A method and device for matching the historical trajectory of the vehicle in front in a V2V environment

Abstract

A method for matching the historical trajectory of the vehicle in front on a curve in a V2V environment, comprising the following steps: S1, the vehicle in front obtains the actual historical trajectory of the vehicle in front of it and calculates its actual historical trajectory; S2, the present vehicle performs path matching with the historical trajectory of the vehicle in front; S3, after the present vehicle matches the actual historical trajectory of the vehicle in front, the actual distance from the true position of the vehicle in front is calculated through the historical trajectory of the vehicle in front, and the earliest and latest safety warning distances are calculated based on this actual distance. Through the real-time information interaction between vehicles in a vehicle-to-vehicle collaborative environment, when the vehicle is about to enter a curve scenario, the present vehicle can obtain the historical trajectory point information of the vehicle in front. After calculation, storage and broadcast transmission, the present vehicle can perform map matching on the historical trajectory information of the vehicle in front, judge the positional relationship with the vehicle in front, and provide relevant driving warning services for the safety of the present vehicle.

Description

Technical Field

[0001] The present invention relates to the field of traffic guidance, and specifically relates to a method for matching the historical trajectory of the vehicle in front in a V2V environment, and a device for executing the method. In a vehicle-to-vehicle cooperation environment, when a vehicle is about to enter a curved road section, an auxiliary driving device for the curved road section is provided. After the vehicle in front enters the curve, when the vehicle enters the curve in the same lane in the same direction, curve auxiliary driving services are provided for the driver in the curve scenario. Background Art

[0002] Currently, the active safety technologies include the Collision Warning System (CWS) and the Autonomous Emergency Braking (AEB) system. Both of these systems detect the vehicle or obstacle in front through a camera or radar, use the collected data of the vehicle in front and the vehicle itself as the system input, calculate, analyze and judge through algorithms, and finally output a warning decision. Since the radar or video sensor used can only scan objects within a certain area range, when driving in a curve, it cannot accurately identify and will generate false alarms and missed alarms. These sensors have high requirements for light, and misjudgment will occur under strong light, low light, with obstacles and bad weather conditions.

[0003] Vehicle-road cooperation technology can effectively improve the problems of false alarms and missed alarms caused by environmental interference and limited self-detection distance. By making full use of the low latency of data exchange in vehicle-road cooperation technology and fully mining the effective data of vehicles, the driving safety of vehicles can be further improved. However, in a vehicle-to-vehicle cooperation environment, there is currently no technical solution that uses the historical trajectory of the vehicle in front to provide warnings or other services for the vehicle behind: for example, Patent CN106004873A stores and sends the historical trajectory of the vehicle in a curve scenario to the roadside device, and the roadside device analyzes and models according to a large amount of historical trajectory data, and provides curve warning services for subsequent vehicles according to the road environment, vehicle type, etc.; another example is Patent CN114013506A, which stores its own vehicle trajectory data, compares and analyzes the historical curve trajectory stored in the vehicle itself, determines whether the vehicle enters the curve, and provides curve warning services for the vehicle.

[0004] The device of the present invention utilizes the V2X communication ability of the vehicle to provide a method for storing its own historical trajectory, broadcasts the historical trajectory to surrounding vehicles, and the vehicle behind will judge the position relationship according to its trajectory to provide services for the vehicle behind. Summary of the Invention

[0005] According to the problems raised in the background art, the present invention provides a method for matching the historical trajectory of the vehicle in front in a V2V environment to solve, and the following further elaborates on the present invention.

[0006] A method for matching the historical trajectory of the vehicle in front in a V2V environment, comprising the following steps:

[0007] S1. The vehicle in front obtains the actual historical trajectory of the vehicle in front of it and calculates its actual historical trajectory, including the following steps;

[0008] S101. Initialize parameters, and select the starting point P in the actual historical trajectory of the vehicle starting = P i-2 , the previous point P previous = P i-1 , and the next point P next = P i ;

[0009] S102. Calculate the chord length distance between points P starting and P next . If it is greater than the threshold, directly enter S104;

[0010] S103. Calculate the vertical distance ErrorDist from the actual trajectory point to the chord of the historical trajectory sampling point;

[0011] S104. If ErrorDist is greater than the allowable error range or the chord length distance is greater than the threshold, store the point P previous , and update P starting , P next and P previous coordinates of the three points: P starting = P i-1 , P previous = P i , P next = P i+1 , and then enter S405;

