Vehicle assisted passage method at traffic intersections based on vehicle AI platform

Abstract

The present invention relates to the technical field of intelligent transportation, and specifically to a vehicle-assisted traffic method at a traffic intersection based on a vehicle AI platform, the method comprising: analyzing the accumulated vehicle information corresponding to a first road section and a second road section at the current moment, and obtaining accumulated parameters of traffic flows such as right turns, straight ahead, left turns, and outflows, wherein the first road section and the second road section are adjacent and have opposite driving directions, and when a tidal lane is opened, the inflow width of the first road section becomes narrower; analyzing the vehicle response information corresponding to the first road section and the second road section at the current moment, and obtaining hysteresis parameters of traffic flows such as right turns, straight ahead, left turns, and outflows; obtaining hysteresis parameters of each traffic flow based on the accumulated parameters and hysteresis parameters of traffic flows such as right turns, straight ahead, left turns, and outflows; analyzing the difference in road congestion between the first road section and the second road section based on the hysteresis parameters of each traffic flow, and generating tidal lane control information. The present invention can improve the control effect of tidal lanes.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent transportation, and in particular to a method for assisting vehicle passage at traffic intersections based on a vehicle AI platform. Background Art

[0002] Tidal lanes (also known as variable lanes) are a dynamic traffic management measure that optimizes road resource allocation by dynamically adjusting the driving direction of lanes based on the directional differences in traffic flow at different times (such as morning and evening rush hours).

[0003] At present, tidal lanes mainly rely on mechanized lane changes during preset time periods, and are unable to perceive the dynamic game of vehicles in all directions at the intersection in real time, so they are unable to effectively respond to sudden traffic flow fluctuations. Specifically, the above-mentioned control method for tidal lanes is too dependent on time, and it is easy for the tidal lanes to be rigidly opened when the current road congestion is relatively serious, thereby further reducing the intersection width of the current road, and further aggravating the congestion of the current road; or, it is easy for the tidal lanes to be rigidly kept open when the congestion of the widened lanes has been effectively alleviated, resulting in the lane resources of the tidal lanes not being fully utilized.

[0004] In other words, the existing tidal lane control scheme has poor control effect. Summary of the Invention

[0005] In order to solve the technical problem of poor control effect of existing tidal lane control solutions, the purpose of the present invention is to provide a vehicle assisted traffic method at a traffic intersection based on a vehicle AI platform. The technical solution adopted is as follows:

[0006] In a first aspect, an embodiment of the present invention provides a method for assisting vehicle traffic at a traffic intersection based on a vehicle AI platform, the method comprising:

[0007] Analyze vehicle accumulation information corresponding to the first road section and the second road section at the current moment to obtain a right-turn accumulation parameter corresponding to the right-turning traffic flow to be merged, a straight-moving accumulation parameter corresponding to the straight-moving traffic flow to be merged, a left-turn accumulation parameter corresponding to the left-turning traffic flow to be merged, and an outflow accumulation parameter corresponding to the outflowing traffic flow, wherein the first road section and the second road section are adjacent, the first road section is a road section where the traffic flow merges in, and the second road section is a road section where the traffic flow merges out, and when the tidal lane is opened, the merging width of the first road section becomes narrower;

[0008] Analyze the vehicle response information corresponding to the first road section and the second road section at the current moment to obtain a right turn hysteresis parameter, a straight hysteresis parameter, a left turn hysteresis parameter, and an export hysteresis parameter;

[0009] A right-turn congestion parameter is obtained according to the right-turn accumulation parameter and the right-turn hysteresis parameter, a straight-through congestion parameter is obtained according to the straight-through accumulation parameter and the straight-through hysteresis parameter, a left-turn congestion parameter is obtained according to the left-turn accumulation parameter and the left-turn hysteresis parameter, and an export congestion parameter is obtained according to the export accumulation parameter and the export hysteresis parameter;

[0010] According to the right-turn congestion parameter, the straight-through congestion parameter, the left-turn congestion parameter, and the export congestion parameter, the congestion difference between the first road section and the second road section is analyzed to generate tidal lane control information.

[0011] In one embodiment, the analyzing of the vehicle accumulation information corresponding to the first road segment and the second road segment at the current moment to obtain the right-turn accumulation parameter corresponding to the right-turning traffic flow to be merged, the straight-line accumulation parameter corresponding to the straight-line traffic flow to be merged, the left-turn accumulation parameter corresponding to the left-turning traffic flow to be merged, and the outgoing accumulation parameter corresponding to the outgoing traffic flow include:

[0012] According to the vehicle accumulation information corresponding to the target traffic flow at the current moment, the maximum number of vehicles and the minimum number of vehicles corresponding to the target traffic flow are obtained, wherein the maximum number of vehicles is the maximum number of vehicles accumulated in the target traffic flow during the previous traffic light alternation cycle at the current moment, and the minimum number of vehicles is the minimum number of vehicles accumulated in the previous traffic light alternation cycle at the current moment, and the target traffic flow is a right-turning traffic flow to be merged, a straight-going traffic flow to be merged, a left-turning traffic flow to be merged, or a traffic flow to be merged out;

[0013] Calculate the sum of the maximum number of vehicles and the minimum number of vehicles corresponding to the target traffic flow to obtain a reference value of the number of vehicles in the target traffic flow;

[0014] The ratio of the reference value of the number of vehicles of the target traffic flow to the target cycle value is calculated to obtain the accumulation parameter corresponding to the target traffic flow, wherein the target cycle value is the time value indicated by the traffic light alternation cycle.

[0015] In one embodiment, analyzing the vehicle response information corresponding to the first road segment and the second road segment at the current moment to obtain the right turn hysteresis parameter, the straight hysteresis parameter, the left turn hysteresis parameter, and the export hysteresis parameter includes:

[0016] Obtaining, based on vehicle response information corresponding to the target traffic flow at the current moment, a first parameter, a second parameter, and a third parameter for each vehicle in the target traffic flow, wherein the first parameter is used to indicate the acceleration performance of the corresponding vehicle, the second parameter is used to indicate the reaction delay of the driver of the corresponding vehicle, and the third parameter is used to indicate the distance between the corresponding vehicle and the intersection, where the target traffic flow is a right-turning traffic flow to be merged, a straight-moving traffic flow to be merged, a left-turning traffic flow to be merged, or a traffic flow to be merged out;

[0017] Obtaining a hysteresis factor for each vehicle in the target traffic flow according to the first parameter, the second parameter, and the third parameter of each vehicle in the target traffic flow;

[0018] The average of multiple hysteresis factors of multiple vehicles in the target traffic flow is calculated to obtain a hysteresis parameter corresponding to the target traffic flow.

[0019] In one embodiment, the step of obtaining the first parameter of each vehicle in the target traffic flow includes:

[0020] Analyze multiple movement processes of each vehicle in the target traffic flow within a first time period to obtain multiple starting accelerations corresponding to each vehicle in the target traffic flow, wherein the multiple starting accelerations correspond one-to-one to the multiple movement processes, the starting acceleration is the instantaneous acceleration of the corresponding vehicle at the start of the corresponding movement process, the duration of the first time period is a first preset duration, and the end time of the first time period is the current time;

[0021] An average of a plurality of initial accelerations corresponding to each vehicle in the target traffic flow is calculated to obtain a first parameter of each vehicle in the target traffic flow.

[0022] In one embodiment, the step of obtaining the second parameter of each vehicle in the target traffic flow includes:

[0023] Analyze multiple movement processes of each vehicle in the target traffic flow within a first time period to obtain multiple start data corresponding to each vehicle in the target traffic flow, wherein the multiple start data correspond to the multiple movement processes in a one-to-one manner, and the start data each include a self-start time of the corresponding vehicle and a start time of a preceding vehicle, wherein the self-start time is the starting movement time of the corresponding vehicle in the corresponding movement process, and the preceding vehicle start time is the starting movement time of a vehicle preceding the corresponding vehicle in the corresponding movement process;

[0024] Obtaining, based on a plurality of start data corresponding to each vehicle in the target traffic flow, a plurality of start time differences corresponding to each vehicle in the target traffic flow, wherein the start time difference is the difference between the self-start time in the corresponding start data and the start time of the preceding vehicle;

[0025] The average of multiple starting time differences corresponding to each vehicle in the target traffic flow is calculated to obtain the second parameter of each vehicle in the target traffic flow.

