Traffic information processing method, device, storage medium, controller and vehicle

Abstract

The present invention discloses a traffic information processing method, device, storage medium, controller and vehicle through embodiments; utilizes traffic flow data of the Internet of Vehicles, real-time navigation data, data customized by map vendors, etc. to realize dynamic judgment of the lane characteristics of the target vehicle; further obtains the vehicle traffic model of the relevant intersection through model parameter fitting; based on the predicted target vehicle traffic data, combined with the action threshold of the automatic start-stop system, it can avoid frequent vehicle starting, improve the vehicle's operating stability and acceleration index near zero speed; in addition, the parameters of the relevant models can be optimized through real-time data, so that the system performance is further improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent vehicles, and in particular relates to a traffic information processing method, device, storage medium, controller and vehicle. Background Art

[0002] The automatic start-stop function of automobile engines has been around for many years, but it is often abandoned by vehicle users. The reasons are many: first, repeated starting and stopping of the engine will generate additional acceleration near zero speed; second, frequent starting will increase engine wear; third, stopping the engine may require cutting off high-power loads; fourth, frequent starting and stopping may not necessarily save fuel compared to idling; special operating conditions, such as slopes, wading and refueling, may also lead to irreparable losses. Summary of the Invention

[0003] An embodiment of the present invention discloses a traffic information processing method, device, storage medium, controller and vehicle. A traffic information processing method; the method includes a first information collection step and a second information prediction step; wherein the first information collection step is used to obtain first traffic flow information, and the first traffic flow information includes second historical traffic flow information and third real-time traffic flow information.

[0004] Specifically, the third real-time traffic flow information further includes fourth geographic positioning information, and the fourth geographic positioning information includes geographic location information of the first target vehicle Z. The first target vehicle Z is a vehicle for which the length of time for which it is to be predicted to be forced to stop safely.

[0005] Furthermore, the second information prediction step obtains a first prohibited passage time ts of the first target lane at the first target intersection, where the first target intersection is the first intersection with a red light, a duty post and / or a parking probability greater than zero in the direction of travel of the first target vehicle Z, and the first target lane is the lane through which the first target vehicle Z intends to pass through the first target intersection; wherein the first prohibited passage time ts is a historical value; ts includes the parking time, statistical value or simulation value in the second historical traffic flow information within a preset distance between the first target lane and the first target intersection, and ts also includes the first red light configuration time t0 or the second forced compulsory parking time t1 of the first target lane; if the first prohibited passage time ts is greater than or equal to the preset first ignition-off time threshold, the first ignition-off time threshold is used as the first prediction value (201).

[0006] Furthermore, the first information collection step also obtains the maximum parking time TM at the head of the queue in the first target lane, which is the maximum parking time of vehicles in the first target lane within a preset distance from the first target intersection in the second historical traffic flow information; the second information prediction step also includes outputting the first prediction value to the vehicle controller or updating the first traffic flow information for further processing; wherein, if TM is greater than t0 or TM is greater than t1, t0 is used as the first prediction value.

[0007] Furthermore, the embodiment of the traffic information processing method may also include a third start-stop control step; the third start-stop control step obtains vehicle operation information for decision-making of the automatic start-stop function; wherein, the first prohibited passage time ts is less than or equal to the first red light configuration time t0; and the first traffic flow information can come from a geographic information system GIS, a traffic information data source and / or a laboratory simulation data source.

[0008] Specifically, if the first predicted value is less than the first ignition-off time threshold, the automatic start-stop function of the first target vehicle Z is disabled.

[0009] Furthermore, its first traffic flow information may also include road condition information obtained in real time, which includes a flooded road section existence sign WV and a water level exceeding limit sensor emergency sign WL. When the vehicle's wading depth or the height of the wading surface waves exceeds the preset wading depth threshold or water level threshold within a preset time period, the water level exceeding limit sensor emergency sign WL will be set; if the flooded road section existence sign WV and / or the water level exceeding limit sensor emergency sign WL are valid, the automatic start and stop function will be turned off.

[0010] Furthermore, its second historical traffic flow information includes historical measurement data, and its third real-time traffic flow information includes real-vehicle collected data; its fourth geographic positioning information includes satellite positioning data; its fourth geographic positioning information also includes the fifth queue status information of the first target vehicle Z at the first target intersection; wherein the fifth queue status information includes the first light-on moment T1 (041) information of the first target intersection, that is, the moment when the red light turns on and all vehicles start to slow down and stop; also includes the second steady stop moment T2 (042) information, that is, the moment when the first target vehicle Z (060) stops steadily; also includes the third pre-start moment T3 (043) information, that is, the moment when the first vehicle at the first target intersection (050) starts to start when the green light turns on; also includes the fourth re-start moment T4 (044) information, that is, the moment when the first target vehicle Z starts to start;

[0011] Based on the fifth queue state information, historical data in the first traffic flow information is fitted, and based on the fitted function or model, a first predicted time tp experienced by the first target vehicle Z from stopping to starting again is predicted and / or output.

[0012] Specifically, the traffic information processing method may further include a fourth model construction step; by obtaining the total number of vehicles K on the first target lane between the first target vehicle Z and the first target intersection, a first prediction model based on the first traffic flow information (001) is constructed; wherein the parameters of the first prediction model are obtained by fitting the first traffic flow information; the first prediction model includes a linear model C = A·D2+B, A=-2 / V, B=td; the parameters A and B are obtained by fitting using the least squares method, V is the current traffic speed, obtained from the first traffic flow information, and D2 is the average distance between vehicles traveling.

[0013] Furthermore, the traffic information processing method may also include a fifth dynamic optimization step: taking tp=t0+KC as the first prediction model; wherein t0 is the first red light configuration time of the first target vehicle Z in the first target lane; by comparing the actual parking time tr of the first target vehicle Z with the first predicted time tp, a first prediction error E is obtained; the parameters of the first prediction model are corrected using the first prediction error; wherein the correction method of the first prediction error E includes a gradient descent method.

