A traffic flow collaborative decision-making optimization method and system

Abstract

The present invention provides a traffic flow collaborative decision optimization method and system, which relates to the field of autonomous driving technology, including: obtaining conflicting intersections in a traffic network in a conflicting area; determining the conflict type of the conflicting area according to the conflicting intersections, and when the conflicting area is a complex intersection type, determining all autonomous driving vehicles heading to each conflicting intersection in the crossroads, and calculating the arrival time of each autonomous driving vehicle at the corresponding conflicting intersection, and generating an estimated arrival time matrix; obtaining an arrival order matrix according to the order in which all autonomous driving vehicles pass through the corresponding conflicting intersections; determining a target conflicting intersection according to the arrival order matrix, updating the estimated arrival time matrix, obtaining an updated estimated arrival time matrix, and then making traffic decisions for each autonomous driving vehicle at each conflicting intersection. The present invention improves the problems of traffic flow congestion and inefficiency of autonomous driving vehicles in complex intersection interaction scenarios.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a traffic flow collaborative decision-making optimization method and system. Background Art

[0002] In the field of intelligent transportation systems (ITS), multi-agent interaction technology is committed to improving traffic flow optimization and road use efficiency through coordination and cooperation among agents. Currently, many studies focus on traffic management strategies at single intersections, which mainly focus on how to optimize traffic signal control and the passage order of autonomous vehicles at single intersections.

[0003] However, these studies often ignore the mutual influence between intersections and the complex traffic network formed when multiple intersections are close to each other. With the acceleration of urbanization, the traffic network has become more and more complex, especially in the central area of ​​the city, where multiple intersections are close to each other, forming complex traffic interaction scenarios. When multiple intersections are close to each other, complex intersection scenarios will appear. Traditional single intersection optimization strategies are insufficient to cope with the complexity and dynamic changes of traffic flow in complex intersection interaction scenarios due to the lack of coordination of the overall traffic flow, which seriously affects the traffic flow efficiency and leads to frequent traffic congestion and conflicts. Summary of the invention

[0004] The problem solved by the present invention is how to solve the traffic congestion and low efficiency problems of autonomous driving vehicles in complex intersection interaction scenarios.

[0005] In order to solve the above problems, the present invention provides a traffic flow collaborative decision-making optimization method and system.

[0006] In a first aspect, the present invention provides a traffic flow collaborative decision-making optimization method, comprising:

[0007] Obtain conflicting intersections of the traffic network in the conflicting area;

[0008] Determine a plurality of intersecting roads according to the conflicting intersections, and judge the conflict type of the conflicting area according to the number of the conflicting intersections in the conflicting area, wherein when there are a plurality of the conflicting intersections in the conflicting area, the conflicting area is judged to be a complex intersection type;

[0009] When the conflict area is of the complex intersection type, determining all autonomous driving vehicles on the crossroad heading for each of the conflict intersections, and calculating the arrival time of each of the autonomous driving vehicles at the corresponding conflict intersection;

[0010] Generate an estimated arrival time matrix according to the arrival time of the autonomous driving vehicle, wherein the estimated arrival time matrix includes the arrival time of each autonomous driving vehicle at the corresponding conflicting intersection;

[0011] Obtaining an arrival order matrix according to the order in which all the autonomous driving vehicles pass through the corresponding conflicting intersections;

[0012] Determine a target conflict intersection according to the arrival order matrix, and update the estimated arrival time matrix based on the arrival time corresponding to the autonomous driving vehicle that first arrives at the target conflict intersection to obtain an updated estimated arrival time matrix, wherein when the conflict intersection in the driving routes of all the autonomous driving vehicles passing through the conflict intersection is always the first conflict intersection arrived by the corresponding autonomous driving vehicle, the conflict intersection is determined to be the target conflict intersection;

[0013] According to the updated estimated arrival time matrix, traffic decisions are made for each autonomous driving vehicle at each conflicting intersection.

[0014] Optionally, determining a plurality of intersecting roads according to the conflicting intersection includes:

[0015] Taking the conflicting intersection as an intersection, obtaining coordinate information of the intersection in the conflicting area;

[0016] Determine a plurality of roads passing through the intersection according to the coordinate information of the intersection and in combination with the road topology of the traffic network in the conflict area;

[0017] Each of the roads passing through the intersection is regarded as the intersection road.

[0018] Optionally, the determining all the autonomous driving vehicles on the crossroad heading for each of the conflicting intersections, and respectively calculating the arrival time of each of the autonomous driving vehicles at the corresponding conflicting intersection, includes:

[0019] Obtaining basic information of each autonomous driving vehicle on the intersection road, the basic information including the real-time position coordinates, driving speed, driving direction, surrounding road conditions, and traffic signal status of the autonomous driving vehicle;

[0020] Determine, according to the driving direction of the autonomous driving vehicle, all the autonomous driving vehicles on the crossroad heading towards the conflicting intersection;

[0021] Calculate the driving distance of each autonomous driving vehicle to the conflicting intersection according to the real-time position coordinates of each autonomous driving vehicle;

[0022] The arrival time of each of the autonomous driving vehicles heading for the conflicting intersection at the corresponding conflicting intersection is determined based on the driving speed, the driving distance, the surrounding road conditions, and the traffic signal status of each of the autonomous driving vehicles.

[0023] Optionally, generating an estimated arrival time matrix according to the arrival time of the autonomous driving vehicle includes:

[0024] For one of the autonomous driving vehicles, generating a time array of the autonomous driving vehicle arriving at each of the conflicting intersections according to the arrival time of the autonomous driving vehicle at each of the conflicting intersections;

[0025] The estimated arrival time matrix is ​​formed according to the time arrays corresponding to all the autonomous driving vehicles.

[0026] Optionally, obtaining an arrival order matrix according to the order in which all the autonomous driving vehicles pass through the corresponding conflicting intersections includes:

[0027] For one of the autonomous driving vehicles, determining the order in which the autonomous driving vehicle passes through each of the conflicting intersections according to the real-time position coordinates and driving direction of the autonomous driving vehicle;

[0028] According to the passing order corresponding to the autonomous driving vehicle, the conflicting intersections that the autonomous driving vehicle will pass through are sorted to obtain an intersection passing array corresponding to the autonomous driving vehicle;

[0029] The arrival order matrix is ​​obtained based on the intersection passing array of all the autonomous driving vehicles.

[0030] Optionally, the updating of the estimated arrival time matrix based on the arrival time corresponding to the autonomous driving vehicle that first arrives at the target conflict intersection to obtain the updated estimated arrival time matrix includes:

[0031] Determining whether there are multiple target autonomous driving vehicles passing through the target conflict intersection according to the arrival order matrix, wherein the target driving vehicle is an autonomous driving vehicle whose target conflict intersection is the first conflict intersection in its driving route;

[0032] If so, all the target autonomous driving vehicles are sorted according to the arrival time of each target autonomous driving vehicle at the target conflict intersection, and the arrival time of other target autonomous driving vehicles at the target conflict intersection is updated according to the arrival time corresponding to the target autonomous driving vehicle that arrives at the target conflict intersection first, the preset minimum time interval and the sorting result;

[0033] According to the updated arrival time of all the target autonomous driving vehicles at the target conflicting intersection, the arrival time of the target autonomous driving vehicles at other conflicting intersections is updated to generate an updated estimated arrival time matrix.

[0034] Optionally, making a traffic decision for the autonomous driving vehicle at the conflicting intersection according to the updated estimated arrival time matrix includes:

[0035] According to the updated estimated arrival time matrix, the autonomous driving vehicles at all the conflicting intersections are queued to obtain a final sorting result of the autonomous driving vehicles;

[0036] A traffic strategy for the conflicting area is generated according to the final sorting result, and a traffic decision is made for the autonomous driving vehicle at the conflicting intersection according to the traffic strategy.