[0026] In one embodiment, based on the right-turn congestion parameter, the straight-through congestion parameter, the left-turn congestion parameter, and the export congestion parameter, the difference in congestion between the first road segment and the second road segment is analyzed to generate tidal lane control information, including:

[0027] Obtain multiple straight-through efficiency values ​​of the straight-through green light period of the traffic flow waiting to merge into in the second time period, multiple left-turn efficiency values ​​of the left-turn green light period of the traffic flow waiting to merge into in the second time period, multiple straight-through green light accumulation parameters of the right-turn traffic flow waiting to merge into in the second time period, and multiple left-turn green light accumulation parameters of the right-turn traffic flow waiting to merge into in the second time period, wherein the straight-through efficiency value is the ratio of the number of vehicles passing through the corresponding straight-through green light period of the traffic flow waiting to merge into and the duration of the corresponding straight-through green light period, and the left-turn efficiency value is the ratio of the number of vehicles passing through the corresponding left-turn green light period of the traffic flow waiting to merge into and the duration of the corresponding left-turn green light period, the duration of the second time period is a second preset duration, and the end time of the second time period is the current time;

[0028] performing a correlation analysis on the plurality of straight-through traffic efficiency values ​​and the plurality of straight-through green light accumulation parameters to obtain a straight-right correlation value, and performing a correlation analysis on the plurality of left-turn traffic efficiency values ​​and the plurality of left-turn green light accumulation parameters to obtain a left-right correlation value;

[0029] According to the straight-right correlation value and the left-right correlation value, the right-turn congestion parameter, the straight-through congestion parameter, and the left-turn congestion parameter are respectively corrected to obtain a right-turn correction parameter, a straight-through correction parameter, and a left-turn correction parameter;

[0030] According to the right-turn correction parameter, the straight-ahead correction parameter, the left-turn correction parameter, and the exported congestion parameter, a congestion difference between the first road section and the second road section is analyzed to generate tidal lane control information.

[0031] In one embodiment, the right-turn congestion parameter, the straight-through congestion parameter, and the left-turn congestion parameter are respectively corrected according to the straight-right correlation value and the left-right correlation value to obtain the right-turn correction parameter, the straight-through correction parameter, and the left-turn correction parameter, including:

[0032] According to the straight-right correlation value and the left-right correlation value, a straight-moving correction value, a left-turn correction value, and a right-turn correction value are obtained, wherein the sum of the straight-right correlation value and the straight-moving correction value is 1, the sum of the left-right correlation value and the left-turn correction value is 1, and the right-turn correction value is the sum of the straight-right correlation value and the left-right correlation value;

[0033] The product of the right-turn congestion parameter and the right-turn correction value is determined as the right-turn correction parameter, the product of the left-turn congestion parameter and the left-turn correction value is determined as the left-turn correction parameter, and the product of the straight-line congestion parameter and the straight-line correction value is determined as the straight-line correction parameter.

[0034] In one embodiment, the step of analyzing the congestion difference between the first road segment and the second road segment based on the right-turn correction parameter, the straight-ahead correction parameter, the left-turn correction parameter, and the exported congestion parameter to generate the tidal lane control information includes:

[0035] When the first road section and the second road section match the congestion condition, calculating the average of the right-turn correction parameter, the straight-going correction parameter, and the left-turn correction parameter to obtain a target congestion parameter;

[0036] generating tidal lane control information instructing to close the tidal lane when the target congestion parameter is greater than the outgoing congestion parameter;

[0037] When the target congestion parameter is less than the outgoing congestion parameter, tidal lane control information indicating opening the tidal lane is generated.

[0038] In one embodiment, before calculating the average of the right-turn correction parameter, the straight-ahead correction parameter, and the left-turn correction parameter to obtain the target congestion parameter, the method further includes:

[0039] Obtain a right-turn curve corresponding to the right-turn correction parameter, a straight-line curve corresponding to the straight-line correction parameter, a left-turn curve corresponding to the left-turn correction parameter, and an outflow curve corresponding to the outflow congestion parameter, wherein the right-turn curve is a curve showing a change in the congestion level of the right-turning traffic to be merging in a third time period, the right-turn correction parameter indicates the congestion level of the right-turning traffic to be merging in at a current moment, the straight-line curve is a curve showing a change in the congestion level of the straight-line traffic to be merging in a third time period, the straight-line correction parameter indicates the congestion level of the straight-line traffic to be merging in at a current moment, the left-turn curve is a curve showing a change in the congestion level of the left-turning traffic to be merging in a third time period, the left-turn correction parameter indicates the congestion level of the left-turning traffic to be merging in at a current moment, the outflow curve is a curve showing a change in the congestion level of the outflow traffic to be merged in a third time period, the outflow congestion parameter indicates the congestion level of the outflow traffic to be merged in at a current moment, the duration of the third time period is a third preset duration, and the end time of the third time period is the current moment;

[0040] Obtain multiple right-turn curve slopes at multiple sampling moments within a third time period based on the right-turn curve, obtain multiple straight-line curve slopes at multiple sampling moments based on the straight-line curve, obtain multiple left-turn curve slopes at multiple sampling moments based on the left-turn curve, and obtain multiple export curve slopes at multiple sampling moments based on the export curve;

[0041] The average value of the slopes of the plurality of right-turn curves is determined as the right-turn slope value, the average value of the slopes of the plurality of straight-line curves is determined as the straight-line slope value, the average value of the slopes of the plurality of left-turn curves is determined as the left-turn slope value, and the average value of the slopes of the plurality of outgoing curves is determined as the outgoing slope value;

[0042] The product of the right-turn slope value and the right-turn correction parameter is determined as the right-turn congestion value, the product of the straight-line slope value and the straight-line correction parameter is determined as the straight-line congestion value, the product of the left-turn slope value and the left-turn correction parameter is determined as the left-turn congestion value, and the product of the export slope value and the export congestion parameter is determined as the export congestion value;

[0043] The right-turn congestion value, the straight-through congestion value, the left-turn congestion value, and the export congestion value are analyzed to determine whether the first road segment and the second road segment match the congestion condition.

[0044] In one embodiment, analyzing the right-turn congestion value, the straight-through congestion value, the left-turn congestion value, and the export congestion value to determine whether the first road segment and the second road segment match the congestion condition includes:

[0045] Calculate the average of the right-turn congestion value, the straight-through congestion value, the left-turn congestion value, and the export congestion value to obtain the congestion index value;

[0046] If the congestion index value is greater than or equal to a congestion threshold, determining that the first road section and the second road section match a congestion condition;

[0047] When the congestion index value is less than a congestion threshold, it is determined that the first road section and the second road section do not match a congestion condition.

[0048] In a second aspect, another embodiment of the present invention provides a vehicle assisted traffic system at a traffic intersection based on a vehicle AI platform, the system comprising:

[0049] A first analysis module is configured to analyze vehicle accumulation information corresponding to a first road section and a second road section at a current moment, and obtain right-turn accumulation parameters corresponding to a right-turning traffic flow to be merged into, straight-moving accumulation parameters corresponding to a straight-moving traffic flow to be merged into, left-turn accumulation parameters corresponding to a left-turning traffic flow to be merged into, and outgoing accumulation parameters corresponding to an outgoing traffic flow to be merged into, wherein the first road section and the second road section are adjacent, the first road section is a road section where traffic flows merge into, the second road section is a road section where traffic flows merge out, and when a tidal lane is enabled, the merging width of the first road section becomes narrower;

[0050] The second analysis module is used to analyze the vehicle response information corresponding to the first road section and the second road section at the current moment to obtain a right turn hysteresis parameter, a straight hysteresis parameter, a left turn hysteresis parameter and an export hysteresis parameter;

[0051] a parameter calculation module, configured to obtain a right-turn congestion parameter based on a right-turn accumulation parameter and a right-turn hysteresis parameter, obtain a straight-through congestion parameter based on a straight-through accumulation parameter and a straight-through hysteresis parameter, obtain a left-turn congestion parameter based on a left-turn accumulation parameter and a left-turn hysteresis parameter, and obtain an export congestion parameter based on an export accumulation parameter and an export hysteresis parameter;

[0052] The congestion analysis module is used to analyze the congestion difference between the first road section and the second road section according to the right-turn congestion parameter, the straight-through congestion parameter, the left-turn congestion parameter and the exported congestion parameter, and generate tidal lane control information.

[0053] In a third aspect, another embodiment of the present invention further provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the method described in the first aspect when executed by the processor.

