This paper presents a novel approach to address the lateral control issue in trajectory tracking for autonomous cars. Traditional model-free adaptive control algorithms have some limitations, prompting the development of an enhanced data-driven model-free adaptive control scheme. More precisely, inside a specified trajectory, the vehicle's issue of lateral control is converted into a problem of tracking the preview-deviation-angle. A dynamic linearized data model is developed for the lateral control of autonomous cars, taking into account the input and output data. To enhance the steering angle control rate, a time-varying proportional control term is incorporated. An enhanced trajectory tracking control technique for the MFAC system is proposed, followed by a demonstration of its convergence. The simulation findings demonstrate that the IMFAC scheme exhibits a superior output response compared to both MFAC and PID in both low-speed and highspeed scenarios. Furthermore, the IMFAC system is capable of achieving autonomous vehicle path tracking.
Trajectory tracking control of autonomous vehicles based on improved model-free adaptive control
Third International Conference on Testing Technology and Automation Engineering (TTAE 2023) ; 2023 ; Xi-an, China
Proc. SPIE ; 13079
2024-03-15
Conference paper
Electronic Resource
English
LSTM-Based Trajectory Tracking Control for Autonomous Vehicles
SAE Technical Papers | 2022
|LSTM-Based Trajectory Tracking Control for Autonomous Vehicles
British Library Conference Proceedings | 2022
|Trajectory tracking control of autonomous vehicles based on event‐triggered model predictive control
DOAJ | 2024
|