Fractional-Order PI-Like Trajectory Tracking Control of Four-Wheel Skid Steering Small Autonomous Ground Vehicles.

Area of Science:

• Robotics
• Control Systems Engineering
• Autonomous Vehicles

Background:

• Autonomous ground vehicles (AGVs) face challenges in trajectory tracking due to uncertainties like friction and external disturbances.
• Existing AGV tracking controllers often lack explicit performance guarantees or are too complex for onboard implementation.

Purpose of the Study:

• To develop a novel controller for four-wheel skid-steering AGVs (FWSAGVs) that ensures explicit tracking error constraints.
• To enhance controller simplicity for practical onboard application in FWSAGVs.

Main Methods:

• Proposed a prescribed-performance fractional-order PI-like controller (PP-FOPIC) integrating prescribed performance control (PPC) with a fractional-order PI-like structure.
• Utilized particle swarm optimization (PSO) for automated tuning of controller parameters.
• Validated the controller through comparative simulations and real-world experiments.

Main Results:

• The PP-FOPIC demonstrated superior tracking accuracy and robustness against uncertainties compared to PP-IOPIC and PID controllers.
• Achieved significantly lower actuator overload rates.
• Real-world hexagonal trajectory experiments confirmed practical applicability with an average onboard execution time of 8.62 ms.

Conclusions:

• The PP-FOPIC offers a low-complexity, effective solution for FWSAGV trajectory tracking with guaranteed performance.
• The method provides quantitative performance guarantees and practical advantages for autonomous vehicle control.