[0054] In a fourth aspect, another embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described in the first aspect are implemented.

[0055] The present invention has the following beneficial effects:

[0056] The present invention first analyzes the vehicle accumulation information corresponding to the first road section and the second road section at the current moment to obtain the right turn accumulation parameter corresponding to the right-turn traffic flow to be merged, the straight-ahead accumulation parameter corresponding to the straight-ahead traffic flow to be merged, the left turn accumulation parameter corresponding to the left-turn traffic flow to be merged, and the outflow accumulation parameter corresponding to the outflow traffic flow, that is, to determine the accumulation situation of the traffic flow to be merged into the first road section by turning right, going straight, and turning left, respectively, and to determine the accumulation situation of the traffic flow to be outflowed from the second road section, and then further analyzes the vehicle response information corresponding to the first road section and the second road section at the current moment to obtain the right turn hysteresis parameter, the straight-ahead hysteresis parameter, the left turn hysteresis parameter, and the outflow hysteresis parameter, that is, to determine the traffic flow to be merged through The flow hysteresis of the traffic flow turning right, going straight and turning left into the first section respectively, as well as the flow hysteresis of the traffic flow to be discharged from the second section, are respectively calculated. Then, the corresponding congestion parameters are obtained by comprehensively combining the accumulation parameters and hysteresis parameters to accurately evaluate the congestion conditions of the corresponding traffic flow from two aspects: the accumulation conditions and the flow hysteresis conditions. Finally, the difference in section congestion between the first section and the second section is analyzed, that is, the difference in congestion conditions between the incoming traffic flow and the outgoing traffic flow is analyzed, so as to flexibly complete the opening and closing of the tidal lane while ensuring the accuracy of control, so as to better adapt to complex lane conditions, improve the control effect of the tidal lane, and make the tidal lane more effectively used. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1 A schematic flow chart of a method for assisting vehicle passage at a traffic intersection based on a vehicle AI platform provided by one embodiment of the present invention;

[0059] Figure 2 A schematic diagram of the structure of a vehicle assisted traffic system at a traffic intersection based on a vehicle AI platform provided by one embodiment of the present invention;

[0060] Figure 3 The present invention provides a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0061] To further illustrate the technical means and effectiveness of the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of a vehicle-assisted traffic intersection method based on a vehicle AI platform proposed by the present invention. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0062] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0063] The following describes in detail a specific solution of a vehicle assisted passage method at a traffic intersection based on a vehicle AI platform provided by the present invention in conjunction with the accompanying drawings.

[0064] This invention proposes a vehicle-assisted traffic method at a traffic intersection based on a vehicle AI platform. Figure 1 , which shows a schematic flow chart of a method for assisting vehicle passage at a traffic intersection based on a vehicle AI platform provided by one embodiment of the present invention, the method comprising the following steps:

[0065] Step S1: Analyze the vehicle accumulation information corresponding to the first and second road sections at the current moment to obtain the right-turn accumulation parameters corresponding to the right-turn traffic flow to be merged, the straight-line accumulation parameters corresponding to the straight-line traffic flow to be merged, the left-turn accumulation parameters corresponding to the left-turn traffic flow to be merged, and the outgoing accumulation parameters corresponding to the outgoing traffic flow.

[0066] The first road section and the second road section are adjacent to each other, the first road section is the section where traffic converges, and the second road section is the section where traffic converges. When the tidal lane is opened, the confluence width of the first road section becomes narrower.

[0067] The first road section and the second road section mentioned above can be understood as any road equipped with a tidal road, wherein the first road section is the road section that becomes narrower after the tidal lane is opened (that is, the road section where the traffic flows in), and the second road section is the road section that becomes wider after the tidal lane is opened (that is, the road section where the traffic flows out). The first road section and the second road section are separated by a guardrail, and the guardrail is a movable guardrail near the intersection. The movable guardrail is powered by mains electricity and is initially located in the middle of the road (that is, The dividing line between the first section and the second section), after receiving the tidal lane opening command, the movable guardrail moves laterally from the initial position to the target position (the distance between the initial position and the target position is the standard lane distance). At this time, the first section becomes narrower and the second section becomes wider, and the tidal lane is used as the left turn lane of the second section; after receiving the tidal lane closing command, the movable guardrail moves laterally from the target position to the initial position. At this time, the first section returns to its normal width and the second section becomes narrower, and the tidal lane is used as the lane of the first section.

[0068] Exemplarily, the movable guardrail is a tidal lane robot.

[0069] It should be understood that the right-turning traffic waiting to merge is the traffic waiting to merge into the first section by turning right; the left-turning traffic waiting to merge is the traffic waiting to merge into the first section by turning left; the straight-going traffic waiting to merge is the traffic waiting to merge into the first section by going straight; and the traffic waiting to merge out is the traffic waiting to exit from the second section (or the traffic waiting to exit from the left-turn lane of the second section).

[0070] In the application, the vehicle accumulation information corresponding to the first and second sections at the current moment (the queue length or number of vehicles to be merged into or out of the vehicle flow at each moment) can be obtained based on the laser radar scanning device and binocular camera installed at the road intersection and on both sides.

[0071] For example, the queue length of each traffic flow can be scanned at a frequency of 10 Hz by using a laser radar scanning device to obtain the queue length of each traffic flow at each moment.

[0072] Alternatively, binocular cameras are used to periodically collect images of queued vehicles in each traffic flow, and vehicle detection is performed on the collected queued vehicle images in combination with deep learning YOLOv7 to determine the number of vehicles included in each traffic flow at each moment, and adjacent queued vehicle images are analyzed to determine the physical parameters of each vehicle at each moment, such as speed, position, and acceleration.

[0073] Among them, the right-turn accumulation parameter is used to indicate the severity of the accumulation of vehicles in the right-turn traffic flow waiting to merge at the current moment (that is, the growth rate of the number of vehicles in the right-turn traffic flow waiting to merge). The larger the right-turn accumulation parameter is, the more serious the accumulation of vehicles in the right-turn traffic flow waiting to merge at the current moment (that is, the faster the growth rate of the number of vehicles in the right-turn traffic flow waiting to merge), and the more a wider intersection is needed for passage. The meanings of the straight accumulation parameter, left turn accumulation parameter, and outgoing accumulation parameter are similar to those of the right-turn accumulation parameter, and will not be repeated here to avoid repetition.

[0074] Furthermore, the analysis of the vehicle accumulation information corresponding to the first road section and the second road section at the current moment to obtain the right turn accumulation parameter corresponding to the right-turning traffic flow to be merged, the straight-going accumulation parameter corresponding to the straight-going traffic flow to be merged, the left turn accumulation parameter corresponding to the left-turning traffic flow to be merged, and the outgoing accumulation parameter corresponding to the outgoing traffic flow, includes:

[0075] According to the vehicle accumulation information corresponding to the target traffic flow at the current moment, the maximum number of vehicles and the minimum number of vehicles corresponding to the target traffic flow are obtained, wherein the maximum number of vehicles is the maximum number of vehicles accumulated in the target traffic flow during the previous traffic light alternation cycle at the current moment, and the minimum number of vehicles is the minimum number of vehicles accumulated in the previous traffic light alternation cycle at the current moment, and the target traffic flow is a right-turning traffic flow to be merged, a straight-going traffic flow to be merged, a left-turning traffic flow to be merged, or a traffic flow to be merged out;

[0076] Calculate the sum of the maximum number of vehicles and the minimum number of vehicles corresponding to the target traffic flow to obtain a reference value of the number of vehicles in the target traffic flow;

[0077] The ratio of the reference value of the number of vehicles of the target traffic flow to the target cycle value is calculated to obtain the accumulation parameter corresponding to the target traffic flow, wherein the target cycle value is the time value indicated by the traffic light alternation cycle.

[0078] When a certain traffic flow waiting to enter or exit is congested, the number of vehicles in the red light waiting period and the number of vehicles in the green light release interval will increase. Therefore, in this process, the accumulation of vehicles in the corresponding traffic flow is dynamically reflected by counting the sum of the maximum number of vehicles and the minimum number of vehicles in the traffic light alternation cycle, and calculating the ratio of this sum to the target cycle value. Among them, selecting the previous traffic light alternation cycle at the current moment as the statistical period can maximally accurately reflect the vehicle detention situation in the corresponding traffic flow at the current moment, making the determined accumulation parameters more accurate and reliable.

[0079] The traffic light alternation cycle should be understood as the time it takes for the lane where the traffic flow is located to completely go through a red light-green light process. For example, the target cycle value can be 100 seconds.