[0037] Optionally, the conflict type further includes a common intersection type, and judging the conflict type of the conflict area according to the number of the conflict intersections in the conflict area further includes:

[0038] When there is a conflicting intersection in the conflicting area, it is determined that the conflicting area is a common intersection type.

[0039] Optionally, it also includes:

[0040] When the conflict area is of the ordinary intersection type, determining all autonomous driving vehicles on the crossroad heading toward the conflict intersection, and calculating the arrival time of each autonomous driving vehicle at the conflict intersection;

[0041] Generate an estimated arrival time matrix corresponding to the conflicting intersection according to the arrival time of the autonomous driving vehicle;

[0042] A traffic strategy is generated based on the estimated arrival time matrix, and a traffic decision is made for the autonomous driving vehicles at the conflicting intersection.

[0043] In a second aspect, the present invention provides a traffic flow collaborative decision-making optimization system, comprising:

[0044] An intersection acquisition unit, used for acquiring conflicting intersections of a traffic network in a conflicting area;

[0045] a type identification unit, configured to determine a plurality of intersecting roads according to the conflicting intersection, and to judge the conflict type of the conflicting area according to the number of the conflicting intersections in the conflicting area, wherein when there are a plurality of the conflicting intersections in the conflicting area, the conflicting area is determined to be a complex intersection type;

[0046] a calculation unit, configured to, when the conflict area is of the complex intersection type, determine all autonomous driving vehicles in the crossroad heading for each of the conflict intersections, and respectively calculate an arrival time of each of the autonomous driving vehicles at the corresponding conflict intersection;

[0047] A time matrix construction unit, configured to generate an estimated arrival time matrix according to the arrival time of the autonomous driving vehicle, wherein the estimated arrival time matrix includes the arrival time of each autonomous driving vehicle at the corresponding conflicting intersection;

[0048] An order matrix construction unit is used to obtain an arrival order matrix according to the order in which all the autonomous driving vehicles pass through the corresponding conflicting intersections;

[0049] an updating unit, configured to determine a target conflict intersection according to the arrival order matrix, and update the estimated arrival time matrix based on the arrival time corresponding to the autonomous driving vehicle that first arrives at the target conflict intersection, to obtain the updated estimated arrival time matrix, wherein when the conflict intersection in the driving routes of all the autonomous driving vehicles passing through the conflict intersection is always the first conflict intersection arrived by the corresponding autonomous driving vehicle, the conflict intersection is determined to be the target conflict intersection;

[0050] A decision-making unit is used to make traffic decisions for each autonomous driving vehicle at each of the conflicting intersections based on the updated estimated arrival time matrix.

[0051] The traffic flow collaborative decision optimization method and system of the present invention accurately locates the traffic conflict area by identifying the conflict intersections in the traffic network and determining the crossroads. Since multiple conflict intersections influence each other, the conflict area can be classified by the number of conflict intersections. When multiple conflict intersections are identified, the area is determined to be a complex intersection type. At this time, complex processing logic is required to deal with the interaction problem of multiple intersections. Specifically, by calculating the estimated time for each vehicle to arrive at each conflict intersection, a comprehensive estimated arrival time matrix is ​​generated, which provides a data basis for subsequent traffic sorting and decision-making. Then, by sorting all autonomous driving vehicles, an arrival order matrix is ​​obtained, and the order in which the autonomous driving vehicles pass through the corresponding conflict intersections is determined, thereby reducing congestion at the intersection. In addition, the mutual influence of multiple intersection areas at complex intersections is comprehensively considered, avoiding the multi-vehicle conflict blocking and locking phenomenon caused by the queuing method of ordinary intersections. By updating the estimated arrival time matrix in real time, the smooth passage of vehicles at complex intersections is ensured, and the traffic efficiency of traffic flow is improved. In addition, the present invention is also practical and can adapt to different traffic scenarios, such as the situation where multiple intersections appear within a short distance. Finally, according to the updated estimated arrival time matrix, traffic decisions are made for the autonomous driving vehicles at each conflicting intersection, such as executing acceleration and deceleration actions, ensuring the smoothness and efficiency of traffic flow and avoiding the problem of low traffic efficiency caused by queuing and blocking. The present invention improves the intelligent level of traffic management and ensures the efficiency and safety of traffic flow through precise time calculation, autonomous driving vehicle sorting and dynamic time update. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A flow chart of a traffic flow collaborative decision-making optimization method according to an embodiment of the present invention;

[0053] Figure 2 A schematic diagram of a complex intersection scene according to an embodiment of the present invention;

[0054] Figure 3 This is one of the schematic diagrams of the estimated arrival time matrix according to an embodiment of the present invention;

[0055] Figure 4 This is a schematic diagram of the order in which an autonomous driving vehicle arrives at an intersection according to an embodiment of the present invention;

[0056] Figure 5 This is a second schematic diagram of an estimated arrival time matrix according to an embodiment of the present invention;

[0057] Figure 6 A schematic diagram of a common intersection scene according to an embodiment of the present invention;

[0058] Figure 7 It is a structural block diagram of the traffic flow collaborative decision-making optimization system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be interpreted as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention.

[0060] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0061] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to"; the term "based on" means "based at least in part on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0062] It should be noted that the modifications of "one" and "plurality" mentioned in the present invention are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0063] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes, and are not used to limit the scope of these messages or information.

[0064] Combination Figure 1 As shown, the present invention provides a traffic flow collaborative decision-making optimization method, comprising:

[0065] Get the conflicting intersections of the traffic network in the conflict area.

[0066] Specifically, a database or a real-time monitoring system of a traffic management system identifies intersections where traffic conflicts exist in the traffic network.

[0067] A plurality of intersecting roads are determined according to the conflicting intersections, and the conflict type of the conflicting area is judged according to the number of the conflicting intersections in the conflicting area, wherein when there are a plurality of the conflicting intersections in the conflicting area, the conflicting area is determined to be a complex intersection type.

[0068] In a preferred embodiment of the present invention, the conflicting areas can be classified according to the number of conflicting intersections. Figure 2 As shown in the figure, for a complex intersection scene, there are two adjacent intersection conflict points X1 and X2, which will affect the decision results of the autonomous driving vehicles on each road. Therefore, this area is regarded as a complex intersection type.

[0069] Specifically, when conflicting intersections are identified, the system analyzes the road layout around these intersections, determines all roads that intersect with them, and determines the complexity of the area by counting the number of conflicting intersections; if there are multiple conflicting intersections, such as two or more, the area is determined to be a complex intersection type, and more complex decision-making logic is adopted to handle traffic flow.

[0070] When the conflict area is of the complex intersection type, all autonomous driving vehicles on the crossroad heading for each of the conflict intersections are determined, and the arrival time of each of the autonomous driving vehicles at the corresponding conflict intersection is calculated respectively.

[0071] Specifically, in complex intersection type areas, the system will monitor all autonomous vehicles approaching these conflicting intersections. Using the real-time position, speed, and road conditions of the autonomous vehicles, the system calculates the estimated time for each vehicle to arrive at each conflicting intersection.

[0072] Based on the arrival time of the autonomous driving vehicles, an estimated arrival time matrix is ​​generated, wherein the estimated arrival time matrix includes the arrival time of each of the autonomous driving vehicles at the corresponding conflicting intersection.

[0073] Specifically, the arrival times of all autonomous vehicles are integrated into an estimated arrival time matrix, where the rows of the matrix represent a conflicting intersection, the columns represent an autonomous vehicle, and each element in the matrix is ​​the estimated time for the autonomous vehicle to arrive at the intersection. The estimated arrival time matrix provides a quantitative data basis for traffic decision-making.

[0074] An arrival order matrix is ​​obtained according to the order in which all the autonomous driving vehicles pass through the corresponding conflicting intersections.

