Cooperative decision making and control
Aiming at the three characteristics of the system of vehicle-vehicle cooperation, which are borderless, dispersion of individual behavioral goals and concentration of group behavioral goals, taking into account optimization for both individual and system, and addressing the difficulties of high-frequency small perturbations in traffic flow and maintaining stability, an invisible constraint predictive control method based on road space domain is proposed, and the linear stability and system stability of the method are proved theoretically.
The method has been applied in some practical projects.
1.The development of V2X based CACC for passenger cars (Cooperator: SAIC)
In order to cope with the vehicle platoon tasks under different scenarios and functional requirements, the V2X based CACC system was constructed based on a rolling horizon optimal control method, which balanced the accuracy, comfort, energy consumption and the reliability of the controller. After the verification of both simulation and field tests of various scenarios, the results show that the longitudinal/lateral error is under 30 cm.
2.The simulation tests and development support of intelligent truck platoon (Cooperator: SAIC)
In order to cope with the tests and verification problem of intelligent truck platooning controller, the integral simulation platform for modelling, testing and the method of evaluation was constructed. Furthermore, calculating method and the multi-dimensional evaluation was proposed. The tests under various scenarios and driving circles were conducted based on the platform. Moreover, the iterative development mechanism was formed with the help of this platform, which is beneficial for subsequent updating of the algorithm. The platoon system and the simulation platform has been utilized in “5G+L4” intelligent truck project, which has reached 20000 TEU/Y in the transportation of Donghai bridge linking Yangshan port.