[0014] Through an embodiment, the present invention also discloses a traffic information processing device, including a first information collection unit and a second information prediction unit; wherein the first information collection unit obtains first traffic flow information, the first traffic flow information includes second historical traffic flow information and third real-time traffic flow information, and the third real-time traffic flow information also includes fourth geographic positioning information, and the fourth geographic positioning information includes the geographic location information of the first target vehicle Z, and the first target vehicle Z is the vehicle for which the length of time it is forced to stop safely is to be predicted.

[0015] Furthermore, the second information prediction unit obtains the first prohibited passage time ts of the first target lane at the first target intersection, where the first target intersection is the first intersection with a red light, a duty post and / or a parking probability greater than zero in the traveling direction of the first target vehicle Z, and the first target lane is the lane through which the first target vehicle Z intends to pass through the first target intersection; the first prohibited passage time ts is a historical value; ts includes the parking time, statistical value or simulation value in the second historical traffic flow information within a preset distance between the first target lane and the first target intersection, and ts may also include the first red light configuration time t0 or the second forced compulsory parking time t1 of the first target lane; if the first prohibited passage time ts is greater than or equal to the preset first ignition off time threshold, the first ignition off time threshold is used as the first predicted value.

[0016] Specifically, its first information collection unit can also obtain the maximum parking time TM at the head of the queue in the first target lane, which is the maximum parking time of vehicles within a preset distance from the first target lane to the first target intersection in the second historical traffic flow information; the second information prediction unit can also output the first prediction value to the vehicle controller or update the first traffic flow information for further processing; wherein, if TM is greater than t0 or TM is greater than t1, t0 is used as the first prediction value.

[0017] Furthermore, the traffic information processing device of this embodiment may also include a third start-stop control unit; the third start-stop control unit obtains vehicle operation information for decision-making of the automatic start-stop function; wherein the first prohibited passage time ts is less than or equal to the first red light configuration time t0; its first traffic flow information comes from a geographic information system GIS, a traffic information data source and / or laboratory simulation data; if the first predicted value is less than the first ignition off time threshold, the automatic start-stop function of the first target vehicle Z (060) is disabled; its first traffic flow information may also include road condition information obtained in real time, the road condition information includes a flooded road section existence flag WV, a water level over-limit sensor emergency flag WL, when the vehicle wading depth or the height of the wading surface wave exceeds the preset wading depth threshold or water level threshold within a preset time length, the water level over-limit sensor emergency flag WL is set; if the flooded road section existence flag WV and / or the water level over-limit sensor emergency flag WL are valid, the automatic start-stop function of the first target vehicle Z is turned off.

[0018] Specifically, its second historical traffic flow information may include historical measurement data, its third real-time traffic flow information may include real vehicle collected data; its fourth geographic positioning information may include satellite positioning data; its fourth geographic positioning information may also include the fifth queue status information of the first target vehicle Z at the first target intersection.

[0019] Specifically, its fifth queue status information includes the first light-on time T1 of the first target intersection, that is, the time when the red light comes on and all vehicles start to slow down and stop; it also includes the second stable stop time T2, that is, the time when the first target vehicle Z stops steadily; it also includes the third pre-start time T3, that is, the time when the green light comes on and the first vehicle at the first target intersection starts to start; it also includes the fourth re-start time T4, that is, the time when the first target vehicle Z starts to start.

[0020] Furthermore, based on the fifth queue state information, historical data in the first traffic flow information is fitted, and based on the fitted function or model, a first predicted time tp experienced by the first target vehicle Z from stopping to starting again is predicted and / or output.

[0021] Furthermore, the traffic information processing device may also include a fourth model construction unit; by obtaining the total number of vehicles K on the first target lane between the first target vehicle Z and the first target intersection, a first prediction model based on the first traffic flow information is constructed; wherein the parameters of the first prediction model are obtained after fitting the first traffic flow information.

[0022] Specifically, the first prediction model can be a linear model C = A·D2+B, A=-2 / V, B=td; wherein parameters A and B are obtained by least squares fitting, V is the current traffic speed provided by the first traffic flow information, and D2 is the average distance between vehicles.

[0023] Furthermore, the traffic information processing device of this embodiment may also be provided with a fifth dynamic optimization unit: and may adopt the first prediction model tp=t0+KC for optimization; wherein t0 is the first red light configuration time of the first target vehicle Z in the first target lane.

[0024] Furthermore, by comparing the actual parking time tr of the first target vehicle Z with the first predicted time tp, a first prediction error E is obtained; the parameters of the first prediction model are corrected using the first prediction error E; wherein, the correction method of the first prediction error E includes a gradient descent method.

[0025] Furthermore, under the premise of adopting the same inventive concept, an embodiment of the present invention also discloses a computer storage medium, a controller and a vehicle, wherein the storage medium includes a storage medium body for storing a computer program; when the computer program is executed by a microprocessor, it can implement the processing process of the above method.

[0026] In addition, its controller may include any of the above traffic information processing devices; if the first predicted value is less than a preset threshold value, the automatic start-stop function of the first target vehicle Z is disabled, the vehicle is prohibited from starting and / or the vehicle is prohibited from secondary starting; wherein, secondary starting is defined as the process of re-igniting the engine after it is shut down from the starting state; the controller may also include a special operating condition detection module, which is used to detect the operating conditions to be avoided that have the risk of secondary starting; specifically, the operating conditions to be avoided include wading conditions, climbing conditions, etc.

[0027] Furthermore, the above methods and products can be used for technical upgrades of vehicles. The implementation process is similar to that of the above products and will not be repeated here.

[0028] In summary, the present invention utilizes traffic flow data from the Internet of Vehicles, real-time navigation data, data customized by map vendors, etc. to achieve dynamic judgment of the lane characteristics of the target vehicle; it also obtains the vehicle traffic model of the relevant intersection through model parameter fitting; based on the predicted target vehicle traffic data, combined with the action threshold of the automatic start-stop system, it can avoid frequent vehicle starts, improve the vehicle's operating stability and acceleration indicators near zero speed; in addition, the parameters of the relevant models can be optimized through real-time data, so that the system performance is further improved.

