Cloud Control of Connected Vehicle Under Stochastic Bi-Directional Delay: An Asynchronized Predictor-Observer Method

In this paper, to compensate time varying bi-directional delay in the connected vehicle cloud control system, an asynchronized predictor-observer structured control strategy is studied, which novelly predict the system states to compensate for the bi-directional delay based on only the boundary information of the delay. The stability of the proposed control strategy under a certain delay length and model uncertainty is analyzed with Lyapunov-Krasovskii Functional (LKF) method, and the stability condition is given in the form of matrix inequality. With the help of Cone Complementary Linearization (CCL), an algorithm to find stabilized controller and observer gains are also optionally provided. Ultimately, simulations encompassing both lateral and longitudinal control of connected vehicles, incorporating high-fidelity vehicle dynamics models, are executed. The simulation outcomes substantiate the feasibility and efficacy of the proposed strategy for delay compensation. Additionally, vehicle tests conducted within a lateral control scenario further affirm the engineering feasibility of the proposed method.