Cloud Control of Connected Vehicle Under Delayed Sensing Information: A Realization of Predictor–Observer Structured Controller

Abstract

This article addresses the problem of controlling connected vehicles under delayed sensing information with the prediction-based control method. First, the connected vehicle cloud control system under delayed sensing is modeled as a discrete linear output feedback system with output delay. Second, synchronized and asynchronized predictor–observer structured controllers are adopted as the delay compensation strategy for known and unknown delay cases. The stability analysis of both prediction strategies under specific delay boundary and bounded model uncertainty is conducted with the help of the scaled bounded real lemma (SBRL) and Lyapunov–Krasovskii functional (LKF) methods, respectively. By utilizing the cone complementary linearization (CCL) technique, an algorithm is developed to seek the stabilized controller and observer parameters iteratively. Finally, the proposed methods are verified through connected vehicle lateral and longitudinal control scenarios built by Simulink–CarSim, where the capability and performance of the proposed delay compensation control strategy are assessed under a relatively actual dynamic.