[0029] It should be noted that the terms "first", "second" and similar terms used in this article are only for describing the various components of the technical solution, and do not constitute a limitation of the technical solution, nor can they be understood as an indication or suggestion of the importance of the corresponding elements; elements with terms such as "first", "second" and similar terms indicate that the corresponding technical solution contains at least one of the element. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solution of the present invention and facilitate a further understanding of the technical effects, technical features and purposes of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings. The accompanying drawings constitute an essential part of the specification and are used together with the embodiments of the present invention to illustrate the technical solution of the present invention, but do not constitute a limitation to the present invention.

[0031] The same reference numerals in the accompanying drawings represent the same components, specifically:

[0032] Figure 1 This is a schematic diagram of an initial model of a vehicle platoon according to an embodiment of the present invention;

[0033] Figure 2 This is a schematic diagram of the workflow of the method and product embodiment of the present invention;

[0034] Figure 3 This is a schematic diagram of a red light waiting time prediction model according to an embodiment of the method of the present invention;

[0035] Figure 4 This is a schematic diagram of the red light waiting time prediction process of the method embodiment of the present invention;

[0036] Figure 5 This is a comparison chart of the predicted and measured red light waiting time for the method and product embodiments of the present invention;

[0037] Figure 6 Schematic diagram of the process of the present invention;

[0038] Figure 7 This is a schematic diagram of the structure of an embodiment of the device of the present invention;

[0039] Figure 8Schematic diagram of the structure of each product embodiment of the present invention Figure 1 ;

[0040] Figure 9 Schematic diagram of the structure of each product embodiment of the present invention Figure 2 ;

[0041] Figure 10 Schematic diagram of the structure of each product embodiment of the present invention Figure 3 ;

[0042] Figure 11 Schematic diagram of the structure of each product embodiment of the present invention Figure 4 .

[0043] in:

[0044] 001-First traffic flow information,

[0045] 002-Second historical traffic flow information,

[0046] 003-Third real-time traffic flow information,

[0047] 004-Fourth geolocation information,

[0048] 006-Stop line, the rest are not marked, indicating the position where the vehicle speed must reach zero.

[0049] 007-Vehicle movement sign, the rest are not marked, indicating that the vehicle's speed towards the intersection is greater than zero.

[0050] 008- When the red light is on, the vehicle AAA is about to stop before the stop line of the target lane.

[0051] 009- When the red light is about to go out, the vehicle BBB enters the same target lane.

[0052] 010-A vehicle between vehicle AAA and vehicle BBB, the rest are not marked,

[0053] 011- The target lane’s red light duration is 1, the red light duration is 1,

[0054] 012- The second time of the red light in the target lane, the time of the red light when it is about to end,

[0055] 022-actual parking time t,

[0056] 033-first prediction time tp,

[0057] 040-First target lane,

[0058] 041-First lighting moment T1,

[0059] 042-Second stable stop time T2,

[0060] 043-The third pre-action moment T3,

[0061] 044-Fourth re-action time T4,

[0062] 050-First target intersection, only the intersection at time T1 is marked.

[0063] 051-Red light countdown one,

[0064] 052-Red light countdown 2,

[0065] 053-Green light countdown one,

[0066] 054-Green light countdown 2,

[0067] 060-The first target vehicle Z, only the time T3 is marked, the rest of the time is not marked,

[0068] 061-Current traffic speed V, related to traffic density,

[0069] 062-average length of general vehicles L,

[0070] 063-The average distance between vehicles traveling, D2, is related to the traffic density.

[0071] 064-The distance S that the first target vehicle Z decelerates from the red light to the point where it stops.

[0072] 065- The distance X between the vehicle and the center of the intersection at the time of parking,

[0073] 066- The distance between the front and rear vehicles when the vehicle is generally stopped is D1.

[0074] 067-The total number of vehicles ahead of the first target vehicle Z at the intersection K,

[0075] 077-Timeline,

[0076] 081-15s red light simulation comparison,

[0077] 082-30s red light simulation comparison,

[0078] 083-60s red light simulation comparison,

[0079] 088-three groups of sample points (15s, 30s, 60s),

[0080] 100-First information collection step,

[0081] 200-Second information prediction step,

[0082] 201-second real-time prediction error,

[0083] 300- The third start-stop control step,

[0084] 301-the third start-stop control signal,

[0085] 400-Fourth model construction step,

[0086] 500-Fifth dynamic optimization step,

[0087] 555-fifth queue status information,

[0088] 610-First information collection unit,

[0089] 620-second information prediction unit,

[0090] 630-3rd start-stop control unit,

[0091] 640-Fourth Model Construction Unit,

[0092] 650-Fifth Dynamic Optimization Unit,

[0093] 900-vehicles,

[0094] 901-Controller,

[0095] 902- Traffic information processing device,

[0096] 903-Storage medium,

[0097] 904- GIS data source or map provider data source,

[0098] 905-There are other intelligent units for information exchange. DETAILED DESCRIPTION

[0099] The present invention will be further described in detail below with reference to the accompanying drawings and examples. Of course, the specific embodiments described below are only intended to explain the technical solutions of the present invention, rather than to limit the present invention. In addition, the parts described in the embodiments or drawings are merely illustrative of the relevant parts of the present invention, rather than the entire present invention.

[0100] like Figure 1 As shown, due to the above-mentioned technical problems, some drivers are unwilling to turn on this function. This embodiment realizes information prediction during the intelligent start-stop process based on GPS and traffic light information; a preliminary prediction of the waiting time Tr of the first target vehicle Z at the red light can be made with some traffic flow information; and through modeling, parameter identification is performed based on offline data to finally obtain an offline prediction model; and then the parameters are iteratively updated based on the latest online data to finally obtain the predicted time result.

[0101] As shown above, with the predicted time Tp, the driver or developer can set a threshold. If the predicted time Tp is lower than this threshold, the stop will not trigger the start-stop function, thereby reducing the number of unnecessary stops and improving the driver's driving experience.

