CN-116959276-B - Intersection signal priority method, system, equipment and medium based on intelligent network-connected vehicle track prediction function

CN116959276BCN 116959276 BCN116959276 BCN 116959276BCN-116959276-B

Abstract

The invention provides an intersection signal priority method based on an intelligent network vehicle track prediction function, which comprises the steps of obtaining historical track data of motor vehicles at an intersection and real-time track data of signal priority vehicles, training a track prediction model according to the historical track data to obtain a trained track prediction model, wherein the trained track prediction model is used for predicting time required by the vehicles to reach an intersection stop line, obtaining time required by the signal priority vehicles to reach the intersection stop line according to the real-time track data of the signal priority vehicles and the trained track prediction model, obtaining a signal timing scheme of the intersection, dynamically adjusting the signal timing scheme according to the time required by the signal priority vehicles to reach the intersection stop line, obtaining the signal priority timing scheme, and controlling signal timing according to the signal priority timing scheme.

Inventors

YUAN YUEMING
CAI HUIXING
GAO YIHONG

Assignees

云控智行(上海)汽车科技有限公司

Dates

Publication Date: 20260512
Application Date: 20230724

Claims (8)

1. An intersection signal priority method based on intelligent network-connected vehicle track prediction function is characterized by comprising the following steps: acquiring historical track data of motor vehicles at an intersection and real-time track data of signal priority vehicles, wherein the signal priority vehicles are vehicles needing to pass preferentially; Training a track prediction model according to the historical track data to obtain a trained track prediction model, wherein the trained track prediction model is used for predicting the time required by a vehicle to reach an intersection stop line; obtaining the time length required by the signal priority vehicle to reach the stop line of the intersection according to the real-time track data of the signal priority vehicle and the trained track prediction model; Acquiring a signal timing scheme of an intersection, and dynamically adjusting the signal timing scheme according to the length of time required by the signal priority vehicle to reach an intersection stop line to obtain the signal priority timing scheme; Signal lamp timing control is carried out according to the signal priority timing scheme; the training of the track prediction model according to the historical track data comprises the following steps: dividing the intersection range of the signal lamp to obtain a plurality of sub-road sections; mapping the track sequence data of the vehicle to the sub-road sections, and determining a probability function according to the number of track positions in the sub-road sections; Calculating the track similarity between the real-time track data of the signal priority vehicle and the track prediction model according to the probability function; Determining a calculation result with highest similarity among calculation results based on track similarity, and taking the calculation result as a vehicle track prediction set; according to the probability function, and calculating the track similarity comprises: Determining the probability function through non-parameter fitting, and calculating the track similarity between the real-time track data of the signal priority vehicle and the track prediction model through similarity calculation; and taking the time required by the vehicle track prediction set to reach the end point as the time required by the signal to prioritize the vehicle to reach the intersection stop line.
2. The intersection signal prioritization method of claim 1, wherein the acquiring historical track data of intersection motor vehicles and real-time track data of signal prioritized vehicles includes: And acquiring track sequence data of the vehicle in the range of the signal lamp intersection uploaded by the road side perception system in real time through a cloud control platform in an intelligent networking environment.
3. The intersection signal prioritization method of claim 1, wherein the dynamically adjusting the signal timing scheme based on the length of time required for the signal-prioritized vehicle to reach an intersection stop line includes: Obtaining the current green light remaining time of the intersection and the green light time of the intersection according to the timing scheme; comparing the time required by the signal priority vehicle to reach the stop line of the intersection with the current green light remaining time of the intersection; When the time required by the signal priority vehicle to reach the stop line of the intersection is smaller than the current green light remaining time of the intersection, keeping the current signal timing scheme unchanged; When the time required by the signal priority vehicle to reach the intersection stop line is longer than the current green light remaining time of the intersection, calculating the difference value between the time required by the signal priority vehicle to reach the intersection stop line and the current green light remaining time of the intersection, adding the difference value and the intersection timing green light time to obtain a sum value, if the sum value is smaller than the maximum green light time allowed by the intersection, prolonging the green light timing of the signal timing scheme, and if the sum value is larger than the maximum green light time allowed by the intersection, keeping the current signal timing scheme unchanged.
4. The intersection signal prioritization method of claim 1, wherein the dynamically adjusting the signal timing scheme based on the length of time required for the signal-prioritized vehicle to reach an intersection stop line includes: Obtaining the current red light remaining time of the intersection and the time length of the red light when the intersection is timed according to the time-timing scheme; Comparing the time required by the signal priority vehicle to reach the stop line of the intersection with the current red light remaining time of the intersection; when the signal priority vehicle reaches the stop line of the intersection, the length of time required by the signal priority vehicle to reach the stop line of the intersection is longer than the current red light remaining time of the intersection, and the current signal timing scheme is kept unchanged; When the time required by the signal priority vehicle to reach the stop line of the intersection is smaller than the current red light remaining time of the intersection, calculating a difference value between the time required by the signal priority vehicle to reach the stop line of the intersection and the current red light remaining time of the intersection, subtracting the time required by the signal priority vehicle to reach the current red light remaining time of the intersection to obtain a difference value, if the difference value is larger than the minimum red light time allowed by the intersection, reducing the red light timing of the signal timing scheme, and if the difference value is smaller than the minimum red light time allowed by the intersection, keeping the current signal timing scheme unchanged.
5. The intersection signal prioritization method of claim 1, wherein the signal timing control according to the signal priority timing scheme includes: and sharing the signal priority timing scheme to a traffic signal control system of a traffic department through a cloud control platform, and performing signal timing control.
6. An intersection signal priority system based on intelligent network-connected vehicle track prediction function, characterized in that the system comprises: The prediction module is used for acquiring historical track data of motor vehicles at the intersection and real-time track data of signal priority vehicles, wherein the signal priority vehicles are vehicles needing to pass preferentially; training a track prediction model according to the historical track data to obtain a trained track prediction model, wherein the trained track prediction model is used for predicting the time required by a vehicle to reach an intersection stop line; the signal timing control module is used for dynamically adjusting the signal timing scheme according to the length required by the signal priority vehicle to reach the stop line of the intersection to obtain the signal priority timing scheme; the training of the track prediction model according to the historical track data comprises the following steps: dividing the intersection range of the signal lamp to obtain a plurality of sub-road sections; mapping the track sequence data of the vehicle to the sub-road sections, and determining a probability function according to the number of track positions in the sub-road sections; Calculating the track similarity between the real-time track data of the signal priority vehicle and the track prediction model according to the probability function; Determining a calculation result with highest similarity among calculation results based on track similarity, and taking the calculation result as a vehicle track prediction set; according to the probability function, and calculating the track similarity comprises: Determining the probability function through non-parameter fitting, and calculating the track similarity between the real-time track data of the signal priority vehicle and the track prediction model through similarity calculation; and taking the time required by the vehicle track prediction set to reach the end point as the time required by the signal to prioritize the vehicle to reach the intersection stop line.
7. An intelligent traffic signal control apparatus, the apparatus comprising: One or more processors, and a memory storing computer program instructions that, when executed, cause the processor to perform the method of any of claims 1-5.
8. A computer readable medium having stored thereon computer program instructions executable by a processor to implement the method of any of claims 1-5.

