Research Of The Cooperative Control Method Of The Intersection For Autonomous Connected Vehicle

The emergence of autonomous vehicles with precise control and sensing capabilities has brought new hope for building a more safety and efficient transportation system. Existing artificial intelligence technology has been able to solve the driving problem effectively in an open road environment. However, when face to the intersection area that has running track conflict and competition prevailing restrictions, both the safety and efficiency of the autonomous vehicles cannot be guaranteed. In recent years, the connected vehicle technology that is aim at cooperatively integrating various elements of the traffic system, gradually become a hot topic in the field of intelligent transportation. It is considered to be a powerful technical to solve traffic problems and intersection control is one of its key attention content.Based on the technology feature that the vehicle and road infrastructure can achieve real-time communication under connected vehicle environment, this paper designed a cooperative control method for autonomous vehicle under connected vehicle environment, which can improve the efficiency and safety of the autonomous vehicle in urban intersection by fully exert autonomous vehicle’s high precision control and strong networking capabilities. The main contents are as follows:(1) Modeling the intersection networked cooperative control problem. This paper designed a cooperative control method based on "buffer-reservation" mechanism by dividing the road network into three logical control area(free driving area, core areas, buffer area), then pointed out the two key steps that to achieve "buffer-reservation" mechanism: crossing time-space resource allocation and the vehicle speed adjustment. For achieving the two key steps, established intersection time span allocation model and vehicle speed control function model. Moreover, a control failure handling mechanism is proposed for improving the fault tolerance of "buffer-reservation" mechanism.(2) Design Algorithm for the networking cooperative control model. For the intersection time span allocation model, this paper proposed an algorithm based on set-projection theory to find the optimal solution of crossing time span; for vehicle speed control function model that related to the functional analysis, this paper use an algorithm based on the three-piece-linear decomposition for the speed control function to get the dynamics control parameters.(3) Develop the simulation system and make simulation. We via C#.NET programming language to make a secondary development of VISSIM COM interfaces, to build a simulation environment – intersection networking cooperative control simulation system. This system that has parallel dual network system architecture, can exchange basis data with VISSIM, and we make a series of simulation experiments based on this system to verify the effectiveness and feasibility of the proposed method. By the collection and analysis of the simulation data, the results demonstrate that the proposed method can significantly increase the capacity of intersection and reduce travel delay, stop time, fuel consumption, and gas emissions.