[0075] Specifically, the estimated arrival time matrix is ​​used to sort the arrival time of each vehicle at each conflicting intersection to determine the order in which the vehicles pass through the intersection; and an arrival order list is created for each conflicting intersection based on the order in which the vehicles arrive, which is used to reflect the priority of the vehicles passing through the intersection; the arrival order matrix determines the order in which the autonomous vehicles pass through the intersection, aiming to optimize traffic flow and reduce congestion. By generating the arrival order matrix, the order in which each autonomous vehicle passes through each conflicting intersection can be clearly determined, and the vehicle's running path and time can be optimized accordingly, reducing the waiting time at the intersection, thereby making the management and scheduling of traffic flow more accurate and efficient, and improving the overall traffic flow efficiency.

[0076] The target conflict intersection is determined according to the arrival order matrix, and the estimated arrival time matrix is ​​updated based on the arrival time corresponding to the autonomous driving vehicle that first arrives at the target conflict intersection to obtain the updated estimated arrival time matrix, wherein when the conflict intersection in the driving routes of all the autonomous driving vehicles passing through the conflict intersection is always the first conflict intersection arrived by the corresponding autonomous driving vehicle, the conflict intersection is determined to be the target conflict intersection.

[0077] Specifically, by calculating the time it takes for each vehicle to arrive at each intersection, an estimated arrival time matrix is ​​constructed, and the order in which each vehicle arrives at each intersection is determined to obtain an arrival order matrix; and on this basis, the intersections with the first arrival order are prioritized, and the arrival time of the vehicle is updated according to the minimum time interval between two vehicles, and finally the final arrival time matrix is ​​output, and each intersection is sorted according to the final arrival time matrix.

[0078] The present invention can simultaneously meet the multi-vehicle interactive decision-making requirements of ordinary intersections and complex intersections.

[0079] For example, for ordinary intersections, the vehicle information near the intersection can be transmitted to the central control platform through V2X technology. The central control platform sorts the vehicles in the system according to the time when each vehicle arrives at the intersection to form a virtual queue, and controls the speed of each vehicle to ensure that the vehicles pass through the intersection in sequence and ensure safety.

[0080] For complex intersections, taking into account the mutual influence of multiple intersection areas, a vehicle may pass through two intersections. The expected arrival time of the first arrival point is obtained according to the sorting rules, and then the time of the second arrival point is updated according to the distance between the first arrival point and the second arrival point. Finally, the queue of the second arrival point is sorted, which fully considers the mutual influence relationship in the scenario where there are two adjacent intersections in the system.

[0081] According to the updated estimated arrival time matrix, traffic decisions are made for each autonomous driving vehicle at each conflicting intersection.

[0082] Specifically, based on the updated estimated arrival time matrix, a decision is made on the passage of each autonomous driving vehicle at each conflicting intersection.

[0083] The traffic flow collaborative decision optimization method of the present invention accurately locates the traffic conflict area by identifying the conflict intersections in the traffic network and determining the crossroads. Since multiple conflict intersections influence each other, the conflict area can be classified by the number of conflict intersections. When multiple conflict intersections are identified, the area is determined to be a complex intersection type. At this time, complex processing logic is required to deal with the interaction problem of multiple intersections. Specifically, by calculating the estimated time for each vehicle to arrive at each conflict intersection, a comprehensive estimated arrival time matrix is ​​generated, which provides a data basis for subsequent traffic sorting and decision-making. Then, by sorting all autonomous driving vehicles, an arrival order matrix is ​​obtained, and the order in which the autonomous driving vehicles pass through the corresponding conflict intersections is determined, thereby reducing the congestion of the intersection. In addition, the mutual influence of multiple intersection areas at complex intersections is comprehensively considered, avoiding the multi-vehicle conflict blocking and locking phenomenon caused by the queuing method of ordinary intersections. By updating the estimated arrival time matrix in real time, the smooth passage of vehicles at complex intersections is ensured, and the traffic efficiency of traffic flow is improved. In addition, the present invention is also practical and can adapt to different traffic scenarios, such as the situation where multiple intersections appear within a short distance. Finally, according to the updated estimated arrival time matrix, traffic decisions are made for the autonomous driving vehicles at each conflicting intersection, such as executing acceleration and deceleration actions, ensuring the smoothness and efficiency of traffic flow and avoiding the problem of low traffic efficiency caused by queuing and blocking. The present invention improves the intelligent level of traffic management and ensures the efficiency and safety of traffic flow through precise time calculation, autonomous driving vehicle sorting and dynamic time update.

[0084] Optionally, determining a plurality of intersecting roads according to the conflicting intersection includes:

[0085] Taking the conflicting intersection as an intersection, obtaining coordinate information of the intersection in the conflicting area;

[0086] Determine a plurality of roads passing through the intersection according to the coordinate information of the intersection and in combination with the road topology of the traffic network in the conflict area;

[0087] Each of the roads passing through the intersection is regarded as the intersection road.

[0088] Specifically, first, the system regards each conflicting intersection as an intersection, and obtains the specific coordinate information of these intersections in the conflict area through the geographic information system (GIS) or the V2X communication equipment of the autonomous driving vehicle, thereby ensuring the accuracy and real-time nature of the coordinate information. Then, combined with the road topology of the traffic network in the conflict area, all roads passing through these intersections are analyzed and determined, such as: main roads, secondary roads, and possible detours. By comprehensively considering the layout of the roads and traffic flow, all intersecting roads that may affect the traffic flow are fully identified; finally, these roads passing through the intersections are defined as intersecting roads, and specific identification is assigned to each road for precise management and scheduling in the subsequent decision-making process.

[0089] In this optional embodiment, by accurately locating intersections, the traffic management system can develop more accurate traffic flow optimization strategies for specific intersections and related roads. For example, the system can send real-time navigation instructions to the autonomous driving vehicle based on the traffic conditions of the intersections to select the optimal route, thereby avoiding congestion. This embodiment significantly improves the efficiency and safety of traffic flow. By accurately identifying intersections, the system can better predict and manage traffic flow and reduce delays caused by traffic congestion or accidents. In addition, it can also ensure that autonomous driving vehicles can pass safely in an optimized order and time interval when passing through complex intersections, which helps reduce the occurrence of traffic accidents.

[0090] Optionally, the determining all the autonomous driving vehicles on the crossroad heading for each of the conflicting intersections, and respectively calculating the arrival time of each of the autonomous driving vehicles at the corresponding conflicting intersection, includes:

[0091] Obtaining basic information of each autonomous driving vehicle on the intersection road, the basic information including the real-time position coordinates, driving speed, driving direction, surrounding road conditions, and traffic signal status of the autonomous driving vehicle;

[0092] Determine, according to the driving direction of the autonomous driving vehicle, all the autonomous driving vehicles on the crossroad heading towards the conflicting intersection;

[0093] Calculate the driving distance of each autonomous driving vehicle to the conflicting intersection according to the real-time position coordinates of each autonomous driving vehicle;

[0094] The arrival time of each of the autonomous driving vehicles heading for the conflicting intersection at the corresponding conflicting intersection is determined based on the driving speed, the driving distance, the surrounding road conditions, and the traffic signal status of each of the autonomous driving vehicles.

[0095] Specifically, the basic information of each autonomous vehicle in the crossroad is first obtained through the V2X communication device or traffic monitoring system of the autonomous vehicle. This information includes the real-time position coordinates, driving speed, driving direction, surrounding road conditions and traffic signal status of the autonomous vehicle. These data are used to provide comprehensive dynamic information of autonomous vehicles for the traffic management system, which is the basis for achieving accurate traffic flow optimization. Then, according to the driving direction of the autonomous vehicle, the autonomous vehicle position and road topology data are used to determine which autonomous vehicles are heading towards a specific conflict intersection, thereby ensuring that the system only focuses on autonomous vehicles that will truly affect the traffic flow at the conflict intersection, thereby saving resources. Next, according to the real-time position coordinates of each autonomous vehicle, combined with the straight line or actual driving distance between the current position of the autonomous vehicle and the conflict intersection, the driving distance of each autonomous vehicle heading to the conflict intersection to reach the corresponding conflict intersection is calculated. Finally, the estimated arrival time of each autonomous vehicle at the corresponding conflict intersection is determined by comprehensively considering the driving speed, surrounding road conditions and traffic signal status of each autonomous vehicle.

