Study On A Novel Merging Trajectory Planning Framework For Automated Vehicles In A Connected Vehicles Environment

In recent years,with the expansion of the Internet of Vehicles technology and the improvement of the level of automation,the automated vehicle has received more and more attention from Internet companies and society.Automated driving technology benefits society,drivers and pedestrians,and can significantly reduce the overall traffic accident rate.In addition,the driving mode of the automated vehicle is more energy-efficient and efficient,and traffic congestion and traffic pollution will be reduced.It is considered to be an important means to solve traffic problems in the future.Therefore,it has also attracted the attention of many scholars in the field of transportation.Among them,the lane-changing and merging technology of the automated vehicle is one of the key issues of the automated driving technology.At present,the research on the merging of the automated vehicle has achieved certain results,but the trajectory planning model of the merging is not perfect.The freeway ramp junction area is a common accident-prone area,and it is easy to cause traffic disorder and reduce traffic efficiency.In order to improve the safety and efficiency of the merging of autonomous vehicles in the confluence area of expressway ramps,a trajectory planning model for the merging of the automated vehicle based on the environment of connected vehicles is proposed.In the Internet of Vehicles environment,the automated vehicle on the ramp can obtain real-time position and speed information of several vehicles on the main lane to obtain the gaps.First,several gaps are selected through the safety rule model,redundant rule model and limit rule model established in this paper,and then execute the optimal merging trajectory planning for the merging gaps obtained after screening.Second,the collision avoidance model is used to generate the collision area that exists in the process of the automated vehicle merging,and let the trajectory decision model adjust the merging trajectory in the conflict area to ensure the safety of the merging trajectory.Third,we introduce a cost function based on efficiency and comfort to compare and analyze each merging trajectory,and then obtain an optimal merging trajectory with the highest efficiency and safety.Finally,this article uses Python simulation to verify the adaptability of our model to different scenarios.We analyzed the influence of multiple factors in different scenarios on the success rate,speed and time of automated vehicles merging into the main lane.The results show that,compared with the success rate of the automated vehicle without model control,the success rate of the vehicle with model control is greatly improved by 24%in the same scenario.In addition,the average merging time of the automated vehicle with model control is reduced by 3.4s,which greatly improves the merging efficiency.It proves that this model established in this article has better applicability.Second,under the control of the model in this article,different initial speeds and initial accelerations have no significant influence on the success rate and merging speed of the vehicles.However,they have a significant influence on the merging time.Third,our simulation shows that different traffic flow densities have no significant impact on the merging speed and the merging time of the automated vehicles.However,they have a significant impact on the merging success rate.Finally,our simulation also analyzes how the the length of the ramp affects the merging behavior of the automated vehicle.The results show that when the length of the ramp continues to increase,the merging success rate of the vehicle is also increasing.Furthermore,when the ramp exceeds a certain length,the length of the ramp will no longer affect the success rate of the vehicle.In the Internet of Vehicles environment,this critical length provides a certain reference value for ramp planning and design.