[0080] It should be understood that when the target traffic flow is a right-turning traffic flow waiting to merge, the accumulated parameter corresponding to the target traffic flow is the right-turn accumulated parameter. The other three situations of the target traffic flow can be deduced accordingly. To avoid repetition, they will not be described again.

[0081] Step S2: Analyze the vehicle response information corresponding to the first road section and the second road section at the current moment to obtain a right turn hysteresis parameter, a straight hysteresis parameter, a left turn hysteresis parameter, and an export hysteresis parameter.

[0082] The vehicle response information is used to indicate the performance of incoming and outgoing vehicles in responding to traffic lights, such as when they pass or wait. This information can be expressed through physical parameters such as the speed, position, and acceleration of each vehicle at each moment.

[0083] The right-turn hysteresis parameter is used to reversely indicate the traffic efficiency of vehicles turning right at the current moment in the traffic flow waiting to merge. The larger the right-turn hysteresis parameter, the lower the traffic efficiency of vehicles turning right at the current moment in the traffic flow waiting to merge (specifically, during the green light release, the fewer vehicles pass through per unit time), and the wider the intersection needs to be for passage. The meanings of the straight hysteresis parameter, left turn hysteresis parameter, and merging hysteresis parameter are similar to those of the right-turn hysteresis parameter and are not repeated here to avoid repetition.

[0084] Furthermore, the analyzing the vehicle response information corresponding to the first road section and the second road section at the current moment to obtain the right turn hysteresis parameter, the straight hysteresis parameter, the left turn hysteresis parameter and the export hysteresis parameter includes:

[0085] Obtaining, based on vehicle response information corresponding to the target traffic flow at the current moment, a first parameter, a second parameter, and a third parameter for each vehicle in the target traffic flow, wherein the first parameter is used to indicate the acceleration performance of the corresponding vehicle, the second parameter is used to indicate the reaction delay of the driver of the corresponding vehicle, and the third parameter is used to indicate the distance between the corresponding vehicle and the intersection, where the target traffic flow is a right-turning traffic flow to be merged, a straight-moving traffic flow to be merged, a left-turning traffic flow to be merged, or a traffic flow to be merged out;

[0086] Obtaining a hysteresis factor for each vehicle in the target traffic flow according to the first parameter, the second parameter, and the third parameter of each vehicle in the target traffic flow;

[0087] The average of multiple hysteresis factors of multiple vehicles in the target traffic flow is calculated to obtain a hysteresis parameter corresponding to the target traffic flow.

[0088] In the above process, based on the vehicle response information corresponding to the target traffic flow at the current moment, the first parameter, second parameter and third parameter of each vehicle in the target traffic flow are analyzed, that is, the acceleration performance of each vehicle in the target traffic flow, the reaction delay of the corresponding driver and the distance between the vehicle and the intersection are analyzed. The degree to which each vehicle in the corresponding traffic flow affects the traffic flow is comprehensively evaluated from three aspects: the vehicle's dynamic performance, the reaction ability of the driver who controls the vehicle, and the impact of the vehicle's position on other vehicles in the traffic flow. The average of multiple impact levels of multiple vehicles is calculated and used as the slowness of the corresponding traffic flow, which can make the determined hysteresis parameter more accurate and reliable.

[0089] It should be noted that the larger the first parameter is, the stronger the acceleration performance of the corresponding vehicle is, and the lower the probability that the corresponding vehicle starts slowly in the traffic flow and affects the overall traffic efficiency of the traffic flow, and vice versa. Therefore, the first parameter and the hysteresis factor are negatively correlated.

[0090] The larger the second parameter is, the longer the reaction delay of the driver of the corresponding vehicle is, the longer it takes for the corresponding vehicle to respond to the green light and start in the traffic flow, and the higher the probability that the corresponding vehicle will affect the overall traffic efficiency of the traffic flow due to slow start, and vice versa. Therefore, the second parameter and the hysteresis factor are positively correlated.

[0091] The larger the third parameter is, the greater the distance between the corresponding vehicle and the intersection, the fewer other vehicles are queued behind the corresponding vehicle, the fewer vehicles are affected by the corresponding vehicle's slow start due to various factors, and the less impact the corresponding vehicle has on the overall traffic efficiency due to its slow start, and vice versa. Therefore, the third parameter and the hysteresis factor are negatively correlated.

[0092] In one example, if the current time is set to time t, the hysteresis factor of the j-th vehicle in the target traffic flow can be expressed as:

[0093]

[0094] in, represents the hysteresis factor of the j-th vehicle in the target traffic flow at time t, represents the second parameter of the j-th vehicle in the target traffic flow at time t, represents the first parameter of the j-th vehicle in the target traffic flow at time t, The third parameter representing the j-th vehicle in the target traffic flow at time t.

[0095] Furthermore, the step of obtaining the first parameter of each vehicle in the target traffic flow includes:

[0096] Analyze multiple movement processes of each vehicle in the target traffic flow within a first time period to obtain multiple starting accelerations corresponding to each vehicle in the target traffic flow, wherein the multiple starting accelerations correspond one-to-one to the multiple movement processes, the starting acceleration is the instantaneous acceleration of the corresponding vehicle at the start of the corresponding movement process, the duration of the first time period is a first preset duration, and the end time of the first time period is the current time;

[0097] An average of a plurality of initial accelerations corresponding to each vehicle in the target traffic flow is calculated to obtain a first parameter of each vehicle in the target traffic flow.

[0098] In this process, multiple movement processes of the corresponding vehicle within a first time period are monitored, and the instantaneous accelerations exhibited by the vehicle in multiple starting stages during the multiple movement processes are counted to calculate the average value as the first parameter representing the acceleration performance of the vehicle, which can relatively accurately reflect the acceleration performance of the vehicle.

[0099] Exemplarily, the first preset duration may be 3 seconds or 10 seconds.

[0100] In the application, the process of obtaining the instantaneous acceleration of the vehicle at the starting stage of the corresponding moving process can be:

[0101] Determining the starting movement time of the corresponding vehicle in the corresponding movement process;

[0102] Among a plurality of continuous video frames corresponding to the first time period, determining a video frame having the same acquisition moment as the start movement moment as the start video frame;

[0103] Determining, among a plurality of consecutive video frames corresponding to the first time period, a video frame next to the start video frame as an adjacent video frame;

[0104] Calculating the difference between the starting position of the vehicle in the starting video frame and its adjacent positions in adjacent video frames to obtain pixel position differences;

[0105] Convert pixel position differences based on pre-configured mapping information to obtain real position differences;

[0106] A ratio of the actual position difference to a time interval between two adjacent video frames in a plurality of continuous video frames is determined as an instantaneous acceleration of the corresponding vehicle in a starting phase of the corresponding moving process.

[0107] Among them, multiple continuous video frames can be multiple video frames captured by the binocular camera at 30 frames per second in the first time. In this case, the time interval between two adjacent video frames in the multiple continuous video frames can be 33.3 milliseconds. The position of the vehicle in the video frame can be understood as the center point position of the vehicle detection anchor frame determined by YOLOv7 in the corresponding video frame, and the mapping information is used to indicate the actual distance corresponding to the unit pixel length / width in the video frame.

[0108] Furthermore, the step of obtaining the second parameter of each vehicle in the target traffic flow includes:

[0109] Analyze multiple movement processes of each vehicle in the target traffic flow within a first time period to obtain multiple start data corresponding to each vehicle in the target traffic flow, wherein the multiple start data correspond to the multiple movement processes in a one-to-one manner, and the start data each include a self-start time of the corresponding vehicle and a start time of a preceding vehicle, wherein the self-start time is the starting movement time of the corresponding vehicle in the corresponding movement process, and the preceding vehicle start time is the starting movement time of a vehicle preceding the corresponding vehicle in the corresponding movement process;

[0110] Obtaining, based on a plurality of start data corresponding to each vehicle in the target traffic flow, a plurality of start time differences corresponding to each vehicle in the target traffic flow, wherein the start time difference is the difference between the self-start time in the corresponding start data and the start time of the preceding vehicle;

[0111] The average of multiple starting time differences corresponding to each vehicle in the target traffic flow is calculated to obtain the second parameter of each vehicle in the target traffic flow.