[0096] In this optional embodiment, by obtaining the basic information of the autonomous driving vehicle and calculating the arrival time, it is possible to provide the traffic management system with an accurate prediction of the arrival time of the autonomous driving vehicle; this enables the system to more effectively perform traffic scheduling, such as sending speed adjustment instructions to the autonomous driving vehicle to optimize traffic flow. By accurately predicting the arrival time of the autonomous driving vehicle, the system can reduce the waiting time of the autonomous driving vehicle at the intersection and reduce traffic congestion. At the same time, it also ensures that the autonomous driving vehicle passes safely in an optimized order and time interval when passing through complex intersections, which helps prevent traffic accidents. In addition, this embodiment can also improve the traffic management system's ability to respond to emergencies, such as traffic accidents or road construction, by adjusting the driving path and speed of the autonomous driving vehicle in real time to reduce the impact on traffic flow.

[0097] Optionally, generating an estimated arrival time matrix according to the arrival time of the autonomous driving vehicle includes:

[0098] For one of the autonomous driving vehicles, generating a time array of the autonomous driving vehicle arriving at each of the conflicting intersections according to the arrival time of the autonomous driving vehicle at each of the conflicting intersections;

[0099] The estimated arrival time matrix is ​​formed according to the time arrays corresponding to all the autonomous driving vehicles.

[0100] Specifically, for each autonomous driving vehicle, through a comprehensive analysis of the driving speed, road conditions, traffic signal status and possible traffic interference factors of each autonomous driving vehicle, the arrival time of the vehicle at each conflicting intersection is calculated and recorded to form a time array for the vehicle to arrive at each conflicting intersection; then the time arrays corresponding to all autonomous driving vehicles are aggregated to form an estimated arrival time matrix, which contains the estimated arrival time of all vehicles at each intersection to ensure the accuracy of the estimated time and provide data support for subsequent decision-making; and this estimated arrival time matrix not only includes the time information of the autonomous driving vehicle's arrival at each intersection, but also provides a clear and easy-to-process data structure for the traffic management system in the form of a matrix for subsequent traffic decision-making and optimization.

[0101] In a preferred embodiment of the present invention, Figure 2 As shown in the figure, during the peak hours in a traffic network in the center of a city, autonomous vehicles converge at these intersections from different directions, causing traffic congestion and potential conflict risks. For example, there are four vehicles (A0, A1, B1, C1) driving on two intersecting roads. They will pass through two adjacent intersections X1 and X2 respectively. This autonomous vehicle interaction scenario is more complicated than a single intersection, because the driving of autonomous vehicles is not only affected by a single intersection, but also by adjacent intersections. First, the intersection information is calculated based on the road topology to obtain intersections X1, X2...Xn, and then the arrival of autonomous vehicles at each intersection is calculated, and the corresponding arrival time is recorded as Among them, Ri represents the i-th vehicle on road R, and Xj represents the j-th intersection; then the estimated arrival time matrix is ​​constructed, and we get Figure 3 The estimated arrival time matrix shown in Figure 1 is Represents the time when vehicle A0 arrives at intersection X1. The above matrix is ​​constructed based on the actual number of intersections and the participating autonomous driving vehicles. The characteristic of this composite intersection decision is that a vehicle exists in multiple intersection queues, thus bringing a systematic impact on the decisions of different intersection areas.

[0102] In this optional embodiment, by generating an estimated arrival time matrix containing all conflicting intersections, the behavior of traffic flow in complex intersection areas can be analyzed and predicted; the estimated arrival time matrix enables the system to quickly identify potential traffic congestion points and autonomous vehicles that may need priority passage, and the estimated arrival time matrix enables the traffic management system to respond in advance, by sending real-time navigation instructions to autonomous vehicles or implementing other traffic guidance measures, effectively avoiding traffic congestion and conflicts, thereby achieving more refined traffic control. In addition, this embodiment also helps to reduce the waiting time of autonomous vehicles, reduce energy consumption and emissions, and improve the overall efficiency of road use.

[0103] Optionally, obtaining an arrival order matrix according to the order in which all the autonomous driving vehicles pass through the corresponding conflicting intersections includes:

[0104] For one of the autonomous driving vehicles, determining the order in which the autonomous driving vehicle passes through each of the conflicting intersections according to the real-time position coordinates and driving direction of the autonomous driving vehicle;

[0105] According to the passing order corresponding to the autonomous driving vehicle, the conflicting intersections that the autonomous driving vehicle will pass through are sorted to obtain an intersection passing array corresponding to the autonomous driving vehicle;

[0106] The arrival order matrix is ​​obtained based on the intersection passing array of all the autonomous driving vehicles.

[0107] Specifically, first, according to the real-time position coordinates and driving direction of the vehicle, the passing order of each conflicting intersection that the vehicle will pass is determined. Then, according to the passing order of each autonomous driving vehicle, the conflicting intersections that the vehicle will pass are sorted, and the intersection passing array corresponding to the vehicle is generated, and it is ensured that the passing order of each intersection matches its actual driving path. Finally, the intersection passing arrays of all autonomous driving vehicles are summarized to form a complete arrival order matrix. This arrival order matrix integrates the intersection passing order of all vehicles and provides an important reference for subsequent traffic management and decision-making. This embodiment can provide a clear driving path planning for each vehicle by accurately determining the order in which each autonomous driving vehicle passes each conflicting intersection, which not only takes into account the real-time position and driving direction of the vehicle, but also takes into account the complexity of the road network, including the existence of multiple intersections and potential conflicts between vehicles. It can ensure that each vehicle can pass through the intersection in the optimal order, thereby reducing the possibility of traffic congestion and conflict.

[0108] In a preferred embodiment of the present invention, the order in which each vehicle arrives at the intersection is determined according to the different arrival times. In combination with the above preferred embodiment, A1 passes through two intersections, so in combination with Figure 3As shown, judge That is, car A1 arrives at point X2 first, and then passes through point X1. Therefore, the arrival order of each car can be obtained as follows: Figure 4 As shown in , if the corresponding arrival order of a certain intersection is 1, the row is sorted first. Figure 4 As shown, first, the estimated arrival time matrix is ​​used to determine the time when each vehicle arrives at each intersection. For example, vehicle A1 is expected to arrive at intersection X2 first, and then at intersection X1. Then, based on the difference in the time when the vehicles arrive at the intersections, their arrival order is determined. Figure 4 In the arrival order matrix, if the arrival order of all vehicles to a certain intersection is 1, it means that these vehicles will pass through the intersection. For intersections with an arrival order of 1, the system will prioritize sorting. Figure 4 In the example, A1 and C1 pass through the intersection X2, so vehicles A1 and C1 are sorted first. During the sorting process, the actual distance of the vehicle to the intersection is taken into account; taking the X2 intersection as an example, C1 is closer to X2, so C1 will be scheduled to pass through the X2 intersection before A1. After determining that C1 arrives at X2 before A1, the system will update A1's arrival time based on the time C1 arrives at X2 and the preset minimum time interval. After A1 passes through the X2 intersection, its arrival time at the X1 intersection is adjusted accordingly. Then A1 uses the updated arrival time at the X1 intersection to participate in the arrival order sorting of the X1 intersection together with vehicles A0 and B1.