Description

Intersection signal priority method, system, equipment and medium based on intelligent network-connected vehicle track prediction function Technical Field The application relates to the field of intelligent traffic, in particular to an intersection signal priority method, system, equipment and readable medium based on an intelligent network-connected vehicle track prediction function. Background With the development of the automobile industry and the acceleration of the urban process, the problems of urban traffic jam, traffic accidents and the like become increasingly prominent. Traffic signal control is a common traffic management mode and has become an important means for solving the problems of urban traffic jams, traffic accidents and the like. In the technical field of intelligent transportation, with the continuous development of intelligent and automatic technologies, intelligent network-connected automobiles are also increasingly widely applied. The intelligent network-connected vehicle has the capability of acquiring the vehicle and road condition information in real time, and can communicate and cooperate with traffic facilities and other vehicles to realize more efficient, safe and intelligent transportation. However, at traffic bottlenecks such as intersections, the advantages of intelligent network-connected vehicles are not fully exerted, because the intersection signal control system is usually based on a static timing scheme, cannot be dynamically adjusted according to real-time traffic conditions, and is difficult to support the signal preferential traffic of the intelligent network-connected vehicles. Disclosure of Invention The application aims to provide an intersection signal priority algorithm, an intersection signal priority system, intersection signal priority equipment and an intersection signal priority device based on an intelligent network vehicle track prediction function, which are at least used for enabling the method to dynamically predict vehicle tracks, improving the passing efficiency of vehicles needing to pass preferentially and relieving traffic jams. The application provides an intersection signal priority method based on an intelligent network vehicle track prediction function, which comprises the steps of obtaining historical track data of motor vehicles at an intersection and real-time track data of signal priority vehicles, training a track prediction model according to the historical track data to obtain a trained track prediction model, wherein the trained track prediction model is used for predicting time required by the vehicles to reach an intersection stop line, obtaining time required by the signal priority vehicles to reach the intersection stop line according to the real-time track data of the signal priority vehicles and the trained track prediction model, obtaining a signal timing scheme of the intersection, dynamically adjusting the signal timing scheme according to the time required by the signal priority vehicles to reach the intersection stop line, obtaining the signal priority timing scheme, and controlling signal timing according to the signal priority timing. Further, the step of acquiring the historical track data of the motor vehicles at the intersection and the real-time track data of the signal priority vehicles comprises the step of acquiring the track sequence data of the vehicles in the signal lamp intersection range uploaded by the road side perception system in real time through a cloud control platform in an intelligent networking environment. Further, the training of the track prediction model according to the historical track data comprises dividing the signal lamp intersection range to obtain a plurality of sub-road sections; the method comprises the steps of mapping track sequence data of the vehicles to the sub-road sections, determining probability functions according to the number of track positions in the sub-road sections, calculating track similarity between real-time track data of the vehicles with signal priority and the track prediction model according to the probability functions, and determining a calculation result with highest similarity in calculation results based on the track similarity as a vehicle track prediction set. Further, the determining the probability function according to the random variable and calculating the track similarity comprises the steps of determining the probability function of the random variable through non-parametric fitting and calculating the track similarity between the real-time track data of the signal priority vehicle and the track prediction model through similarity calculation. Further, the signal timing scheme is dynamically adjusted according to the required length of the signal priority vehicle reaching the intersection stop line, the signal timing scheme comprises the steps of obtaining the current green light remaining time of the intersection and the intersection timing green l