[0112] In this process, the multiple movement processes of the corresponding vehicle within the first time period are monitored, and the difference in the starting time of the vehicle compared with the preceding vehicle during the multiple movement processes is counted, so as to calculate the average value as the second parameter representing the reaction delay of the vehicle driver. This can relatively accurately reflect the operation delay of the vehicle driver in the process of following the traffic flow.

[0113] Step S3: obtaining a right-turn congestion parameter based on the right-turn accumulation parameter and the right-turn hysteresis parameter, obtaining a straight-through congestion parameter based on the straight-through accumulation parameter and the straight-through hysteresis parameter, obtaining a left-turn congestion parameter based on the left-turn accumulation parameter and the left-turn hysteresis parameter, and obtaining an export congestion parameter based on the export accumulation parameter and the export hysteresis parameter.

[0114] The larger the accumulation parameter, the faster the growth rate of the number of vehicles included in the corresponding traffic flow, that is, the number of vehicles waiting to pass in the corresponding traffic flow is on the rise, and the higher the congestion level of the corresponding traffic flow; similarly, the larger the hysteresis parameter, the lower the overall travel efficiency of each vehicle in the corresponding traffic flow, the lower the overall traffic efficiency during the green light period, and the higher the congestion level of the corresponding vehicles.

[0115] Therefore, the right-turn congestion parameter can be the product of the right-turn accumulation parameter and the right-turn hysteresis parameter. Correspondingly, the left-turn congestion parameter can be the product of the left-turn accumulation parameter and the left-turn hysteresis parameter. The straight-through congestion parameter can be the product of the straight-through accumulation parameter and the straight-through hysteresis parameter. The export congestion parameter can be the product of the export accumulation parameter and the export hysteresis parameter.

[0116] Step S4: Analyze the congestion difference between the first road section and the second road section based on the right-turn congestion parameter, the straight-through congestion parameter, the left-turn congestion parameter, and the exported congestion parameter to generate tidal lane control information.

[0117] In this process, the congestion levels of the traffic flows waiting to merge into the first section can be analyzed based on the right-turn congestion parameters, the straight-ahead congestion parameters, and the left-turn congestion parameters. The congestion levels of the traffic flows waiting to merge out of the second section can then be shared. By comparing the difference between the two, the tidal lane can be adaptively controlled to open or close to balance the congestion difference between the first section and the second section, so that the tidal lane can be used relatively effectively.

[0118] Furthermore, the generating of tidal lane control information by analyzing the difference in congestion between the first road section and the second road section based on the right-turn congestion parameter, the straight-through congestion parameter, the left-turn congestion parameter, and the export congestion parameter includes:

[0119] Obtain multiple straight-through efficiency values ​​of the straight-through green light period of the traffic flow waiting to merge into in the second time period, multiple left-turn efficiency values ​​of the left-turn green light period of the traffic flow waiting to merge into in the second time period, multiple straight-through green light accumulation parameters of the right-turn traffic flow waiting to merge into in the second time period, and multiple left-turn green light accumulation parameters of the right-turn traffic flow waiting to merge into in the second time period, wherein the straight-through efficiency value is the ratio of the number of vehicles passing through the corresponding straight-through green light period of the traffic flow waiting to merge into and the duration of the corresponding straight-through green light period, and the left-turn efficiency value is the ratio of the number of vehicles passing through the corresponding left-turn green light period of the traffic flow waiting to merge into and the duration of the corresponding left-turn green light period, the duration of the second time period is a second preset duration, and the end time of the second time period is the current time;

[0120] performing a correlation analysis on the plurality of straight-through traffic efficiency values ​​and the plurality of straight-through green light accumulation parameters to obtain a straight-right correlation value, and performing a correlation analysis on the plurality of left-turn traffic efficiency values ​​and the plurality of left-turn green light accumulation parameters to obtain a left-right correlation value;

[0121] According to the straight-right correlation value and the left-right correlation value, the right-turn congestion parameter, the straight-through congestion parameter, and the left-turn congestion parameter are respectively corrected to obtain a right-turn correction parameter, a straight-through correction parameter, and a left-turn correction parameter;

[0122] According to the right-turn correction parameter, the straight-ahead correction parameter, the left-turn correction parameter, and the exported congestion parameter, a congestion difference between the first road section and the second road section is analyzed to generate tidal lane control information.

[0123] Exemplarily, the second preset duration may be 12 hours, and the duration of the straight green light period and the duration of the left turn green light period may both be 30 seconds.

[0124] The process of obtaining the above-mentioned straight green light accumulation parameters can be as follows:

[0125] Obtain the first maximum number of vehicles and the first minimum number of vehicles in the right-turning traffic flow waiting to merge during the corresponding straight-ahead green light period;

[0126] Calculating a sum of a first maximum number of vehicles and a first minimum number of vehicles to obtain a first vehicle number sum value;

[0127] The ratio of the first vehicle number and value to the duration of the through green light period is calculated to obtain a through green light accumulation parameter of the right-turn merging traffic flow in the corresponding through green light period.

[0128] The process of obtaining the left turn green light accumulation parameters can be as follows:

[0129] Obtain the second maximum number of vehicles and the second minimum number of vehicles in the right-turning traffic flow waiting to merge during the corresponding left-turn green light period;

[0130] calculating a sum of a second maximum number of vehicles and a second minimum number of vehicles to obtain a second vehicle number sum value;

[0131] The ratio of the second vehicle number and value to the duration of the straight green light period is calculated to obtain the left-turn green light accumulation parameter of the right-turn merging traffic flow in the corresponding left-turn green light period.

[0132] Exemplarily, the straight-right correlation value can be the Pearson correlation coefficient between the multiple straight-through traffic efficiency values ​​and the multiple straight-through green light accumulation parameters. Similarly, the left-right correlation value can be the Pearson correlation coefficient between the multiple left-turn traffic efficiency values ​​and the multiple left-turn green light accumulation parameters. In this case, the closer the straight-right correlation value is to 1, the greater the degree of influence on the traffic efficiency of the right-turn traffic flow to be merging and the straight-through traffic flow to be merging. Similarly, the closer the left-right correlation value is to 1, the greater the degree of influence on the traffic efficiency of the right-turn traffic flow to be merging and the left-turn traffic flow to be merging.

[0133] Since the vehicles waiting to merge on the left-turn road and the through road are strictly controlled by traffic lights, while the vehicles waiting to merge on the right-turn road can directly merge into the first road section when the right-turn green light is not on, the traffic efficiency of the left-turn waiting to merge and the straight-going waiting to merge traffic will be affected by the right-turn waiting to merge traffic. Based on this, through the above settings, the green light traffic efficiency of the left-turn waiting to merge traffic and the accumulated parameters of the right-turn waiting to merge traffic in the corresponding left-turn green light period are analyzed, and the green light traffic efficiency of the straight-going waiting to merge traffic and the accumulated parameters of the right-turn waiting to merge traffic in the corresponding straight-going green light period are analyzed to evaluate the impact of the right-turn waiting to merge traffic and the straight-going waiting to merge traffic on the traffic efficiency, and accordingly correct the right-turn congestion parameters, the straight-going congestion parameters and the left-turn congestion parameters to accurately reflect the congestion impact caused by the right-turn waiting to merge traffic, the straight-going waiting to merge traffic and the left-turn waiting to merge traffic in the first road section.

[0134] Furthermore, the right-turn congestion parameter, the straight-through congestion parameter, and the left-turn congestion parameter are respectively corrected according to the straight-right correlation value and the left-right correlation value to obtain the right-turn correction parameter, the straight-through correction parameter, and the left-turn correction parameter, including:

[0135] According to the straight-right correlation value and the left-right correlation value, a straight-moving correction value, a left-turn correction value, and a right-turn correction value are obtained, wherein the sum of the straight-right correlation value and the straight-moving correction value is 1, the sum of the left-right correlation value and the left-turn correction value is 1, and the right-turn correction value is the sum of the straight-right correlation value and the left-right correlation value;

[0136] The product of the right-turn congestion parameter and the right-turn correction value is determined as the right-turn correction parameter, the product of the left-turn congestion parameter and the left-turn correction value is determined as the left-turn correction parameter, and the product of the straight-line congestion parameter and the straight-line correction value is determined as the straight-line correction parameter.

[0137] For example, the right-turn correction parameter, the left-turn correction parameter, and the straight-ahead correction parameter can be expressed as:

[0138]

[0139]

[0140]

[0141] in, Indicates the left turn correction parameter, represents the straight-line correction parameter, Indicates the right turn correction parameter, represents the left-right correlation value, represents the direct right correlation value, Indicates the left turn correction value, Indicates the straight-line correction value. Indicates the right turn correction value, Indicates the left-turn congestion parameter, represents the straight-through congestion parameter, Indicates the right-turn congestion parameter.

