FUZZY-LOGIC

This research presents the design and implementation of an integrated fuzzy logic–based decision-making system for autonomous vehicle navigation, focusing on intelligent speed regulation, steering control, and lane keeping. A three-degree-of-freedom (3-DoF) dual-track vehicle dynamics model was developed in MATLAB/Simulink to capture longitudinal, lateral, and yaw behaviors. The control architecture uses a Takagi–Sugeno fuzzy inference system to process speed error, distance error, and yaw deviation, generating throttle and steering actions that emulate human driving intuition. A simulation-based framework was developed for both ego and target vehicles, enabling the evaluation of inter-vehicle distance, trajectory following, and lane stability across straight and curved road sections. Results show that the fuzzy controller reduced longitudinal speed error to below 0.25 m/s, maintained lateral deviation within ±0.12 m on curved paths, and improved yaw rate tracking with a settling time of 1.8 s compared to 3.1 s without fuzzy control. The controller also limited throttle oscillations to less than 5% and sustained a safe inter-vehicle distance with less than 7% deviation from the desired headway. Overall, the research establishes a computationally efficient fuzzy-logic framework suitable for autonomous-vehicle applications, and the findings confirm the controller's robustness and adaptability in a virtual test environment.