[0012] If ErrorDist is less than the allowable error range, update the coordinates of P next and P previous : P previous = P i , P next = P i+1 ;

[0013] S105. Calculate the cumulative distance totalDist between the actual trajectory points. If the cumulative distance is greater than the preset threshold, delete the last point until totalDist is less than the preset threshold; if the number of stored points exceeds the storage length, delete the last historical trajectory point to ensure that the stored historical trajectory always remains at the target length, and then enter S102 to calculate the actual historical trajectory of the vehicle in front in a loop;

[0014] S2. Perform path matching between the historical trajectories of this vehicle and the vehicle ahead. The matching steps are as follows:

[0015] S201. Screen out the vehicle ahead of the HV of this vehicle;

[0016] S202. Screen out the point closest to this vehicle from the historical trajectory of the vehicle ahead, and calculate the direction angle between this point (lat1, lon1) and the GPS point (lat2, lon2) of this vehicle, and determine the relationship between the vehicle's heading angle and the azimuth angle courseDegree. The calculation method of the direction angle is as follows:

[0017]

[0018] If courseDegree < 0, then courseDegree = courseDegree + 360;

[0019] Suppose the calculated closest trajectory point is P i point, and the calculated angle is α. Judge the relationship between the azimuth angle and the heading angle of point P i :

[0020] If α - HVheading < 90, it means that point P i is behind the vehicle, then select P i+1 point as the point in front of the vehicle HV;

[0021] If α - HVheading > 90, it means that point P i is in front of the vehicle, then select P i-1 point as the point behind the vehicle HV;

[0022] S203. Calculate the azimuth angle β between P i+1 and the HV GPS point:

[0023] If β - heading > 90, it means that this point is the point in front of the vehicle, which conforms to the speculation in S202;

[0024] If β - heading < 90, it means that no valid point is obtained, and exit this calculation;

[0025] S204. Judge whether the direction angle γ of the front and rear points is consistent with the heading angle of this vehicle;

[0026] S205. Calculate the perpendicular distance from the GPS point of this vehicle to the line connecting the vehicle's front trajectory point P i and P i+1 points. If the perpendicular distance is less than the threshold, it means that the vehicle has successfully matched the trajectory of the vehicle ahead; if the trajectory of the vehicle ahead is not matched, the matching fails;

[0027] S3. After the vehicle matches the actual historical trajectory of the vehicle ahead, calculate the actual distance from the true position of the vehicle ahead based on the historical trajectory of the vehicle ahead. Based on this actual distance and the earliest and latest safety warning distances of the vehicle calculated according to GB / T33577, when the actual path distance Dist is between the maximum and minimum safety distances, trigger a forward collision warning.

[0028] A device for matching the historical trajectory of the vehicle ahead in a V2V environment, equipped with an on-vehicle unit OBU of the vehicle, including:

[0029] A data acquisition module, including: a GNSS high-precision positioning module capable of acquiring lane-level high-precision positioning information; a status information acquisition module for real-time acquisition of its own status information, including vehicle longitude and latitude, speed, heading angle and other information; an optional vehicle controller area network module information for real-time acquisition of the vehicle's current braking information, steering wheel angle information, anti-lock braking system information, electronic stability program information of the vehicle body, traction controller and other information;

[0030] A communication module for sending and receiving the vehicle's own status data collected through the V2V method to and from other nearby vehicles;

[0031] A message center module for parsing the received information of the vehicle ahead, the information collected by the vehicle's sensors, and the messages to be sent after being processed by the intelligent analysis module, etc.;

[0032] An intelligent analysis module for analyzing and storing its own vehicle information, making decisions on the information of surrounding vehicles, and sending the warning strategy to the message center module;

[0033] The APP module includes a warning reminder module for displaying and broadcasting the warning information sent by the algorithm module, and a human-machine interaction module for displaying the basic status information of the vehicle.

