Adaptive Optimal Control of Heterogeneous CACC System With Uncertain Dynamics

Abstract

Cooperative adaptive cruise control (CACC), as an extension of adaptive cruise control, connects multiple vehicles in a platoon via wireless communication. In practice, different vehicles may have different dynamic parameters and their exact values are unknown/uncertain to designers. In this brief, we propose a new control structure that uses an estimate of dynamic parameters to transform the heterogeneous CACC problem into the regulation problem of error dynamics for each vehicle in the platoon. An adaptive optimal control is proposed to learn the optimal feedback based on online data. The position transfer function between adjacent vehicles is further analyzed in the frequency domain. By sum of squares programming, the minimum headway values that ensure the vehicle string stability are found. Experiments on numerical and complex systems validate our method.