Event-Triggered Model Predictive Control for an Autonomous Vehicle Based on Tight Constraints

• Published in: IEEE International Conference on Industrial Informatics (INDIN) pp. 1 - 6
• Main Authors: Lin, Xiurui, Chen, Hantuo, Qin, Dongdong, Liu, Andong, Ni, Hongjie, Wang, Ye
• Format: Conference Proceeding
• Language: English
• Published: IEEE 18.08.2024
• Subjects: autonomous vehicle; Autonomous vehicles; Closed loop systems; constraint tightening; Event detection; event-triggered strategy; model predictive control; Optimization; Prediction algorithms; Predictive control; Simulation; Stability analysis; Sufficient conditions; Tires; trajectory tracking
• DOI: 10.1109/INDIN58382.2024.10774245

This paper develops an event-triggered model pre-dictive control strategy based on tight constraints for the control of autonomous vehicle. First, a discrete-time error model is established for a vehicle according to the magic formula-based tire model and the simplified bicycle model. Then, constraints with continuous contraction are constructed based on the tight constraint control mechanism to avoid excessive proximity to initial constraints and improve the vehicle comfort. Besides, an event-triggering condition is derived based on the input state stability (ISS) condition. Finally, the feasibility of the algorithm is proved, and the stability of the closed-loop system is derived. Simulation results show that this strategy can effectively reduce the waste of resources under the premise of ensuring a good vehicle tracking effect.