Predictive Cruise Cloud Control Scheme Design on Notable Vehicles—Under the Perspective of Cyber-Physical Systems

The notable vehicles (NV) predictive cruise cloud control (PCCC) system has great potential in improving the performances of driving safety and energy saving by virtue of advantages on the decent calculation and optimal control capability of cloud control system (CCS). The traditional PCC systems lack the interoperable integration control of cyber layer and physical plane, limited by the range of perception and on-board computing ability, so that the information acquisition and processing are quite restrained, which impede the improvement of control performances to a certain extent. In this study, we propose a CCS layered architecture based on cyber-physical systems (CPS), which effectively addresses the problems of real-time multi-sources heterogeneous information utilization by PCC systems and insufficient computational planning capability of solver of NV. Then, based on the digital information in the cloud control basic platform, we propose an intensified PCCC framework in the hierarchical architecture application platform. In this framework, we introduce a modified deep-learning traffic prediction model to predict the traffic state after pre-factorization processing and the optimization algorithm solver to optimize the proposed scheme under the restrictive constraints via using the prediction information and digital information. We fully evaluate the effectiveness of the architecture under various traffic scenarios through simulation. The results show that the proposed layered architecture has implementation feasibility and great potential in the application of vehicle-cloud layered control.