[0102] Down Figure 2 The figure shows two states of vehicles entering the intersection under normal traffic conditions. The AAA vehicle enters the intersection just when the red light comes on. At this time, the distance between the AAA vehicle and the intersection is very small. The BBB vehicle enters the intersection when the red light is about to end. It can be observed that under normal conditions, there is a negative correlation between the remaining time of the red light or the time the vehicle stops at this intersection and the distance between the vehicle and the intersection; this is also the main basis of the method of the present invention.

[0103] Furthermore, since there is a negative correlation between the vehicle's red light waiting time and the distance to the intersection, the relevant map traffic flow information obtained by the map provider through GPS can be combined with the vehicle status information inside the electronic control unit ECU (Electronic Control Unit) to predict the vehicle's red light waiting time and control the vehicle status to improve the driver's experience of the start-stop function.

[0104] Specifically, if Figure 2 、 Figure 3 As shown, the embodiment of the present invention establishes a vehicle red light waiting time prediction model scenario by modeling the vehicle's own state and the surrounding environment state before the red light.

[0105] We focus on four moments in the model:

[0106] T1: The first light-up moment, i.e. the moment when the red light comes on and all vehicles begin to slow down and stop;

[0107] T2: the second stable stop time, i.e. the time when the first target vehicle Z stops stably;

[0108] T3: The third pre-start time, that is, the moment when the green light comes on and the first car starts to move;

[0109] T4: the fourth restart time, that is, the time when the first target vehicle Z starts to start.

[0110] Since we are predicting the time from the first target vehicle Z stopping to starting, that is, T4-T2, according to the definition of the above time, we can expand T4-T2 into (T4-T3)+(T3-T1)-(T2-T1).

[0111] T4-T3 is the total vehicle start time. The number of vehicles K ahead of the first target vehicle Z at the intersection can be calculated from the vehicle length L, the inter-vehicle distance D1 when the vehicle stops, and the distance X between the vehicle and the center of the intersection at the time of stopping: K=X / (L+D1).

[0112] Therefore, according to the model parameters, the overall vehicle starting time is: T4-T3=K·td=(X / (L+D1))·td.

[0113] T3-T1 is the traffic light configuration time. According to the model parameters, we can obtain: T3-T1=t0.

[0114] T2-T1 is the time required for all vehicles to stop. Since the distance between each vehicle is D2, the time required for the first target vehicle Z to perform uniform deceleration linear motion until it stops is recorded as tsp.

[0115] And because the displacement formula S=V·tsp-a·(tsp)^2, the expression of tsp can be obtained: tsp=2S / V.

[0116] Since S=KD2, the time required for the first target vehicle Z to decelerate and stop can be obtained based on the model parameters:

[0117] T2-T1=tsp=(2KD2) / V=(X·2D2) / ((L+D1)·V);

[0118] Combining the above formula, the final formula for the vehicle red light waiting time prediction model is as follows:

[0119] tp=T4-T2=t0+(X·(td-(2D2) / V)) / (L+D1).

[0120] in:

[0121] t0 is the red light configuration duration;

[0122] L is the average length of a general vehicle;

[0123] td is the general average startup time;

[0124] V is the current traffic speed, which is related to the traffic density;

[0125] D1 is the distance between the front and rear vehicles when the vehicle is generally stopped;

[0126] D2 is the average distance between vehicles, which is also related to the traffic density;

[0127] X is the distance between the vehicle and the center of the intersection at the time of stopping;

[0128] K is the total number of vehicles ahead of the first target vehicle Z at the intersection;

[0129] S is the distance that the first target vehicle Z decelerates from the red light to the point where it stops.

[0130] In this embodiment of the present invention, a traffic scenario simulation model was established in a virtual vehicle laboratory. Nine intersections were set up, with traffic light durations set to 15s, 30s, and 60s, respectively. Vehicles were set to travel through the scenario, and 10 of them were selected to collect driving data, including vehicle status data such as speed, distance between the vehicle and the intersection center at the time of the stop, and actual stop time. After processing this historical data, offline parameter identification of the vehicle model was performed.

[0131] Since the distance X between the vehicle and the center of the intersection, the vehicle length L, and the distance D1 between the front and rear vehicles when the vehicle stops can all be known from historical data, the model can be simplified as follows:

[0132] tp=t0+K·(td-(2D2) / V)=t0+KC; where C=(td-(2D2) / V).

[0133] That is, C is considered to be a linear function related to D2, C = A·D2+B, A=-2 / V, B=td.

[0134] As can be seen from the above formula, the relevant historical data collected in the simulation scenario, such as the parking time tr, are brought into the above formula and fitted using the least squares method to obtain the values ​​of A and B, and finally the initial parameters of the entire model.

[0135] Furthermore, by substituting the parameters V and td into the original formula, a vehicle red light waiting time prediction model can be obtained. The model parameters are input into the ECU software model, and after integrated compilation, the relevant vehicle control can be realized.

[0136] In addition, based on the error E between the actual parking time and the predicted time, the corresponding parameters obtained by previous fitting can be optimized and corrected in real time.

[0137] Specifically, if Figure 6 As shown, the traffic information processing method of the embodiment of the present invention includes a first information collection step 100 and a second information prediction step 200; wherein, the first information collection step 100 obtains first traffic flow information 001, and the first traffic flow information 001 also includes second historical traffic flow information 002 and third real-time traffic flow information 003, and the third real-time traffic flow information 003 also includes fourth geographic positioning information 004, and the fourth geographic positioning information 004 includes geographic location information of the first target vehicle Z, i.e., 060, and the first target vehicle Z, i.e., 060 is a vehicle whose forced safe parking time is to be predicted.

[0138] Further, if Figure 3and Figure 6 As shown, the second information prediction step 200 obtains the first prohibited passage time ts of the first target lane 040 at the first target intersection 050, wherein the first target intersection 050 is the first intersection with a red light, a duty post and / or a parking probability greater than zero in the direction of travel of the first target vehicle Z, i.e., 060, and the first target lane 040 is the lane through which the first target vehicle Z, i.e., 060 intends to pass through the first target intersection 050; the first prohibited passage time ts is a historical value; ts includes the parking time, statistical value or simulation value in the second historical traffic flow information 002 within a preset distance between the first target lane 040 and the first target intersection 050, and ts also includes the first red light configuration time t0 or the second forced compulsory parking time t1 of the first target lane 040; if the first prohibited passage time ts is greater than or equal to the preset first ignition off time threshold, the first ignition off time threshold is used as the first prediction value 201.

