Abstract

Background: In this paper, a Proportional-Integral-Differential (PID) controller that facilitates track maneuvering for self-driving cars is proposed.

Methods: Three different design approaches are used to find and tune the controller hyperparameters. One of them is “WAF-Tune”, which is an ad hoc trial-and-error based technique that is specifically proposed in this paper for this specific application. The proposed controller uses only the Cross-Track-Error (CTE) as an input to the controller, whereas the output is the steering command.

Results: Extensive simulation studies in complex tracks with many sharp turns have been carried out to evaluate the performance of the proposed controller at different speeds.

Conclusion: The analysis shows that the proposed technique outperforms the other ones. The usefulness and the shortcomings of the proposed tuning mechanism are also discussed in details.

Keywords: PID control, self-driving car, autonomous driving, PID tuning, collision avoidance, fuzzy control.