Dynamic Cooperative Vehicle Platoon Control Considering Longitudinal and Lane-changing Dynamics

TL;DR

This paper introduces a distributed control algorithm and a real-time lane-changing model for connected autonomous vehicle platoons, enhancing safety, stability, and adaptability during lane changes under various conditions.

Contribution

It develops a novel DCPID control algorithm with stability analysis and a dynamic lane-changing model integrating MPC, improving safety and robustness in heterogeneous CAV platoons.

Findings

DCPID algorithm achieves robust spacing and speed control in 400 scenarios.

The lane-changing model ensures safe, effective maneuvers even in emergencies.

Mathematical stability conditions are derived for the control algorithm.

Abstract

This paper presents a distributed cascade Proportional Integral Derivate (DCPID) control algorithm for the connected and automated vehicle (CAV) platoon considering the heterogeneity of CAVs in terms of the inertial lag. Furthermore, a real-time dynamic cooperative lane-changing model for CAVs, which can seamlessly combine the DCPID algorithm and the improved sine function is developed. The DCPID algorithm determines the appropriate longitudinal acceleration and speed of the lane-changing vehicle considering the speed fluctuations of the front vehicle on the target lane (TFV). In the meantime, the sine function plans a reference trajectory which is further updated in real time using the model predictive control (MPC) to avoid potential collisions until lane-changing is completed. Both the local and the asymptotic stability conditions of the DCPID algorithm are mathematically derived, and the sensitivity of the DCPID control parameters under different states is analyzed. Simulation experiments are conducted to assess the performance of the proposed model and the results indicate that the DCPID algorithm can provide robust control for tracking and adjusting the desired spacing and velocity for all 400 scenarios, even in the relatively extreme initial state. Besides, the proposed dynamic cooperative lane-changing model can guarantee an effective and safe lane-changing with different speeds and even in emergency situations (such as the sudden deceleration of the TFV).