[0109] In this optional embodiment, an orderly traffic flow management method is provided for autonomous driving vehicles by generating an arrival order matrix. Through accurate path planning and determination of the order of passing through intersections, the traffic efficiency of intersections is effectively improved and the risk of traffic accidents is reduced. By generating an arrival order matrix, the passage of autonomous driving vehicles in complex intersection areas can be accurately sorted and scheduled; thereby helping to reduce the waiting time of autonomous driving vehicles at intersections and improve the continuity and traffic efficiency of traffic flow. The arrival order matrix enables the traffic management system to identify and resolve potential traffic conflicts in advance, and effectively prevent traffic congestion and accidents by optimizing the passage order of autonomous driving vehicles. In addition, this method can also improve the overall operating efficiency of the transportation network and reduce energy consumption to a certain extent.

[0110] Optionally, the updating of the estimated arrival time matrix based on the arrival time corresponding to the autonomous driving vehicle that first arrives at the target conflict intersection to obtain the updated estimated arrival time matrix includes:

[0111] Determining whether there are multiple target autonomous driving vehicles passing through the target conflict intersection according to the arrival order matrix, wherein the target driving vehicle is an autonomous driving vehicle whose target conflict intersection is the first conflict intersection in its driving route;

[0112] If so, all the target autonomous driving vehicles are sorted according to the arrival time of each target autonomous driving vehicle at the target conflict intersection, and the arrival time of other target autonomous driving vehicles at the target conflict intersection is updated according to the arrival time corresponding to the target autonomous driving vehicle that arrives at the target conflict intersection first, the preset minimum time interval and the sorting result;

[0113] According to the updated arrival time of all the target autonomous driving vehicles at the target conflicting intersection, the arrival time of the target autonomous driving vehicles at other conflicting intersections is updated to generate an updated estimated arrival time matrix.

[0114] Specifically, the time for the autonomous driving vehicle to arrive at the target conflict intersection is adjusted first, and other conflict intersections other than the target conflict intersection are processed later. First, all vehicles that use the target conflict intersection as the first conflict intersection are identified through the arrival order matrix; then, these vehicles are sorted according to the time they arrive at the target conflict intersection; after the sorting is completed, the arrival time of the first arriving vehicle is used as a benchmark, combined with the preset minimum time interval, to adjust the arrival time of subsequent vehicles. This ensures that there is enough safety distance between vehicles and prevents potential conflicts. Finally, according to the updated arrival time, the time for these vehicles to arrive at other conflict intersections is updated accordingly to generate a new estimated arrival time matrix. This embodiment determines whether there are multiple target autonomous driving vehicles based on the arrival order matrix. If there are multiple such vehicles, the system will sort them according to the time each vehicle arrives at the target conflict intersection, and update the arrival time of other target autonomous driving vehicles according to the arrival time of the first arriving vehicle, the preset minimum time interval and the sorting result.

[0115] In another embodiment of the present invention, while sorting the target vehicles, the time when other autonomous driving vehicles arrive at the same conflicting intersection is also considered; thereby ensuring that the arrival time of all vehicles can be reasonably arranged to avoid the situation where multiple vehicles arrive at the same intersection at the same time. First, the updated time information of the target autonomous driving vehicle arriving at other conflicting intersections except the target conflicting intersection is collected, and at the same time, the time information of other autonomous driving vehicles arriving at the same target conflicting intersection is also collected. Then, according to the collected time information, the time when all vehicles arrive at each conflicting intersection is sorted to determine the order in which the vehicles pass through each intersection. After the sorting is completed, the arrival time of the vehicle is adjusted based on the arrival order of the vehicle and the preset minimum time interval, thereby ensuring that there is enough time difference between the vehicles to avoid potential conflicts. Among them, the preset minimum time interval is a safety parameter used to ensure that the vehicles maintain a sufficient distance when passing through the intersection. The system will adjust the arrival time of the vehicle according to this interval to ensure that each vehicle has enough time to pass through the same intersection after the previous vehicle passes. Based on the sorting results and the preset minimum time interval, the arrival time of the vehicle is updated, including adjustments to the arrival time of the target vehicle and other related vehicles, such as fine-tuning the vehicle's driving speed (acceleration or deceleration) to ensure that the vehicle can arrive at the conflict intersection at the updated time, thereby ensuring smooth and safe traffic flow; finally, a new estimated arrival time matrix is ​​generated based on the updated arrival time. The estimated arrival time matrix is ​​used to reflect the latest estimated arrival time of all vehicles at each conflict intersection, providing the latest data support for traffic management and vehicle scheduling.

[0116] In this optional embodiment, by accurately updating the estimated arrival time matrix, the system can more effectively manage the driving order of vehicles and reduce the possibility of traffic congestion and conflict. In addition, this embodiment not only takes into account the driving conditions of a single vehicle, but also integrates the arrival time of all related vehicles to achieve comprehensive traffic optimization. This can effectively reduce traffic congestion, reduce the risk of traffic accidents, improve road use efficiency, and reduce energy consumption and environmental pollution. In addition, it also provides a smoother and safer driving experience for autonomous driving vehicles, enhances the reliability of the autonomous driving system and user satisfaction, and provides strong technical support for the development of intelligent transportation systems.

[0117] In a preferred embodiment of the present invention, Figures 2 to 5 As shown, the actual safety distance between A1 and C1 needs to be considered. Therefore, the time for A1 to arrive at point X2 needs to be updated. The update formula is:

[0118]

[0119] Among them, tm represents the updated time, and Interval represents the minimum time interval between two vehicles arriving at the same intersection. Since the time when A1 arrives at X1 has changed, the time when it arrives at X2 also needs to be adjusted accordingly, and is adjusted to:

[0120]

[0121] Combination Figure 5 As shown, a new estimated arrival time matrix is ​​obtained, and then the queue sorting of the two intersection points X1 and X2 is calculated respectively to obtain the final sorting result.

[0122] In this optional embodiment, the traffic management system improves the accuracy of the predicted arrival time of the autonomous vehicle by updating the estimated arrival time matrix; this method ensures that the autonomous vehicles maintain a safe time interval when passing through conflicting intersections, thereby reducing the possibility of traffic conflicts and congestion. The updated estimated arrival time matrix enables the traffic management system to dispatch autonomous vehicles more effectively, reduce waiting time at conflicting intersections, and reduce the risk of traffic accidents.

[0123] Optionally, making a traffic decision for the autonomous driving vehicle at the conflicting intersection according to the updated estimated arrival time matrix includes:

[0124] According to the updated estimated arrival time matrix, the autonomous driving vehicles at all the conflicting intersections are queued to obtain a final sorting result of the autonomous driving vehicles;

[0125] A traffic strategy for the conflicting area is generated according to the final sorting result, and a traffic decision is made for the autonomous driving vehicle at the conflicting intersection according to the traffic strategy.

[0126] Specifically, the arrival time of each autonomous vehicle is compared to determine the priority of autonomous vehicles passing through conflicting intersections. The system uses the updated estimated arrival time matrix to sort the queues of autonomous vehicles at all conflicting intersections; the sorting process also takes into account the arrival order of autonomous vehicles, the minimum time interval, and possible traffic interference factors to ensure the accuracy and practicality of the sorting; after the sorting is completed, the traffic strategy for the conflicting area is generated based on the final sorting results.

[0127] In this optional embodiment, by making traffic decisions based on the updated estimated arrival time matrix, this embodiment can achieve accurate control of traffic flow at conflicting intersections; it ensures that autonomous vehicles pass through intersections in an optimized order, reduces traffic conflicts and congestion, and improves the continuity of traffic flow. Through accurate traffic strategies, it is possible to reduce the waiting time of autonomous vehicles, reduce energy consumption, reduce emissions, and also have a positive impact on environmental protection.

[0128] Optionally, the conflict type further includes a common intersection type, and judging the conflict type of the conflict area according to the number of the conflict intersections in the conflict area further includes:

[0129] When there is a conflicting intersection in the conflicting area, it is determined that the conflicting area is a common intersection type.