[0139] Further, if Figure 6 As shown, the first information collection step 100 of the embodiment of the present invention also obtains the maximum parking time TM at the head of the first target lane 040, where TM is the maximum parking time of vehicles within a preset distance from the first target lane 040 to the first target intersection 050 in the second historical traffic flow information 002.

[0140] Specifically, the second information prediction step 200 may further include outputting the first prediction value 201 to the vehicle controller 901 or updating the first traffic flow information 001 for further processing; wherein, if TM is greater than t0 or TM is greater than t1, t0 is used as the first prediction value 201.

[0141] Furthermore, an embodiment of the method of the present invention also includes a third start-stop control step 300; the third start-stop control step 300 obtains vehicle operation information for decision-making of the automatic start-stop function; wherein, the first prohibited passage time ts is less than or equal to the first red light configuration time t0; and its first traffic flow information 001 can be measurement or simulation data from a geographic information system GIS, a traffic information data source and / or a laboratory.

[0142] Furthermore, if the first predicted value 201 is less than the first ignition-off time threshold, the automatic start-stop function of the first target vehicle Z, ie, 060, is disabled.

[0143] Furthermore, the first traffic flow information 001 may also include road condition information obtained in real time, and its road condition information may include a flooded road section existence mark WV and a water level exceeding limit sensor emergency mark WL. When the vehicle's wading depth or the height of the wading surface waves exceeds the preset wading depth threshold or water level threshold within a preset time period, the water level exceeding limit sensor emergency mark WL is set; if the flooded road section existence mark WV and / or the water level exceeding limit sensor emergency mark WL are valid, the automatic start and stop function of the first target vehicle Z is turned off.

[0144] Specifically, its second historical traffic flow information 002 can be historical measurement data 101, and its third real-time traffic flow information 003 includes real vehicle collected data 102; its fourth geographic positioning information 004 can be satellite positioning data; its fourth geographic positioning information 004 can also include the fifth queue status information 555 of the first target vehicle Z, i.e. 060, at the first target intersection 050; the fifth queue status information 555 includes the first light-on moment T1 of the first target intersection 050, code-named 041, that is, the moment when the red light comes on and all vehicles start to slow down and stop; it also includes the second stable stop moment T2, code-named 042, that is, the moment when the first target vehicle Z, i.e. 060 stops steadily; it also includes the third pre-start moment T3, code-named 043, that is, the moment when the green light comes on and the first vehicle at the first target intersection 050 starts to start; it also includes the fourth re-start moment T4, code-named 044, that is, the moment when the first target vehicle Z, i.e. 060 starts to start.

[0145] Further, if Figure 3 As shown, based on the fifth queue state information 555, the historical data in the first traffic flow information 001 is fitted, and based on the fitted function or model, the first predicted time tp experienced by the first target vehicle Z, i.e. 060, from stopping to starting again is predicted and / or output.

[0146] Further, if Figure 6 As shown, the present method embodiment further includes a fourth model construction step 400; by obtaining the first target vehicle Z, that is, the total number of vehicles K on the first target lane 040 between 060 and the first target intersection 050,

[0147] A first prediction model based on the first traffic flow information 001 is constructed; wherein parameters of the first prediction model are obtained by fitting the first traffic flow information 001.

[0148] Specifically, its first prediction model can be a linear model C =A·D2+B, A=-2 / V, B=td; wherein, parameters A and B can be obtained by least squares fitting, V is the current traffic speed, which can be provided by the first traffic flow information 001, and D2 is the average driving distance of the traffic flow where the first target vehicle Z is located.

[0149] Further, if Figure 6 As shown, the embodiment of the method also includes a fifth dynamic optimization step 500: if tp=t0+KC is used as the first prediction model; wherein t0 is the first red light configuration time of the first target vehicle Z, i.e., the first target lane 040 where 060 is located; the first prediction error E can be obtained by comparing the actual parking time tr of the first target vehicle Z, i.e., 060, with the first predicted time tp; and the parameters of the first prediction model are corrected by the first prediction error E; wherein the correction method of the first prediction error E can adopt the gradient descent method.

[0150] like Figure 7 As shown, an embodiment of the present invention further discloses a traffic information processing device, including a first information collection unit 610 and a second information prediction unit 620; wherein, the first information collection unit 610 obtains first traffic flow information 001, and the first traffic flow information 001 also includes second historical traffic flow information 002 and third real-time traffic flow information 003, and the third real-time traffic flow information 003 also includes fourth geographic positioning information 004, and the fourth geographic positioning information 004 includes geographic location information of the first target vehicle Z, i.e., 060; similarly, the first target vehicle Z, i.e., 060 is a vehicle for which the length of time of forced safe parking is to be predicted.

[0151] Furthermore, if Figure 6 As shown, its second information prediction unit 620 obtains the first prohibited passage time ts of the first target lane 040 at the first target intersection 050, and the first target intersection 050 is the first intersection with a red light, a duty post and / or a parking probability greater than zero in the direction of travel of the first target vehicle Z, i.e., 060, and the first target lane 040 is the lane through which the first target vehicle Z, i.e., 060 intends to pass through the first target intersection 050; the first prohibited passage time ts is a historical value; the ts includes the parking time, statistical value or simulation value in the second historical traffic flow information 002 within a preset distance between the first target lane 040 and the first target intersection 050, and the ts also includes the first red light configuration time t0 or the second forced compulsory parking time t1 of the first target lane 040; if the first prohibited passage time ts is greater than or equal to the preset first ignition off time threshold, the first ignition off time threshold is used as the first prediction value 201.

