Research On Real-time Estimation Of Lane-level Queue Length On Expressways Based On Instrumented Vehicles Data

With the advantages of high efficiency,safety and strong traffic capacity,expressways have developed into the arteries of the urban transportation system.With the increase in car ownership,traffic congestion on expressways is frequent.Congestion not only reduces the capacity of expressways,but also easily leads to traffic accidents.Therefore,it is necessary to conduct real-time monitoring of traffic congestion in order to implement timely control measures.However,due to the random burstiness in time and space,traditional fixed-point detection-based methods cannot effectively monitor the occurrence and development of traffic congestion.In recent years,with the rapid development of intelligent networked driving technology,instrumented vehicles(such as Advanced Driving Assistance Vehicles and autonomous vehicles)will be gradually applied.At that time,benefit from high-precision positioning and environmental perception capabilities,instrumented vehicles are expected to become a new traffic detection technology.Based on the traffic shockwave theory,this paper proposes a real-time estimation method of lane-level queue length by combining the spatiotemporal trajectory characteristics of vehicles under different traffic conditions(congested and free flow)and the characteristics of instrumented vehicle data.The proposed methodology includes five steps:(1)In view of the problem that the trajectory obtained by the instrument vehicle has errors,the trajectory noise reduction method based on wavelet analysis is used for noise reduction;(2)Based on the relationship between the trajectory characteristics and the traffic state,identify the turning points that reflect the sudden change of the traffic state in the trajectory of the instrumented vehicle data;(3)In order to determine the traffic shockwave to which the turning point belongs,a turning point classification algorithm based on the temporal and spatial characteristics of the traffic shockwave is proposed;(4)Calculate traffic shockwave speed based on turning points;(5)Determine the queue length based on turning point and traffic shockwave.In order to test the effect of the proposed method in detecting traffic congestion on expressways,this paper uses the microscopic traffic simulation data under two different congestion scenarios on the expressway to process the trajectory data of different instrumented vehicles penetration rates in the two traffic scenarios.The detection effect of traffic congestion(location and propagation speed)and the queuing delay caused by it are tested under various instrumented vehicles proportions.Experiments show that with no more than 9% of instrumented vehicles,the method proposed in this paper can detect sudden traffic jams on expressways more accurately in real time.