[0034] Beneficial effects: Compared with the prior art, through the real-time information interaction between vehicles in a vehicle-to-vehicle collaborative environment, when the vehicle is about to enter a curved road scenario, the invention obtains the historical trajectory point information of the vehicle ahead. After calculation, storage and broadcast, the vehicle can perform map matching on the historical trajectory information of the vehicle ahead, judge the positional relationship with the vehicle ahead, and provide relevant driving warning services for the safety of the vehicle. Description of the Drawings

[0035] Figure 1 : Schematic diagram of the modules of the device of the present invention;

[0036] Figure 2 : Schematic diagram of vehicle trajectory sampling points and actual trajectory points;

[0037] Figure 3:Schematic diagram of trajectory point selection method;

[0038] Figure 4 :Flowchart of historical trajectory screening method;

[0039] Figure 5 :Schematic diagram of historical trajectory path matching. Specific implementation manner

[0040] Next, a specific embodiment of the present invention will be elaborated in detail with reference to the accompanying drawings.

[0041] Refer to the attached Figure 1 Figure, a device for matching the historical trajectory of the vehicle in front in a curve based on the V2V environment (hereinafter referred to as "the device of the present invention") is provided with a vehicle on-board unit OBU, and the on-board unit can broadcast and receive vehicle information outward; wherein, the on-board device OBU includes:

[0042] Data acquisition module, including: GNSS high-precision positioning module, which can collect lane-level high-precision positioning information; status information acquisition module, which is used to obtain its own status information in real time, including vehicle longitude and latitude, speed, heading angle and other information; optional vehicle controller area network (Controller Area Network, CAN) module information, which can obtain the current braking information, steering wheel angle information, antilock brake system (Antilock Brake System, ABS) information, electronic stability program (Electronic Stability Program, ESP) information, traction controller (Traction Control System, TSC) and other information of the vehicle in real time;

[0043] Communication module, which is used to send and receive the collected vehicle own status data to other nearby vehicles in a V2V manner;

[0044] Message center module, which is used to analyze the received information of the vehicle in front, the information collected by the sensors of the vehicle itself, and the messages to be sent after being processed by the intelligent analysis module, etc.;

[0045] Intelligent analysis module, which is used to analyze and store the information of its own vehicle and make decisions on the information of surrounding vehicles, and send the early warning strategy to the message center module;

[0046] APP module, including an early warning reminder module, which is used to display and broadcast the early warning information sent by the algorithm module, and a man-machine interaction module, which is used to display the basic status information of the vehicle itself.

[0047] In the vehicle networking environment, the BSM (Base Safe Message) basic safety message set can be transmitted through V2X technology. Relevant safety messages of the vehicle itself, such as the historical trajectory of the vehicle itself, can be carried in the BSM basic safety message set.

[0048] In the present invention, the GNSS high-precision positioning module of the data acquisition module records the effective GNSS trajectory points collected during the current vehicle positioning, stores the effective GNSS trajectory points in order in a list, and stores the trajectory list in the corresponding data structure. Under the traditional recording method, the historical trajectory of the vehicle will have overly dense trajectories on straight roads, resulting in waste of memory space, while in the curved road scenario, there will be sparse trajectory points, and the driving trajectory characteristics of the vehicle cannot be accurately represented. Compared with the traditional recording method, a method for screening trajectories based on time and distance constraints is provided in the intelligent analysis module of this device, which can effectively screen the historical trajectory points of the vehicle, making the selected trajectory points have certain representativeness and practicality.

[0049] Specifically, referring to the attached Figure 2 , what is shown is the set of trajectory points of the vehicle driving in a curved road. Among them, the hollow points represent the subset of trajectory point samples during the actual driving process of the vehicle, the solid points represent the historical trajectory sampling points, and the chord length between two consecutive black points approximately represents the approximate value of the vehicle trajectory driving section. For the selected historical trajectory sampling points, it is required to ensure that the vertical distance from the path point on the actual vehicle to the chord of the trajectory sampling point is less than ErrorDist. At the same time, the size of the historical trajectory sampling points of the vehicle in the buffer area can be adjusted. When driving in a curved road, it is necessary to meet both the requirements of the lateral distance of the vehicle and the requirement that the chord length meets the fixed-length distance MinFixDist.

[0050] Assume that the vehicle path is composed of the two closest preceding vehicle trajectory points of the vehicle. ErrorDist is defined as the vertical distance from the actual trajectory point to the chord of the historical trajectory sampling point. As the actual trajectory points of the vehicle continue to increase, the trajectory points on the actual path may become historical trajectory sampling points.