[0152] Furthermore, if Figure 7-11As shown, the traffic information processing device of this embodiment, wherein the first information acquisition unit 610 can also obtain the maximum parking time TM at the head of the queue of the first target lane 040, where TM is the maximum parking time of vehicles within a preset distance from the first target lane 040 to the first target intersection 050 in the second historical traffic flow information 002; its second information prediction unit 620 can also output the first prediction value 201 to the vehicle controller 901 or update the first traffic flow information 001 for further processing; wherein, if TM is greater than t0 or TM is greater than t1, t0 is used as the first prediction value 201.

[0153] Furthermore, an embodiment of the present device may also include a third start-stop control unit 630; the third start-stop control unit 630 obtains vehicle operation information for decision-making of the automatic start-stop function; wherein, the first prohibited passage time ts is less than or equal to the first red light configuration time t0; the first traffic flow information 001 comes from the geographic information system GIS, the traffic information data source and / or laboratory simulation data; if the first predicted value 201 is less than the first ignition off time threshold, the automatic start-stop function of the first target vehicle Z, i.e., 060, is disabled.

[0154] Furthermore, its first traffic flow information 001 may also include road condition information obtained in real time, which includes a flooded road section existence mark WV and a water level exceeding limit sensor emergency mark WL. When the vehicle's wading depth or the height of the wading surface waves exceeds the preset wading depth threshold or water level threshold within a preset time period, the water level exceeding limit sensor emergency mark WL is set; if the flooded road section existence mark WV and / or the water level exceeding limit sensor emergency mark WL are valid, the automatic start and stop function of the first target vehicle Z, i.e., 060, is turned off.

[0155] Specifically, if Figure 1 ,like Figure 4 As shown, the second historical traffic flow information 002 may be historical measurement data 101 , the third real-time traffic flow information 003 may be actual vehicle collected data 102 ; and the fourth geographic positioning information 004 may be satellite positioning data.

[0156] Among them, such as Figure 3 As shown, its fourth geographic positioning information 004 may also include the fifth queue status information 555 of the first target vehicle Z, i.e. 060, at the first target intersection 050; its fifth queue status information 555 may be the first lighting moment T1 of the first target intersection 050, code-named 041, i.e. the moment when the red light comes on and all vehicles start to slow down and stop; it may also include the second stable stop moment T2, code-named 042, i.e. the moment when the first target vehicle Z, i.e. 060 comes to a stable stop; it may also include the third pre-start moment T3, code-named 043, i.e. the moment when the green light comes on and the first vehicle at the first target intersection 050 starts to start; it also includes the fourth re-start moment T4, code-named 044, i.e. the moment when the first target vehicle Z, i.e. 060 starts to start.

[0157] Specifically, based on the fifth queue status information 555, the historical data in the first traffic flow information 001 can be fitted, and based on the fitted function or model, the first predicted time tp of its first target vehicle Z, that is, 060, from stopping to starting again can be predicted and / or output.

[0158] Further, if Figure 7 As shown, the embodiment of the device of the present invention may also include a fourth model construction unit 640; by obtaining the first target vehicle Z, that is, the total number of vehicles K on the first target lane 040 between 060 and the first target intersection 050, and constructing a first prediction model based on the first traffic flow information 001; wherein, the parameters of the first prediction model can be obtained after fitting the first traffic flow information 001.

[0159] Specifically, the first prediction model can be a linear model C = A·D2+B, A=-2 / V, B=td; wherein, parameters A and B can be obtained by least squares fitting, V is the current traffic speed provided by the first traffic flow information 001, and D2 is the average distance between vehicles.

[0160] Further, if Figure 7 As shown, the embodiment of the device of the present invention may also include a fifth dynamic optimization unit 650: and use the first prediction model tp=t0+KC for optimization; wherein t0 is the first red light configuration time of the first target lane 040 of the first target vehicle Z, i.e. 060.

[0161] Specifically, the first prediction error E can be obtained by comparing the actual parking time tr of the first target vehicle Z, i.e., 060, with the first predicted time tp; and the parameters of the first prediction model can be corrected using the first prediction error E; wherein, the correction method of the first prediction error E can be the gradient descent method.

[0162] Further, if Figure 2 、 Figures 8 to 11 As shown, an embodiment of the present invention further provides a computer storage medium, a controller, and a vehicle structure; the storage medium includes a storage medium body for storing a computer program; so that when the computer program is executed by a microprocessor, the relevant method disclosed in the present invention can be implemented; similarly, the controller includes any of the above-mentioned traffic information processing devices. If the first prediction value 201 is less than the preset threshold, the automatic start-stop function of the first target vehicle Z, that is, 060, must be disabled, or the vehicle's start and / or secondary start function must be prohibited; the secondary start here is defined as the process of re-igniting the engine after it is turned off from the starting state; in addition, a special operating condition detection module can be used, and the special operating condition detection module can be used to detect operating conditions to be avoided that have the risk of secondary start, such as wading conditions.

[0163] like Figure 7-11 As shown, vehicles using the above-mentioned devices, storage media or controllers naturally fall within the scope of protection of the present invention, and their information processing processes and methods are similar to those of the above-mentioned product embodiments, which will not be repeated here.

[0164] The prediction method and product based on the embodiments of the present invention can combine information provided by the map provider with the original functions of the controller, such as the ECU, to reasonably control the vehicle's start and stop functions. Based on the traffic intersection model and historical parking data, a predictive relationship between vehicle network V2X (vehicle to X) information and parking time can be established.

[0165] Furthermore, through online correction, the predicted parking time of the first target vehicle Z can be obtained, so that the developer or driver's threshold for parking time can be calibrated. If the parking time is less than the threshold, the automatic start-stop function is prohibited, thereby effectively reducing invalid parking during driving, saving some fuel and improving the driver's driving experience.

[0166] like Figure 5 As shown in , after the simulation model obtains relevant data, a part of it is selected as the verification data set, and the remaining is used as the sample set to fit the corresponding parameters, which are then substituted into the prediction model; then the verification set is selected for verification, and its verification results can be obtained; as shown in Figure 5 As shown, the embodiment of the present invention preliminarily sets parking time of less than 10 seconds as invalid parking based on driver experience, sets 10 seconds as the threshold, and considers the prediction correct if both the predicted time and the actual parking time are greater than or less than 10 seconds. The algorithm accuracy rate is 86.87% based on standard verification.

