduling mechanism using fuzzy inference to automatically tune PID parameters (Kp, Ki, Kd) in real-time, reducing the need for manual calibration. The fuzzy logic controller employsa7×7Mamdani-typerulebasewithtriangularmembershipfunctions, taking position error and error derivative as inputs and outputting optimized gain adjustments. The system was validated through both simulation (MATLAB/Simulink) and hardware experiments using a custom differential drive robot with incremental encoders and DC motors. Experimental results demonstrate that the hybrid Fuzzy-PID controller reduces average trajectory error by 43.7% compared to conventional PID (from 12.8 cm to 7.2 cm in circular path tracking), improves settling time by 31%, and achieves 89.4%reduction in steady-state error. The controller maintains stability across varying speeds (0.1-0.5 m/s) and different trajectory shapes (straight, circular). The proposed hybrid approach significantly enhances trajectory tracking accuracy while maintaining computational efficiency suitable for real-time embedded implementation.