[0051] As Figure 3 shown, for the actual trajectory points P1, P2, P3, etc. of the vehicle, ErrorDist will change according to the position of the points selected in the curved road path. When ErrorDist exceeds the set threshold, the points that meet the threshold conditions will be stored as the latest historical trajectory points, so that the historical trajectory points include the trajectory points of the current position of the vehicle. As Figure 3 shown, the actual chord length is defined as the actual chord length distance ChordLength between the GNSS points in two historical trajectories of the vehicle trajectory in the curved road path. ChordLength is the distance between two GNSS trajectory points, and each trajectory point is represented by its corresponding latitude and longitude.

[0052] For the trajectory point P1(lon1, lat1), lon1 and lat1 represent its longitude and latitude. Similarly for P2 and P3, and all are in radians. The radius of the Earth's equator is R Earth . Then the calculation formula for the actual distance of the chord is as follows:

[0053] ChordLength = R Earth * cos -1 [cos(lat1)cos(lat2)cos(lon2 - lon1) + sin(lat1)sin(lat2)];

[0054] Where R Earth = 6378137m.

[0055] The specific calculation steps are as follows:

[0056] Step S401: Initialize parameters. There are at least three points in the actual historical trajectory of the vehicle. Select the starting point P starting = P i-2 , the previous point P previous = P i-1 , and the next point P next = P i , and store the first trajectory point P starting in the cache list PathHistoryList, totalDist = 0.

[0057] Step S402: Calculate the chord length distance between the P_starting and P_next points respectively according to the calculation method of ChordLength. If ChordLength is greater than the threshold CHRODLENGTH, directly enter S404.

[0058] CHRODLENGTH is the distance threshold between two trajectory points. For example, if the maximum storage length of historical trajectory points in a 100m trajectory is 10, then the distance threshold for this historical trajectory point can be set to 10m.

[0059] Step S403: Calculate ErrorDist. The distance calculation formula can first convert the longitude and latitude coordinates into plane coordinates through the Mercator longitude and latitude coordinate conversion formula, and then calculate the distance from a point to a line through the three - point coordinates using Heron's formula. The calculation method is as follows:

[0060] Mercator conversion formula:

[0061] x = 2π * R earth * lon / 180;

[0062] y = 2π * R earth*log(tan((90+lat)*π / 360)) / (π / 180);

[0063] where lon is the longitude value, lat is the latitude value, and R earth is the radius of the Earth's equatorial ellipsoid, in m.

[0064] Heron's formula:

[0065] P = (ChordLength1 + ChordLength2 + ChordLength3) / 2;

[0066]

[0067] where: ChordLength1 is the chord length distance between P starting and P previous points, ChordLength2 is the chord length distance between P previous and P next points, and ChordLength3 is the chord length distance between P starting and P next points.

[0068] Step S404: If ErrorDist is greater than the allowed error range ERRORDIST or ChordLength is greater than the threshold CHRODLENGTH, then place P previous in the PathHistoryList list and update P starting , P next and P previous coordinates of the three points:

[0069] P starting = P i-1 , P previous = P i , P next = P i+1 ,

[0070] then proceed to S405;

[0071] If ErrorDist is less than the allowed error range ERRORDIST, then update P next and P previous coordinates:

[0072] P previous = P i , P next = P i+1 .

[0073] Step S405: The number of stored points is restricted by the storage space size. It is necessary to calculate the cumulative distance totalDist between each actual trajectory point. If the cumulative distance is greater than SUMDIST, then delete the last point until totalDist is less than SUMDIST; if the number of stored points exceeds the storage length, then delete the last historical trajectory point to ensure that the stored historical trajectory always remains within 10 (determined according to the storage length design), and then enter S402.

[0074] Through the above trajectory screening method, the effective historical trajectory during the vehicle driving process can be obtained. This effective historical trajectory data will be broadcast through the additional safety message field in the BSM message. The nearby vehicles will receive the BSM message sent by it and obtain the historical trajectory information of the nearby vehicles by parsing the message set information. Whether in a straight or curved road scenario, when the host vehicle receives the historical trajectory sent by the preceding vehicle, it can perform path matching through this historical trajectory to determine whether the driving trajectory of the host vehicle matches that of the preceding vehicle. If the trajectories match, it means that the host vehicle and the preceding vehicle belong to the same lane, and it can effectively and accurately make judgments for some dangerous collision scenarios.