[0142] Furthermore, the generating of tidal lane control information by analyzing the difference in congestion between the first road section and the second road section based on the right-turn correction parameter, the straight-ahead correction parameter, the left-turn correction parameter, and the exported congestion parameter includes:

[0143] When the first road section and the second road section match the congestion condition, calculating the average of the right-turn correction parameter, the straight-going correction parameter, and the left-turn correction parameter to obtain a target congestion parameter;

[0144] generating tidal lane control information instructing to close the tidal lane when the target congestion parameter is greater than the outgoing congestion parameter;

[0145] When the target congestion parameter is less than the outgoing congestion parameter, tidal lane control information indicating opening the tidal lane is generated.

[0146] It should be noted that when the target congestion parameter is equal to the exported congestion parameter, tidal lane control information is generated to indicate maintaining the current state of the tidal lane. Maintaining the current state of the tidal lane is specifically: when the tidal lane is open, the tidal lane is controlled to remain open; and when the tidal lane is closed, the tidal lane is controlled to remain closed.

[0147] The congestion condition that matches the first road section and the second road section can be understood as: the overall congestion condition of the road where the first road section and the second road section are located is relatively serious (that is, tidal lanes need to be opened or closed to alleviate traffic congestion on the road).

[0148] In this process, matching judgment is made based on congestion conditions to avoid traffic congestion caused by frequent opening and closing of tidal lanes, so that tidal lanes can be used reasonably.

[0149] In the application, a condition judgment operation can be performed every 20 minutes or 30 minutes to determine whether the first road section and the second road section match the congestion conditions. If they match, it is further determined whether the tidal lane needs to be opened or closed. If they do not match, the current state of the tidal lane is maintained unchanged.

[0150] Before comparing the target congestion parameters with the exported congestion parameters, the target congestion parameters and the exported congestion parameters can be normalized respectively to obtain the input normalized parameters and the exported normalized parameters, and then the numerical differences between the input normalized parameters and the exported normalized parameters are compared to avoid errors introduced by the numerical calculation differences between the target congestion parameters and the exported congestion parameters, thereby improving the accuracy of judging the congestion difference between the first road section and the second road section, and thereby improving the control accuracy of the tidal lane.

[0151] Among them, when the target congestion parameter is greater than the outgoing congestion parameter, it can be regarded that the overall congestion situation of the incoming traffic flow of the first section is more serious than the overall congestion situation of the outgoing traffic flow of the second section. At this time, it is chosen to close the tidal lane so that the first section can maintain / obtain a wider intersection for vehicles to merge in, so as to alleviate the overall congestion situation of the incoming traffic flow of the first section and enhance the improvement effect of the tidal lane on the overall congestion situation of the roads where the first section and the second section are located.

[0152] When the target congestion parameter is less than the outgoing congestion parameter, it can be considered that the overall congestion of the outgoing traffic flow on the second section is more serious than the overall congestion of the incoming traffic flow on the first section. At this time, it is chosen to open the tidal lane so that the second section can maintain / obtain a wider intersection for vehicles to merge, so as to alleviate the overall congestion of the outgoing traffic flow on the second section and enhance the improvement effect of the tidal lane on the overall congestion of the roads where the first and second sections are located.

[0153] Specifically, before calculating the average of the right-turn correction parameter, the straight-going correction parameter, and the left-turn correction parameter to obtain the target congestion parameter, the method further includes:

[0154] Obtain a right-turn curve corresponding to the right-turn correction parameter, a straight-line curve corresponding to the straight-line correction parameter, a left-turn curve corresponding to the left-turn correction parameter, and an outflow curve corresponding to the outflow congestion parameter, wherein the right-turn curve is a curve showing a change in the congestion level of the right-turning traffic to be merging in a third time period, the right-turn correction parameter indicates the congestion level of the right-turning traffic to be merging in at a current moment, the straight-line curve is a curve showing a change in the congestion level of the straight-line traffic to be merging in a third time period, the straight-line correction parameter indicates the congestion level of the straight-line traffic to be merging in at a current moment, the left-turn curve is a curve showing a change in the congestion level of the left-turning traffic to be merging in a third time period, the left-turn correction parameter indicates the congestion level of the left-turning traffic to be merging in at a current moment, the outflow curve is a curve showing a change in the congestion level of the outflow traffic to be merged in a third time period, the outflow congestion parameter indicates the congestion level of the outflow traffic to be merged in at a current moment, the duration of the third time period is a third preset duration, and the end time of the third time period is the current moment;

[0155] Obtain multiple right-turn curve slopes at multiple sampling moments within a third time period based on the right-turn curve, obtain multiple straight-line curve slopes at multiple sampling moments based on the straight-line curve, obtain multiple left-turn curve slopes at multiple sampling moments based on the left-turn curve, and obtain multiple export curve slopes at multiple sampling moments based on the export curve;

[0156] The average value of the slopes of the plurality of right-turn curves is determined as the right-turn slope value, the average value of the slopes of the plurality of straight-line curves is determined as the straight-line slope value, the average value of the slopes of the plurality of left-turn curves is determined as the left-turn slope value, and the average value of the slopes of the plurality of outgoing curves is determined as the outgoing slope value;

[0157] The product of the right-turn slope value and the right-turn correction parameter is determined as the right-turn congestion value, the product of the straight-line slope value and the straight-line correction parameter is determined as the straight-line congestion value, the product of the left-turn slope value and the left-turn correction parameter is determined as the left-turn congestion value, and the product of the export slope value and the export congestion parameter is determined as the export congestion value;

[0158] The right-turn congestion value, the straight-through congestion value, the left-turn congestion value, and the export congestion value are analyzed to determine whether the first road segment and the second road segment match the congestion condition.

[0159] Exemplarily, the third preset duration is 20 minutes or 30 minutes.

[0160] It should be understood that each data point in the right-turn curve corresponds to two dimensions, namely the time dimension and the congestion level dimension. In the right-turn curve, the value of the data point corresponding to the current moment in the congestion level dimension is the right-turn correction parameter. The values ​​of other data points in the right-turn curve in the congestion level dimension can refer to the calculation process of the right-turn correction parameter. Similarly, the straight curve, left-turn curve and export curve are similar to the right-turn curve. In order to avoid repetition, they will not be repeated.

[0161] In the above process, a curve is constructed showing the time-varying congestion level of each traffic flow in the most recent period (i.e., the third time period), and the average of the slopes of the curves at each point on the curve is calculated as the rate of change (i.e., the slope value) of the congestion level of the corresponding traffic flow in the most recent period. Furthermore, the product of the rate of change and a correction parameter / congestion parameter indicating the current congestion level is calculated to comprehensively determine the congestion status of the corresponding traffic flow from both the congestion level and its degree of change. This allows for a more accurate quantitative representation of the congestion status. The more severe the congestion status of the corresponding traffic flow, the greater the congestion value of the corresponding traffic flow.

[0162] Specifically, analyzing the right-turn congestion value, the straight-ahead congestion value, the left-turn congestion value, and the export congestion value to determine whether the first road section and the second road section match the congestion condition includes:

[0163] Calculate the average of the right-turn congestion value, the straight-through congestion value, the left-turn congestion value, and the export congestion value to obtain the congestion index value;

[0164] If the congestion index value is greater than or equal to a congestion threshold, determining that the first road section and the second road section match a congestion condition;

[0165] When the congestion index value is less than a congestion threshold, it is determined that the first road section and the second road section do not match a congestion condition.

[0166] The above congestion index value can be understood as the congestion condition of the traffic waiting to merge into the first road section. The larger the congestion index value, the more serious the congestion condition of the traffic waiting to merge into the first road section, and vice versa.

[0167] The process of obtaining the above congestion threshold can be:

[0168] Obtain multiple historical congestion values ​​corresponding to the past year or the past three months. The calculation process of historical congestion values ​​and congestion index values ​​is the same. The difference between the two is that the time corresponding to the historical congestion value is the time when the tidal lane is activated according to the established rules. The above established rules are to activate the tidal lane during a set time period (such as 6:00 to 8:00 every morning);

[0169] Among multiple historical congestion values, the minimum historical congestion value is determined as the congestion threshold.

