Comparative Performance of Fuzzy Logic and PID Steering Control for Improved Swerve Autonomous Vehicles

This study presents a performance comparison of a fuzzy logic controller with a proportional-integral-derivative (PID) controller in an autonomous vehicle steering controller based on an improved swerve drive. The advantage of this swerve drive system is that it provides high maneuverability in tight spaces by utilizing nonlinear kinematic behavior and strong coupling between translational and rotational motions. This is a challenge for conventional control strategies. To overcome this problem, a fuzzy logic controller is used, which has the ability to work in more dynamic conditions. To support the control system for precision, a good structural design is required. The feasibility of the proposed improved swerve drive mechanical design is verified through finite element-based structural analysis to ensure that the control performance is not limited by mechanical constraints. Testing results show that the configuration of seven membership functions in the fuzzy logic controller provides the best performance, with an overshoot value of 7.33% and a steady-state error of 0.0324. Real-time testing of this electric vehicle prototype was conducted in five scenarios: straight road, 90° turn, parallel parking, obstacle avoidance, and on-the-spot maneuvering. The testing results also show that the fuzzy logic controller consistently outperforms the PID controller by reducing tracking error, minimizing overshoot, and achieving faster settling times, especially under complex motion conditions. Structural validation also confirms that this improved swerve drive, operating within the elastic limits of the material, supports the implementation of reliable control strategies.