[0130] Specifically, this embodiment can not only handle complex intersection types, that is, situations where there are multiple conflicting intersections, but also identify and handle common intersection types, that is, situations where there is only one conflicting intersection in the conflict area; when the system identifies that there is only one conflicting intersection in the conflict area through traffic monitoring equipment or database information, the area is determined to be a common intersection type. For common intersection types, the system will adopt a simplified traffic decision process. This includes collecting the autonomous driving vehicle information of the single conflicting intersection, calculating the arrival time of the autonomous driving vehicle, and generating an estimated arrival time matrix based on this information; then, based on this data, the autonomous driving vehicles are sorted to form an arrival order matrix, and traffic decisions are made for the autonomous driving vehicles according to the traffic strategy, such as sending speed adjustment instructions to the autonomous driving vehicles.

[0131] In this optional embodiment, by distinguishing between ordinary intersection types and complex intersection types, the method of the present invention can flexibly adapt to traffic scenarios of different scales and complexities; for ordinary intersections, the system can respond quickly, reduce unnecessary calculations and resource consumption, and ensure smooth traffic flow; and achieve effective traffic management at a lower computing cost, thereby improving the efficiency of traffic decision-making. In addition, by providing customized traffic decisions for different types of intersections, the overall performance of the traffic network can be further improved, providing support for the efficient operation of urban traffic.

[0132] Optionally, the traffic flow collaborative decision-making optimization method further includes:

[0133] When the conflict area is of the ordinary intersection type, determining all autonomous driving vehicles on the crossroad heading toward the conflict intersection, and calculating the arrival time of each autonomous driving vehicle at the conflict intersection;

[0134] Generate an estimated arrival time matrix corresponding to the conflicting intersection according to the arrival time of the autonomous driving vehicle;

[0135] A traffic strategy is generated based on the estimated arrival time matrix, and a traffic decision is made for the autonomous driving vehicles at the conflicting intersection.

[0136] Specifically, if the conflict area is identified as a common intersection type, that is, there is only one conflict intersection in the area, the system will first identify all autonomous vehicles heading to the conflict intersection; and obtain the basic information of these autonomous vehicles, such as real-time location coordinates, driving speed and driving direction, through the GPS module or V2X communication system of the autonomous vehicle. Then, based on factors such as the current speed of the autonomous vehicle, the distance to the intersection, and the road conditions, the estimated time for each autonomous vehicle to reach the single conflict intersection is calculated.

[0137] In a preferred embodiment of the present invention, an ordinary intersection may be a standard cross-shaped intersection with a traffic light, and there is only one intersection within the range. A complex intersection refers to an intersection without a traffic light, and the intersection angles exist in various states, and multiple intersections appear within a relatively short distance (for example, more than 2 intersections appear within a distance of 30m). The final sorting result is calculated as follows: a queue is formed at the intersection X1, and three vehicles A0 / A1 / B1 participate in the sorting. The order of A0 / A1 / B1 can be determined by comparing the time it takes to reach the corresponding row of X1; a queue is formed at the intersection X2, and A1 / C1 participate in the sorting. Similarly, the order of A1 / C1 can be determined by comparing the time it takes to reach the corresponding row of X2.

[0138] Combination Figure 6 As shown in the figure, the central circle in the figure represents an intersection, that is, four roads, A, B, C, and D, converge here; A1, A2, B1, and C1 represent autonomous vehicles on each road; there are autonomous vehicles (A1, A2, B1, and C1) approaching the intersection on each road; Ti represents time, that is, the theoretical time for roads A / B / C / D to reach position S0 at the corresponding position at normal speed; Ts is the monitoring time distance, Tc is the control time distance, and T1, T2, and T3 are time equipotential points, indicating the time to reach the intersection calculated based on the current speed; the arrow indicates the driving direction of the autonomous vehicle, and all autonomous vehicles are driving towards the intersection. The system collects the real-time position coordinates, driving speed, and driving direction of each vehicle through the GPS module or V2X communication system of the autonomous vehicle.

[0139] For each vehicle, the time it takes to reach the intersection is calculated using the time formula:

[0140]

[0141] in, is the time it takes for the i-th vehicle on road R to reach the intersection area, is the distance that the i-th vehicle on road R reaches the intersection area, is the corresponding speed. For a single intersection conflict scenario, it is easy to get the corresponding arrival time, and the queue will be sorted according to the size of the arrival time, so that they can pass through the intersection in sequence. For example, if the autonomous driving vehicle A1 is on road A, 100 meters away from the intersection, and the speed is 20 meters / second, then its estimated arrival time is 5 seconds. An estimated arrival time matrix is ​​generated for each vehicle to record the time they arrive at the intersection; based on the estimated arrival time matrix, a traffic strategy is generated. For example, if the autonomous driving vehicles A1 and B1 are expected to arrive at the same time, the autonomous driving vehicles A1 and B1 perform acceleration and deceleration actions according to the received instructions, such as letting A1 go first and B1 pass later. V2X communication sends the traffic strategy to each vehicle, guiding them to pass through the intersection at the scheduled time to ensure smooth traffic. Through precise calculation and scheduling, the waiting time of autonomous driving vehicles at the intersection is reduced, and congestion is reduced; it ensures that autonomous driving vehicles pass through the intersection in the optimal order, and improves the traffic efficiency of traffic flow.

[0142] In this optional embodiment, by generating an estimated arrival time matrix for ordinary intersection types, accurate arrival time information of autonomous vehicles can be provided to the traffic management system; the system can effectively manage and schedule traffic flow at a single intersection, reduce the waiting time of autonomous vehicles, and improve the traffic efficiency of intersections. For conflict areas of ordinary intersection types, the traffic efficiency and safety of traffic flow can be significantly improved. By accurately calculating the arrival time of autonomous vehicles and generating traffic strategies, the system can ensure that autonomous vehicles pass through intersections in the optimal order, reducing traffic congestion and conflicts. In addition, according to the traffic strategy, autonomous vehicles can pass through intersections more smoothly, reduce unnecessary acceleration and deceleration, and help reduce energy consumption and environmental pollution.

[0143] At ordinary crossroads, the present invention transmits the information of autonomous driving vehicles near the intersection to the central control platform through V2X technology, and then the central control platform sorts the autonomous driving vehicles in the system according to the time when each vehicle in the system arrives at the intersection to form a virtual queue, and controls the speed of each vehicle based on the idea of ​​the queue to ensure that each vehicle does not collide with the vehicle in front, and ensures that the autonomous driving vehicles at the intersection pass in sequence to ensure safety. In addition, when there are two or more intersections in a system, it is considered a complex intersection. At this time, a vehicle may pass through two intersections. Therefore, according to the above sorting rules, the expected arrival time of the first arrival point can be obtained, and then the time of the second arrival point is updated according to the distance between the first arrival point and the second arrival point, and then the queue of the second arrival point is sorted. The present invention fully considers the mutual influence relationship in the scenario where there are two adjacent intersections in the system, so that the sorting result is more in line with the actual situation and has practical executability.

