2021/06/23 by Hassan Jafarzadeh, Jafarzadeh, Hassan, Cody Fleming +1 · 2 citations
Engineering · #Advanced Control Systems Optimization #FOS: Electrical engineering #Real-time simulation and control systems #Systems and Control (eess.SY) #Vehicle Dynamics and Control Systems #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2106.12366
openalex publication_date 2021/06/23 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
In this paper, we present a data-driven Model Predictive Controller that\nleverages a Gaussian Process to generate optimal motion policies for connected\nautonomous vehicles in regions with uncertainty in the wireless channel. The\ncommunication channel between the vehicles of a platoon can be easily\ninfluenced by numerous factors, e.g. the surrounding environment, and the\nrelative states of the connected vehicles, etc. In addition, the trajectories\nof the vehicles depend significantly on the motion policies of the preceding\nvehicle shared via the wireless channel and any delay can impact the safety and\noptimality of its performance. In the presented algorithm, Gaussian Process\nlearns the wireless channel model and is involved in the Model Predictive\nController to generate a control sequence that not only minimizes the\nconventional motion costs, but also minimizes the estimated delay of the\nwireless channel in the future. This results in a farsighted controller that\nmaximizes the amount of transferred information beyond the controller's time\nhorizon, which in turn guarantees the safety and optimality of the generated\ntrajectories in the future. To decrease computational cost, the algorithm finds\nthe reachable set from the current state and focuses on that region to minimize\nthe size of the kernel matrix and related calculations. In addition, we present\nan efficient recursive approach to decrease the time complexity of developing\nthe data-driven model and involving it in Model Predictive Control. We\ndemonstrate the capability of the presented algorithm in a simulated scenario.\n