[0170] In general, the present invention first analyzes the vehicle accumulation information corresponding to the first road section and the second road section at the current moment to obtain the right turn accumulation parameter corresponding to the right-turn traffic flow to be merged, the straight-ahead accumulation parameter corresponding to the straight-ahead traffic flow to be merged, the left turn accumulation parameter corresponding to the left-turn traffic flow to be merged, and the outflow accumulation parameter corresponding to the outflow traffic flow, that is, to determine the accumulation situation of the traffic flow to be merged into the first road section by turning right, going straight, and turning left, respectively, and to determine the accumulation situation of the traffic flow to be outflowed from the second road section, and then further analyzes the vehicle response information corresponding to the first road section and the second road section at the current moment to obtain the right turn hysteresis parameter, the straight-ahead hysteresis parameter, the left turn hysteresis parameter, and the outflow hysteresis parameter, that is, to determine the traffic flow to be merged into the first road section by turning right, going straight, and turning left. The flow hysteresis of the traffic flow entering the first section by turning right, going straight and turning left, as well as the flow hysteresis of the traffic flow to be discharged from the second section, are respectively calculated. Then, the corresponding congestion parameters are obtained by comprehensively integrating the accumulation parameters and hysteresis parameters, so as to accurately evaluate the congestion condition of the corresponding traffic flow from both the accumulation condition and the flow hysteresis condition. Finally, the difference in section congestion between the first section and the second section is analyzed, that is, the difference in congestion condition between the incoming traffic flow and the outgoing traffic flow is analyzed, so as to flexibly complete the opening and closing of the tidal lane while ensuring the accuracy of control, so as to better adapt to complex lane conditions, improve the control effect of the tidal lane, and make the tidal lane more effectively used.

[0171] This invention proposes a vehicle-assisted traffic system at a traffic intersection based on a vehicle AI platform. Figure 2 , which shows a schematic structural diagram of a vehicle assisted traffic system 200 at a traffic intersection based on a vehicle AI platform provided by one embodiment of the present invention. The system includes:

[0172] The first analysis module 201 is configured to analyze the vehicle accumulation information corresponding to the first road section and the second road section at the current moment, and obtain a right-turn accumulation parameter corresponding to the right-turning traffic flow to be merged, a straight-line accumulation parameter corresponding to the straight-line traffic flow to be merged, a left-turn accumulation parameter corresponding to the left-turning traffic flow to be merged, and an outgoing accumulation parameter corresponding to the outgoing traffic flow, wherein the first road section and the second road section are adjacent, the first road section is a road section where the traffic flow merges in, and the second road section is a road section where the traffic flow merges out, and when the tidal lane is opened, the merging width of the first road section becomes narrower;

[0173] The second analysis module 202 is configured to analyze the vehicle response information corresponding to the first road segment and the second road segment at the current moment, and obtain a right-turn hysteresis parameter, a straight-ahead hysteresis parameter, a left-turn hysteresis parameter, and an export hysteresis parameter;

[0174] a parameter calculation module 203 for obtaining a right-turn congestion parameter based on the right-turn accumulation parameter and the right-turn hysteresis parameter, obtaining a straight-through congestion parameter based on the straight-through accumulation parameter and the straight-through hysteresis parameter, obtaining a left-turn congestion parameter based on the left-turn accumulation parameter and the left-turn hysteresis parameter, and obtaining an export congestion parameter based on the export accumulation parameter and the export hysteresis parameter;

[0175] The congestion analysis module 204 is configured to analyze the congestion difference between the first road segment and the second road segment based on the right-turn congestion parameter, the straight-through congestion parameter, the left-turn congestion parameter, and the exported congestion parameter, and generate tidal lane control information.

[0176] It should be noted that the system provided in the above embodiment is merely an example of the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the above embodiment provides a vehicle AI platform-based traffic intersection vehicle assisted passage system and a vehicle AI platform-based traffic intersection vehicle assisted passage method embodiment, which are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0177] The embodiment of the present invention also provides an electronic device. Figure 3 , the electronic device may include a processor 301, a memory 302, and a program 3021 stored in the memory 302 and executable on the processor 301.

[0178] When the program 3021 is executed by the processor 301, it can achieve Figure 1 Any steps in the corresponding method embodiments and achieving the same beneficial effects will not be repeated here.

[0179] Those skilled in the art will appreciate that all or part of the steps of implementing the above-described embodiment method may be accomplished through hardware associated with program instructions, and the program may be stored in a readable medium.

[0180] The embodiment of the present invention further provides a readable storage medium, wherein the readable storage medium stores a computer program, and when the computer program is executed by a processor, the above Figure 1 Any steps in the corresponding method embodiments can achieve the same technical effects and will not be described again here to avoid repetition.

[0181] The computer-readable storage medium of the embodiments of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or component.

[0182] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0183] The program code contained on the storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0184] Computer program code for performing the operations of the present invention may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or terminal. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0185] An embodiment of the present invention also provides a computer program product. When the computer program product is run on a computer, it enables the computer to execute the above-mentioned related steps to implement a vehicle assisted passage method at a traffic intersection based on a vehicle AI platform provided in the above embodiment.

[0186] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0187] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A vehicle-assisted traffic method at a traffic intersection based on a vehicle AI platform, characterized in that: The method comprises: Analyze vehicle accumulation information corresponding to the first road section and the second road section at the current moment to obtain a right-turn accumulation parameter corresponding to the right-turning traffic flow to be merged, a straight-moving accumulation parameter corresponding to the straight-moving traffic flow to be merged, a left-turn accumulation parameter corresponding to the left-turning traffic flow to be merged, and an outflow accumulation parameter corresponding to the outflowing traffic flow, wherein the first road section and the second road section are adjacent, the first road section is a road section where the traffic flow merges in, and the second road section is a road section where the traffic flow merges out, and when the tidal lane is opened, the merging width of the first road section becomes narrower; Analyze the vehicle response information corresponding to the first road section and the second road section at the current moment to obtain a right turn hysteresis parameter, a straight hysteresis parameter, a left turn hysteresis parameter, and an export hysteresis parameter; A right-turn congestion parameter is obtained according to the right-turn accumulation parameter and the right-turn hysteresis parameter, a straight-line congestion parameter is obtained according to the straight-line accumulation parameter and the straight-line hysteresis parameter, a straight-line congestion parameter is obtained according to the left-turn accumulation parameter and the left-turn hysteresis parameter, and an export congestion parameter is obtained according to the export accumulation parameter and the export hysteresis parameter; Obtain multiple straight-through efficiency values ​​of the straight-through green light period of the traffic flow waiting to merge into in the second time period, multiple left-turn efficiency values ​​of the left-turn green light period of the traffic flow waiting to merge into in the second time period, multiple straight-through green light accumulation parameters of the right-turn traffic flow waiting to merge into in the second time period, and multiple left-turn green light accumulation parameters of the right-turn traffic flow waiting to merge into in the second time period, wherein the straight-through efficiency value is the ratio of the number of vehicles passing through the corresponding straight-through green light period of the traffic flow waiting to merge into and the duration of the corresponding straight-through green light period, and the left-turn efficiency value is the ratio of the number of vehicles passing through the corresponding left-turn green light period of the traffic flow waiting to merge into and the duration of the corresponding left-turn green light period, the duration of the second time period is a second preset duration, and the end time of the second time period is the current time; performing a correlation analysis on the plurality of straight-through traffic efficiency values ​​and the plurality of straight-through green light accumulation parameters to obtain a straight-right correlation value, and performing a correlation analysis on the plurality of left-turn traffic efficiency values ​​and the plurality of left-turn green light accumulation parameters to obtain a left-right correlation value; According to the straight-right correlation value and the left-right correlation value, a straight-moving correction value, a left-turn correction value, and a right-turn correction value are obtained, wherein the sum of the straight-right correlation value and the straight-moving correction value is 1, the sum of the left-right correlation value and the left-turn correction value is 1, and the right-turn correction value is the sum of the straight-right correlation value and the left-right correlation value; Determine the product of the right-turn congestion parameter and the right-turn correction value as the right-turn correction parameter, determine the product of the left-turn congestion parameter and the left-turn correction value as the left-turn correction parameter, and determine the product of the straight-through congestion parameter and the straight-through correction value as the straight-through correction parameter; According to the right-turn correction parameter, the straight-ahead correction parameter, the left-turn correction parameter, and the exported congestion parameter, a congestion difference between the first road section and the second road section is analyzed to generate tidal lane control information.