[0144] In summary, the overall process of the traffic flow collaborative decision-making optimization method can be as follows: the system first identifies the conflicting intersections X1 and X2 in the traffic network, and determines that the area is a complex intersection type because there are multiple conflicting intersections. Then determine all the autonomous driving vehicles (A0, A1, B1, C1) heading to each conflicting intersection in the crossroads, and calculate the arrival time of each autonomous driving vehicle at the corresponding conflicting intersection; according to the arrival time of the autonomous driving vehicles, generate an estimated arrival time matrix containing all conflicting intersections. For example, A0 is expected to arrive at X1 at time T1, B1 is expected to arrive at X1 at time T2, A1 is expected to arrive at X2 at time T3, and C1 is expected to arrive at X2 at time T4. Sort the arrival times of all autonomous driving vehicles to obtain the arrival order matrix of the conflicting intersections. For example, A0 may be scheduled to pass X1 before B1, and A1 may pass X2 before C1. Then obtain the first arrival order autonomous driving vehicle corresponding to each conflicting intersection in the arrival order matrix, and start from the first arrival order autonomous driving vehicle, update the estimated arrival time matrix of each autonomous driving vehicle. For example, if A0 is the first arrival autonomous driving vehicle of X1, the system will adjust the arrival time of B1 according to the passing time of A0 to ensure that there is enough safety distance between the two vehicles. Finally, according to the updated estimated arrival time matrix, the system makes traffic decisions for each autonomous driving vehicle at each conflicting intersection. Correspondingly, the autonomous driving vehicles are equipped with corresponding communication equipment to send speed adjustment instructions to the autonomous driving vehicles. Through the existing mature V2X technology, each vehicle transmits its own vehicle status to the central control platform in real time, and then completes the scheduling algorithm in the central control platform, and sends it to each vehicle through V2X communication. Each vehicle performs acceleration and deceleration actions according to the received instructions; to ensure smooth and safe traffic flow.

[0145] The present invention reduces the accumulation of autonomous driving vehicles at intersections by precisely controlling the time when autonomous driving vehicles arrive at X1 and X2, thereby alleviating traffic congestion; and by ensuring that a safe interval is maintained between autonomous driving vehicles, the risk of collision accidents is reduced. Autonomous driving vehicles pass through X1 and X2 according to the optimized order, reducing waiting time and improving the traffic efficiency of intersections. In summary, the present invention can take preventive measures against potential bottlenecks in the traffic network by accurately identifying conflict areas and intersections; the determination of complex intersection types enables the method to adapt to traffic scenes of different scales, further improving the applicability of the method, and through the estimated arrival time matrix and arrival order matrix, it can effectively predict and manage the order of autonomous driving vehicles arriving at intersections, reducing congestion and conflict caused by disorderly competition, and improving the coordination of traffic flow. At the same time, the update of arrival time ensures that autonomous driving vehicles pass through intersections in an orderly manner under the premise of complying with safe intervals, reducing emergency braking and sudden acceleration, thereby improving driving safety. Finally, according to the autonomous driving vehicle sorting decision, traffic guidance measures can be implemented according to the real-time updated arrival time matrix, effectively alleviating traffic congestion and improving the traffic capacity of intersections.

[0146] V2X (Vehicle-to-Everything) communication technology plays a vital role in the traffic flow collaborative decision-making optimization method of autonomous vehicles. The communication equipment installed on each autonomous vehicle can collect the vehicle's status information in real time, including location coordinates, driving speed, driving direction, and vehicle sensor data. Through V2X technology, this status information is transmitted to the central control platform in real time, and the traffic management system monitors the dynamics of each vehicle in real time. After receiving the status information of all vehicles, the central control platform runs the scheduling algorithm to calculate the estimated time for each vehicle to arrive at the intersection and generate an estimated arrival time matrix. Based on the results of the scheduling algorithm, the central control platform generates traffic strategies, such as vehicle speed adjustment or other traffic guidance measures; traffic strategies are sent to each autonomous vehicle through V2X communication technology. Each autonomous vehicle performs corresponding acceleration, deceleration or parking actions according to the received instructions to ensure smooth and safe traffic flow. V2X technology ensures the real-time transmission of information, allowing the traffic management system to respond to changes in traffic conditions in a timely manner. By accurately collecting and processing the status information of each vehicle, the scheduling algorithm can more accurately predict traffic flow, thereby improving the accuracy of traffic decisions. Through precise traffic scheduling, the waiting time of vehicles at intersections is reduced, and the traffic flow efficiency is improved. Autonomous vehicles act according to the instructions of the central control platform, which helps to avoid traffic accidents and enhance driving safety. By optimizing the driving speed and route of the vehicle, unnecessary acceleration and deceleration can be reduced, thereby reducing energy consumption. While reducing energy consumption, it also helps to reduce vehicle emissions and has a positive impact on environmental protection.

[0147] Combination Figure 7 As shown, the present invention provides a traffic flow collaborative decision-making optimization system, comprising:

[0148] An intersection acquisition unit, used for acquiring conflicting intersections of a traffic network in a conflicting area;

[0149] a type identification unit, configured to determine a plurality of intersecting roads according to the conflicting intersection, and to judge the conflict type of the conflicting area according to the number of the conflicting intersections in the conflicting area, wherein when there are a plurality of the conflicting intersections in the conflicting area, the conflicting area is determined to be a complex intersection type;

[0150] a calculation unit, configured to, when the conflict area is of the complex intersection type, determine all autonomous driving vehicles in the crossroad heading for each of the conflict intersections, and respectively calculate an arrival time of each of the autonomous driving vehicles at the corresponding conflict intersection;

[0151] A time matrix construction unit, configured to generate an estimated arrival time matrix according to the arrival time of the autonomous driving vehicle, wherein the estimated arrival time matrix includes the arrival time of each autonomous driving vehicle at the corresponding conflicting intersection;

[0152] An order matrix construction unit is used to obtain an arrival order matrix according to the order in which all the autonomous driving vehicles pass through the corresponding conflicting intersections;

[0153] an updating unit, configured to determine a target conflict intersection according to the arrival order matrix, and update the estimated arrival time matrix based on the arrival time corresponding to the autonomous driving vehicle that first arrives at the target conflict intersection, to obtain the updated estimated arrival time matrix, wherein when the conflict intersection in the driving routes of all the autonomous driving vehicles passing through the conflict intersection is always the first conflict intersection arrived by the corresponding autonomous driving vehicle, the conflict intersection is determined to be the target conflict intersection;

[0154] A decision-making unit is used to make traffic decisions for each autonomous driving vehicle at each of the conflicting intersections based on the updated estimated arrival time matrix.

[0155] The traffic flow collaborative decision optimization system of the present invention accurately locates the traffic conflict area by identifying the conflict intersections in the traffic network and determining the crossroads. Since multiple conflict intersections influence each other, the conflict area can be classified by the number of conflict intersections. When multiple conflict intersections are identified, the area is determined to be a complex intersection type. At this time, complex processing logic is required to deal with the interaction problem of multiple intersections. Specifically, by calculating the estimated time for each vehicle to arrive at each conflict intersection, a comprehensive estimated arrival time matrix is ​​generated, which provides a data basis for subsequent traffic sorting and decision-making. Then, by sorting all autonomous driving vehicles, an arrival order matrix is ​​obtained, and the order in which the autonomous driving vehicles pass through the corresponding conflict intersections is determined, thereby reducing the congestion at the intersection. In addition, the mutual influence of multiple intersection areas at complex intersections is comprehensively considered, avoiding the multi-vehicle conflict blocking and locking phenomenon caused by the queuing method of ordinary intersections. By updating the estimated arrival time matrix in real time, the smooth passage of vehicles at complex intersections is ensured, and the traffic efficiency of traffic flow is improved. In addition, the present invention is also practical and can adapt to different traffic scenarios, such as the situation where multiple intersections appear within a short distance. Finally, according to the updated estimated arrival time matrix, traffic decisions are made for the autonomous driving vehicles at each conflicting intersection, such as executing acceleration and deceleration actions, ensuring the smoothness and efficiency of traffic flow and avoiding the problem of low traffic efficiency caused by queuing and blocking. The present invention improves the intelligent level of traffic management and ensures the efficiency and safety of traffic flow through precise time calculation, autonomous driving vehicle sorting and dynamic time update.