[0167] It should be noted that the above embodiments are only for the purpose of more clearly illustrating the technical solutions of the present invention. Those skilled in the art will understand that the implementation methods of the present invention are not limited to the above contents, and obvious changes, replacements or substitutions based on the above contents do not exceed the scope covered by the technical solutions of the present invention; other implementation methods will also fall within the scope of the present invention without departing from the concept of the present invention.

Claims

1. A traffic information processing method, characterized in that: include: A first information collection step (100), a second information prediction step (200); wherein the first information collection step (100) acquires first traffic flow information (001), the first traffic flow information (001) including second historical traffic flow information (002) and third real-time traffic flow information (003), the third real-time traffic flow information (003) including fourth geographic positioning information (004), the fourth geographic positioning information (004) including geographic location information of a first target vehicle Z (060), the first target vehicle Z (060) being a vehicle for which a forced safe parking time is to be predicted; the fourth geographic positioning information also includes fifth queue status information of the first target vehicle Z at the first target intersection; The second information prediction step (200) obtains a first prohibited passage time ts of the first target lane (040) at the first target intersection (050), wherein the first target intersection (050) is the first intersection with a red light, a duty post and / or a parking probability greater than zero in the direction of travel of the first target vehicle Z (060), and the first target lane (040) is the lane through which the first target vehicle Z (060) intends to pass through the first target intersection (050); the first prohibited passage time ts is a historical value; ts includes the parking duration, statistical value or simulation value of the first target lane (040) within a preset distance from the first target intersection (050) in the second historical traffic flow information (002), and ts also includes the first red light configuration time t0 or the second forced compulsory parking time t1 of the first target lane (040); If the first prohibited passage time ts is greater than or equal to a preset first flameout time threshold, the first flameout time threshold is used as a first prediction value (201); Based on the fifth queue state information, historical data in the first traffic flow information is fitted, where the historical data is parameter A and parameter B, where A = -2 / V and B = td; wherein V represents the current traffic speed and td represents the average start time. Constructing a first prediction model based on the first traffic flow information, and using the first prediction model obtained by fitting as a basis, predicting and / or outputting a first predicted time tp, i.e., a first predicted value, that the first target vehicle Z will experience from stopping to starting again; If the first predicted value is less than the first ignition-off time threshold, the automatic start-stop function of the first target vehicle Z is disabled.

2. The traffic information processing method according to claim 1, wherein: The first information collection step (100) further obtains the maximum parking time TM of the first target lane (040), where TM is the maximum parking time of vehicles within a preset distance from the first target lane (040) to the first target intersection (050) in the second historical traffic flow information (002); The second information prediction step (200) further includes outputting the first prediction value (201) to the vehicle controller (901) or updating the first traffic flow information (001) for further processing; wherein, if TM is greater than t0 or TM is greater than t1, t0 is used as the first prediction value (201).

3. The traffic information processing method according to claim 1, further comprising: A third start-stop control step (300); The third start-stop control step (300) obtains vehicle operation information for decision-making of the automatic start-stop function; wherein the first prohibited passage time ts is less than or equal to the first red light configuration time t0; and the first traffic flow information (001) comes from a geographic information system GIS, a traffic information data source and / or laboratory simulation data.

4. The traffic information processing method according to claim 3, wherein: The first traffic flow information (001) also includes road condition information acquired in real time, the road condition information including a flooded road section presence flag WV and a water level overlimit sensor emergency flag WL, wherein when the wading depth of a vehicle or the height of the wading surface waves exceeds a preset wading depth threshold or water level threshold within a preset time period, the water level overlimit sensor emergency flag WL is set; If the waterlogged road section has a mark WV and / or the water level over-limit sensor emergency mark WL is valid, the automatic start-stop function is turned off.

5. The traffic information processing method according to claim 3, wherein: The second historical traffic flow information (002) includes historical measurement data (101), the third real-time traffic flow information (003) includes real vehicle collected data (102); the fourth geographic positioning information (004) includes satellite positioning data; The fourth geographic positioning information (004) further includes fifth queue status information (555) of the first target vehicle Z (060) at the first target intersection (050); The fifth queue status information (555) includes: the first light-on time T1 (041) of the first target intersection (050), that is, the time when the red light comes on and all vehicles start to slow down and stop; the second stable stop time T2 (042), that is, the time when the first target vehicle Z (060) comes to a stable stop; the third pre-start time T3 (043), that is, the time when the green light comes on and the first vehicle at the first target intersection (050) starts to start; and the fourth re-start time T4 (044), that is, the time when the first target vehicle Z (060) starts to start.

6. The traffic information processing method according to claim 5, further comprising a fourth model construction step (400); Obtaining a total number K of vehicles on the first target lane (040) between the first target vehicle Z (060) and the first target intersection (050), where K is a natural number; Constructing a first prediction model based on the first traffic flow information (001); wherein, The parameters of the first prediction model are obtained by fitting the first traffic flow information (001); The first prediction model includes a linear model C=A·D2+B, A=-2 / V, B=td; wherein parameters A and B are obtained by least squares fitting, V is the current vehicle speed provided by the first traffic flow information (001), td represents the general average start time, and D2 is the average distance between vehicles traveling.

7. The traffic information processing method according to claim 6, further comprising a fifth dynamic optimization step (500): Take tp=t0+KC as the first prediction model; wherein, t0 is the first red light configuration time of the first target vehicle Z (060) and the first target lane (040); Compare the actual parking time tr of the first target vehicle Z (060) with the first predicted time tp to obtain a first prediction error E; correct the parameters of the first prediction model with the first prediction error; wherein the correction method of the first prediction error E includes a gradient descent method.