2. The vehicle-assisted traffic method at a traffic intersection based on a vehicle AI platform according to claim 1 is characterized in that: The analyzing of the vehicle accumulation information corresponding to the first road section and the second road section at the current moment to obtain the right-turn accumulation parameter corresponding to the right-turning traffic flow to be merged, the straight-going accumulation parameter corresponding to the straight-going traffic flow to be merged, the left-turn accumulation parameter corresponding to the left-turning traffic flow to be merged, and the outgoing accumulation parameter corresponding to the outgoing traffic flow, includes: According to the vehicle accumulation information corresponding to the target traffic flow at the current moment, the maximum number of vehicles and the minimum number of vehicles corresponding to the target traffic flow are obtained, wherein the maximum number of vehicles is the maximum number of vehicles accumulated in the target traffic flow during the previous traffic light alternation cycle at the current moment, and the minimum number of vehicles is the minimum number of vehicles accumulated in the previous traffic light alternation cycle at the current moment, and the target traffic flow is a right-turning traffic flow to be merged, a straight-going traffic flow to be merged, a left-turning traffic flow to be merged, or a traffic flow to be merged out; Calculate the sum of the maximum number of vehicles and the minimum number of vehicles corresponding to the target traffic flow to obtain a reference value of the number of vehicles in the target traffic flow; The ratio of the reference value of the number of vehicles of the target traffic flow to the target cycle value is calculated to obtain the accumulation parameter corresponding to the target traffic flow, wherein the target cycle value is the time value indicated by the traffic light alternation cycle.

3. The vehicle-assisted traffic method at a traffic intersection based on a vehicle AI platform according to claim 1, characterized in that: The analyzing the vehicle response information corresponding to the first road section and the second road section at the current moment to obtain the right turn hysteresis parameter, the straight hysteresis parameter, the left turn hysteresis parameter and the export hysteresis parameter includes: Obtaining, based on vehicle response information corresponding to the target traffic flow at the current moment, a first parameter, a second parameter, and a third parameter for each vehicle in the target traffic flow, wherein the first parameter is used to indicate the acceleration performance of the corresponding vehicle, the second parameter is used to indicate the reaction delay of the driver of the corresponding vehicle, and the third parameter is used to indicate the distance between the corresponding vehicle and the intersection, where the target traffic flow is a right-turning traffic flow to be merged, a straight-moving traffic flow to be merged, a left-turning traffic flow to be merged, or a traffic flow to be merged out; Obtaining a hysteresis factor for each vehicle in the target traffic flow according to the first parameter, the second parameter, and the third parameter of each vehicle in the target traffic flow; The average of multiple hysteresis factors of multiple vehicles in the target traffic flow is calculated to obtain a hysteresis parameter corresponding to the target traffic flow.

4. The vehicle-assisted traffic method at a traffic intersection based on a vehicle AI platform according to claim 3 is characterized in that: The step of obtaining the first parameter of each vehicle in the target traffic flow includes: Analyze multiple movement processes of each vehicle in the target traffic flow within a first time period to obtain multiple starting accelerations corresponding to each vehicle in the target traffic flow, wherein the multiple starting accelerations correspond one-to-one to the multiple movement processes, the starting acceleration is the instantaneous acceleration of the corresponding vehicle at the start of the corresponding movement process, the duration of the first time period is a first preset duration, and the end time of the first time period is the current time; An average of a plurality of initial accelerations corresponding to each vehicle in the target traffic flow is calculated to obtain a first parameter of each vehicle in the target traffic flow.

5. The vehicle-assisted traffic method at a traffic intersection based on a vehicle AI platform according to claim 3 is characterized in that: The step of obtaining the second parameter of each vehicle in the target traffic flow includes: Analyze multiple movement processes of each vehicle in the target traffic flow within a first time period to obtain multiple start data corresponding to each vehicle in the target traffic flow, wherein the multiple start data correspond to the multiple movement processes in a one-to-one manner, and the start data each include a self-start time of the corresponding vehicle and a start time of a preceding vehicle, wherein the self-start time is the starting movement time of the corresponding vehicle in the corresponding movement process, and the preceding vehicle start time is the starting movement time of a vehicle preceding the corresponding vehicle in the corresponding movement process; Obtaining, based on a plurality of start data corresponding to each vehicle in the target traffic flow, a plurality of start time differences corresponding to each vehicle in the target traffic flow, wherein the start time difference is the difference between the self-start time in the corresponding start data and the start time of the preceding vehicle; The average of multiple starting time differences corresponding to each vehicle in the target traffic flow is calculated to obtain the second parameter of each vehicle in the target traffic flow.

6. The vehicle-assisted traffic method at a traffic intersection based on a vehicle AI platform according to claim 1, characterized in that: The method of analyzing the congestion difference between the first road section and the second road section according to the right-turn correction parameter, the straight-ahead correction parameter, the left-turn correction parameter, and the exported congestion parameter to generate tidal lane control information includes: When the first road section and the second road section match the congestion condition, calculating the average of the right-turn correction parameter, the straight-going correction parameter, and the left-turn correction parameter to obtain a target congestion parameter; generating tidal lane control information instructing to close the tidal lane when the target congestion parameter is greater than the outgoing congestion parameter; When the target congestion parameter is less than the outgoing congestion parameter, tidal lane control information indicating opening the tidal lane is generated.

7. The vehicle-assisted traffic method at a traffic intersection based on a vehicle AI platform according to claim 6, characterized in that: Before calculating the average of the right-turn correction parameter, the straight-ahead correction parameter, and the left-turn correction parameter to obtain the target congestion parameter, the method further includes: Obtain a right-turn curve corresponding to the right-turn correction parameter, a straight-line curve corresponding to the straight-line correction parameter, a left-turn curve corresponding to the left-turn correction parameter, and an outflow curve corresponding to the outflow congestion parameter, wherein the right-turn curve is a curve showing a change in the congestion level of the right-turning traffic to be merging in a third time period, the right-turn correction parameter indicates the congestion level of the right-turning traffic to be merging in at a current moment, the straight-line curve is a curve showing a change in the congestion level of the straight-line traffic to be merging in a third time period, the straight-line correction parameter indicates the congestion level of the straight-line traffic to be merging in at a current moment, the left-turn curve is a curve showing a change in the congestion level of the left-turning traffic to be merging in a third time period, the left-turn correction parameter indicates the congestion level of the left-turning traffic to be merging in at a current moment, the outflow curve is a curve showing a change in the congestion level of the outflow traffic to be merged in a third time period, the outflow congestion parameter indicates the congestion level of the outflow traffic to be merged in at a current moment, the duration of the third time period is a third preset duration, and the end time of the third time period is the current moment; Obtain multiple right-turn curve slopes at multiple sampling moments within a third time period based on the right-turn curve, obtain multiple straight-line curve slopes at multiple sampling moments based on the straight-line curve, obtain multiple left-turn curve slopes at multiple sampling moments based on the left-turn curve, and obtain multiple export curve slopes at multiple sampling moments based on the export curve; The average value of the slopes of the plurality of right-turn curves is determined as the right-turn slope value, the average value of the slopes of the plurality of straight-line curves is determined as the straight-line slope value, the average value of the slopes of the plurality of left-turn curves is determined as the left-turn slope value, and the average value of the slopes of the plurality of outgoing curves is determined as the outgoing slope value; The product of the right-turn slope value and the right-turn correction parameter is determined as the right-turn congestion value, the product of the straight-line slope value and the straight-line correction parameter is determined as the straight-line congestion value, the product of the left-turn slope value and the left-turn correction parameter is determined as the left-turn congestion value, and the product of the export slope value and the export congestion parameter is determined as the export congestion value; The right-turn congestion value, the straight-through congestion value, the left-turn congestion value, and the export congestion value are analyzed to determine whether the first road segment and the second road segment match the congestion condition.

8. The vehicle-assisted traffic method at a traffic intersection based on a vehicle AI platform according to claim 7 is characterized in that: The analyzing the right-turn congestion value, the straight-ahead congestion value, the left-turn congestion value, and the export congestion value to determine whether the first road section and the second road section match the congestion condition includes: Calculate the average of the right-turn congestion value, the straight-through congestion value, the left-turn congestion value, and the export congestion value to obtain the congestion index value; If the congestion index value is greater than or equal to a congestion threshold, determining that the first road section and the second road section match a congestion condition; When the congestion index value is less than a congestion threshold, it is determined that the first road section and the second road section do not match a congestion condition.

Images