[0156] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A traffic flow collaborative decision-making optimization method, characterized in that: include: Obtain conflicting intersections of the traffic network in the conflicting area; Determine a plurality of intersecting roads according to the conflicting intersections, and judge the conflict type of the conflicting area according to the number of the conflicting intersections in the conflicting area, wherein when there are a plurality of the conflicting intersections in the conflicting area, the conflicting area is judged to be a complex intersection type; When the conflict area is of the complex intersection type, determining all autonomous driving vehicles on the crossroad heading for each of the conflict intersections, and calculating the arrival time of each of the autonomous driving vehicles at the corresponding conflict intersection; Generate an estimated arrival time matrix according to the arrival time of the autonomous driving vehicle, wherein the estimated arrival time matrix includes the arrival time of each autonomous driving vehicle at the corresponding conflicting intersection; Obtaining an arrival order matrix according to the order in which all the autonomous driving vehicles pass through the corresponding conflicting intersections; Determine a target conflict intersection according to the arrival order matrix, and update the estimated arrival time matrix based on the arrival time corresponding to the autonomous driving vehicle that first arrives at the target conflict intersection to obtain an updated estimated arrival time matrix, wherein when the conflict intersection in the driving routes of all the autonomous driving vehicles passing through the conflict intersection is always the first conflict intersection arrived by the corresponding autonomous driving vehicle, the conflict intersection is determined to be the target conflict intersection; According to the updated estimated arrival time matrix, traffic decisions are made for each autonomous driving vehicle at each conflicting intersection.

2. The traffic flow collaborative decision-making optimization method according to claim 1 is characterized in that: The determining of a plurality of intersecting roads according to the conflicting intersection comprises: Taking the conflicting intersection as an intersection, obtaining coordinate information of the intersection in the conflicting area; Determine a plurality of roads passing through the intersection according to the coordinate information of the intersection and in combination with the road topology of the traffic network in the conflict area; Each of the roads passing through the intersection is regarded as the intersection road.

3. The traffic flow collaborative decision-making optimization method according to claim 1 is characterized in that: The determining of all autonomous driving vehicles on the crossroad heading for each of the conflicting intersections, and respectively calculating the arrival time of each of the autonomous driving vehicles at the corresponding conflicting intersection, includes: Obtaining basic information of each autonomous driving vehicle on the intersection road, the basic information including the real-time position coordinates, driving speed, driving direction, surrounding road conditions, and traffic signal status of the autonomous driving vehicle; Determine, according to the driving direction of the autonomous driving vehicle, all the autonomous driving vehicles on the crossroad heading towards the conflicting intersection; Calculate the driving distance of each autonomous driving vehicle to the conflicting intersection according to the real-time position coordinates of each autonomous driving vehicle; The arrival time of each of the autonomous driving vehicles heading for the conflicting intersection at the corresponding conflicting intersection is determined based on the driving speed, the driving distance, the surrounding road conditions, and the traffic signal status of each of the autonomous driving vehicles.

4. The traffic flow collaborative decision-making optimization method according to claim 1 is characterized in that: The generating an estimated arrival time matrix according to the arrival time of the autonomous driving vehicle comprises: For one of the autonomous driving vehicles, generating a time array of the autonomous driving vehicle arriving at each of the conflicting intersections according to the arrival time of the autonomous driving vehicle at each of the conflicting intersections; The estimated arrival time matrix is ​​formed according to the time arrays corresponding to all the autonomous driving vehicles.

5. The traffic flow collaborative decision-making optimization method according to claim 1 is characterized in that: The arrival order matrix is ​​obtained according to the order in which all the autonomous driving vehicles pass through the corresponding conflicting intersections, including: For one of the autonomous driving vehicles, determining the order in which the autonomous driving vehicle passes through each of the conflicting intersections according to the real-time position coordinates and driving direction of the autonomous driving vehicle; According to the passing order corresponding to the autonomous driving vehicle, the conflicting intersections that the autonomous driving vehicle will pass through are sorted to obtain an intersection passing array corresponding to the autonomous driving vehicle; The arrival order matrix is ​​obtained based on the intersection passing array of all the autonomous driving vehicles.

6. The traffic flow collaborative decision-making optimization method according to claim 5 is characterized in that: The updating of the estimated arrival time matrix based on the arrival time corresponding to the autonomous driving vehicle that first arrives at the target conflict intersection to obtain the updated estimated arrival time matrix includes: Determining whether there are multiple target autonomous driving vehicles passing through the target conflict intersection according to the arrival order matrix, wherein the target driving vehicle is an autonomous driving vehicle whose target conflict intersection is the first conflict intersection in its driving route; If so, all the target autonomous driving vehicles are sorted according to the arrival time of each target autonomous driving vehicle at the target conflict intersection, and the arrival time of other target autonomous driving vehicles at the target conflict intersection is updated according to the arrival time corresponding to the target autonomous driving vehicle that arrives at the target conflict intersection first, the preset minimum time interval and the sorting result; According to the updated arrival time of all the target autonomous driving vehicles at the target conflicting intersection, the arrival time of the target autonomous driving vehicles at other conflicting intersections is updated to generate an updated estimated arrival time matrix.

7. The traffic flow collaborative decision-making optimization method according to claim 1 is characterized in that: The step of making a traffic decision for the autonomous driving vehicle at the conflicting intersection according to the updated estimated arrival time matrix includes: According to the updated estimated arrival time matrix, the autonomous driving vehicles at all the conflicting intersections are queued to obtain a final sorting result of the autonomous driving vehicles; A traffic strategy for the conflicting area is generated according to the final sorting result, and a traffic decision is made for the autonomous driving vehicle at the conflicting intersection according to the traffic strategy.

8. The traffic flow collaborative decision-making optimization method according to claim 1 is characterized in that: The conflict type also includes a common intersection type, and judging the conflict type of the conflict area according to the number of the conflict intersections in the conflict area further includes: When there is a conflicting intersection in the conflicting area, it is determined that the conflicting area is a common intersection type.

9. The traffic flow collaborative decision-making optimization method according to claim 8 is characterized in that: Also includes: When the conflict area is of the ordinary intersection type, determining all autonomous driving vehicles on the crossroad heading toward the conflict intersection, and calculating the arrival time of each autonomous driving vehicle at the conflict intersection; Generate an estimated arrival time matrix corresponding to the conflicting intersection according to the arrival time of the autonomous driving vehicle; A traffic strategy is generated based on the estimated arrival time matrix, and a traffic decision is made for the autonomous driving vehicles at the conflicting intersection.

10. A traffic flow collaborative decision-making optimization system, characterized in that: include: An intersection acquisition unit, used for acquiring conflicting intersections of a traffic network in a conflicting area; a type identification unit, configured to determine a plurality of intersecting roads according to the conflicting intersection, and to judge the conflict type of the conflicting area according to the number of the conflicting intersections in the conflicting area, wherein when there are a plurality of the conflicting intersections in the conflicting area, the conflicting area is determined to be a complex intersection type; a calculation unit, configured to, when the conflict area is of the complex intersection type, determine all autonomous driving vehicles in the crossroad heading for each of the conflict intersections, and respectively calculate an arrival time of each of the autonomous driving vehicles at the corresponding conflict intersection; A time matrix construction unit, configured to generate an estimated arrival time matrix according to the arrival time of the autonomous driving vehicle, wherein the estimated arrival time matrix includes the arrival time of each autonomous driving vehicle at the corresponding conflicting intersection; An order matrix construction unit is used to obtain an arrival order matrix according to the order in which all the autonomous driving vehicles pass through the corresponding conflicting intersections; an updating unit, configured to determine a target conflict intersection according to the arrival order matrix, and update the estimated arrival time matrix based on the arrival time corresponding to the autonomous driving vehicle that first arrives at the target conflict intersection, to obtain the updated estimated arrival time matrix, wherein when the conflict intersection in the driving routes of all the autonomous driving vehicles passing through the conflict intersection is always the first conflict intersection arrived by the corresponding autonomous driving vehicle, the conflict intersection is determined to be the target conflict intersection; A decision-making unit is used to make traffic decisions for each autonomous driving vehicle at each of the conflicting intersections based on the updated estimated arrival time matrix.

Images