8. A traffic information processing device, comprising: A first information collection unit (610) and a second information prediction unit (620); The first information collection unit (610) acquires first traffic flow information (001), the first traffic flow information (001) including second historical traffic flow information (002) and third real-time traffic flow information (003), the third real-time traffic flow information (003) including fourth geographic positioning information (004), the fourth geographic positioning information (004) including geographic location information of a first target vehicle Z (060), the first target vehicle Z (060) being a vehicle for which a forced safe parking time is to be predicted; the fourth geographic positioning information also includes fifth queue status information of the first target vehicle Z at the first target intersection; The second information prediction unit (620) obtains a first prohibited passage time ts of the first target lane (040) at the first target intersection (050), wherein the first target intersection (050) is the first intersection with a red light, a duty post and / or a parking probability greater than zero in the direction of travel of the first target vehicle Z (060), and the first target lane (040) is the lane through which the first target vehicle Z (060) intends to pass through the first target intersection (050); the first prohibited passage time ts is a historical value; ts includes the parking time, statistical value or simulation value of the first target lane (040) within a preset distance from the first target intersection (050) in the second historical traffic flow information (002), and ts also includes the first red light configuration time t0 or the second forced compulsory parking time t1 of the first target lane (040); if the first prohibited passage time ts is greater than or equal to a preset first ignition off time threshold, the first ignition off time threshold is used as the first prediction value (201); Based on the fifth queue state information, historical data in the first traffic flow information is fitted, where the historical data is parameter A and parameter B, where A = -2 / V and B = td; wherein V represents the current traffic speed and td represents the average start time. Constructing a first prediction model based on the first traffic flow information, and using the first prediction model obtained by fitting as a basis, predicting and / or outputting a first predicted time tp, i.e., a first predicted value, that the first target vehicle Z will experience from stopping to starting again; If the first predicted value is less than the first ignition-off time threshold, the automatic start-stop function of the first target vehicle Z is disabled.

9. The traffic information processing device according to claim 8, wherein: The first information collection unit (610) further obtains a maximum parking time TM at the head of the first target lane (040), where TM is the maximum parking time of vehicles within a preset distance from the first target lane (040) to the first target intersection (050) in the second historical traffic flow information (002); The second information prediction unit (620) further outputs the first prediction value (201) to the vehicle controller (901) or updates the first traffic flow information (001) for further processing; wherein, if TM is greater than t0 or TM is greater than t1, t0 is used as the first prediction value (201).

10. The traffic information processing device according to claim 8, further comprising: a third start-stop control unit (630); The third start-stop control unit (630) obtains vehicle operation information for decision-making of the automatic start-stop function; wherein the first prohibited passage time ts is less than or equal to the first red light configuration time t0; and the first traffic flow information (001) is obtained from a geographic information system GIS, a traffic information data source and / or laboratory simulation data; If the first predicted value (201) is less than the first ignition-off time threshold, disabling the automatic start-stop function of the first target vehicle Z (060); The first traffic flow information (001) also includes road condition information obtained in real time, and the road condition information includes a flooded road section existence flag WV and a water level exceeding limit sensor emergency flag WL. When the vehicle wading depth or the height of the wading surface wave exceeds a preset wading depth threshold or water level threshold within a preset time period, the water level exceeding limit sensor emergency flag WL is set; if the flooded road section existence flag WV and / or the water level exceeding limit sensor emergency flag WL are valid, the automatic start-stop function is turned off.

11. The traffic information processing device according to claim 10, wherein: The second historical traffic flow information (002) includes historical measurement data (101), the third real-time traffic flow information (003) includes real vehicle collected data (102); the fourth geographic positioning information (004) includes satellite positioning data; The fourth geographic positioning information (004) further includes fifth queue status information (555) of the first target vehicle Z (060) at the first target intersection (050); The fifth queue state information (555) includes: the first light-on time T1 (041) of the first target intersection (050), that is, the time when the red light turns on and all vehicles start to slow down and stop; the second stable stop time T2 (042), that is, the time when the first target vehicle Z (060) stops steadily; the third pre-start time T3 (043), that is, the time when the green light turns on and the first vehicle at the first target intersection (050) starts to start; and the fourth re-start time T4 (044), that is, the time when the first target vehicle Z (060) starts to start. Based on the fifth queue state information (555), historical data in the first traffic flow information (001) is fitted, and based on the fitted function or model, a first predicted time tp that the first target vehicle Z (060) takes from stopping to starting again is predicted and / or output.

12. The traffic information processing device according to claim 11, further comprising a fourth model construction unit (640); By obtaining the total number K of vehicles on the first target lane (040) between the first target vehicle Z (060) and the first target intersection (050), where K is a natural number; Constructing a first prediction model based on the first traffic flow information (001); wherein, The parameters of the first prediction model are obtained by fitting the first traffic flow information (001); The first prediction model includes a linear model C=A·D2+B, A=-2 / V, B=td; wherein parameters A and B are obtained by least squares fitting, V is the current vehicle speed provided by the first traffic flow information (001), td represents the general average start time, and D2 is the average distance between vehicles traveling.

13. The traffic information processing device according to claim 12, further comprising a fifth dynamic optimization unit (650): Take tp=t0+KC as the first prediction model and optimize it; t0 is the first red light configuration time of the first target vehicle Z (060) and the first target lane (040); By comparing the actual parking time tr of the first target vehicle Z (060) with the first predicted time tp, a first prediction error E is obtained; the parameters of the first prediction model are corrected using the first prediction error E; wherein the correction method of the first prediction error E includes a gradient descent method.

14. A computer storage medium comprising: A storage medium for storing a computer program; when the computer program is executed by a microprocessor, the traffic information processing method according to any one of claims 1 to 7 is implemented.

15. A controller comprising: The traffic information processing device according to any one of claims 8 to 13; If the first prediction value (201) is less than a preset threshold, the automatic start-stop function of the first target vehicle Z (060) is disabled, the vehicle is prohibited from starting and / or the vehicle is prohibited from secondary starting; the secondary starting is defined as the process of re-igniting the engine after it is turned off from the starting state; It also includes a special operating condition detection module, which is used to detect the operating condition to be avoided in which the secondary startup risk exists; the operating condition to be avoided includes a wading condition.

16. A vehicle comprising: The traffic information processing device according to any one of claims 8 to 13; and / or the storage medium according to claim 14; And / or a controller as claimed in claim 15.

Images