[0075] Taking the forward collision warning scenario FCW as an example, when the host vehicle does not receive the historical trajectory of the preceding vehicle, the host vehicle can calculate the position relationship with itself through information such as the position, speed, and heading angle of the preceding vehicle. The lateral distance in the direction where the heading angle of the preceding vehicle is located relative to the host vehicle, the longitudinal distance under the actual longitude and latitude conditions of the vehicle, and the safety distance calculated through the relative vehicle speed. According to common sense, when the preceding vehicle enters a curve, the lateral distance between the vehicles will become larger, resulting in the lateral distance between the vehicles exceeding the threshold for matching in the same lane, making it impossible for the host vehicle to accurately determine whether the host vehicle and the preceding vehicle are in the same lane, resulting in an unsatisfactory trigger in the curved road scenario and the situation of no warning. Based on this, this device effectively avoids the situation of calculation errors, false alarms, or missed alarms caused by the jitter of the vehicle heading angle by using the historical trajectory to judge the vehicle's current position relationship.

[0076] According to the foregoing, to effectively improve the judgment accuracy of the vehicle position relationship and solve the problem of no warning in the curved road scenario, the premise is to perform path matching on the historical trajectories of the current vehicle and the vehicle in front. This embodiment provides a path matching method. When the host vehicle receives the historical trajectory of the vehicle in front, the host vehicle calculates the position relationship between the current position of the vehicle and the position in the historical trajectory. The calculation steps are as follows:

[0077] S501: First, screen out the vehicle in front of HV.

[0078] S502: Screen the point closest to the host vehicle from the historical trajectory of the leading vehicle, calculate the direction angle between this point (lat1, lon1) and the HV GPS point (lat2, lon2), and determine the relationship between the vehicle heading angle and the azimuth angle. The calculation method of the direction angle courseDegree is as follows:

[0079]

[0080] If courseDegree < 0, then courseDegree = courseDegree + 360;

[0081] As Figure 5 shown, if the calculated closest trajectory point P i point, the calculated angle is α, determine the relationship between the azimuth angle and the heading angle of point P i :

[0082] If α - HVheading < 90, it means that vehicle P i point is behind the vehicle, then select P i+1 point as the point in front of vehicle HV;

[0083] If α - HVheading > 90, it means that vehicle P i point is in front of the vehicle, then select P i-1 point as the point behind vehicle HV.

[0084] S503: Calculate the azimuth angle β between P i+1 and the HV GPS point,

[0085] If β - heading > 90, it means that this point is the point in front of the vehicle, which conforms to the speculation in Step502;

[0086] If β - heading < 90, it means that no valid point is obtained, and this calculation is exited.

[0087] S504: Determine whether the direction angle γ between the front and rear points is consistent with the host vehicle heading angle.

[0088] S505: Calculate the perpendicular distance from the HV GPS point to the line connecting the leading vehicle trajectory points P i and P i+1 points. If the perpendicular distance is less than the threshold MATCHTHRESHOLD, it means that the vehicle has successfully matched the leading vehicle trajectory; if the leading vehicle trajectory is not matched, the matching fails.

[0089] After the vehicle matches the trajectory of the vehicle in front, the actual distance from the vehicle to the true position of the vehicle in front can be calculated through the historical trajectory of the vehicle in front. The actual distance is composed of the distance from the vehicle HV to the front point that matches the historical trajectory to the actual position of the HV, the distance from the front point that matches to the latest point of the historical trajectory, and the distance from the latest historical trajectory point to the real-time position of the vehicle in front. The calculation method between trajectory points is the same as the above method. The arc length is approximately equal to the chord length between trajectory points, and the actual path distance Dist between vehicles is calculated through the chord length distance between points.

[0090] Taking forward collision warning as an example, when there is an application scenario where a collision is likely to occur, sufficient reaction time should be reserved for the driver to ensure that the vehicle can brake safely and avoid collision accidents. Triggering the warning too early or too late will result in a poor driving experience and may even lead to traffic accidents. A reasonable reaction timing is crucial. Referring to GB / T33577, the average reaction time T of Chinese drivers is between 0.3s and 2s, and the average deceleration a of the driver's braking s is 3.6m / s 2 -7.9m / s 2 . Referring to GB 7258-2012, the braking coordination time t1 ranges from 0.35s to 0.6s, the deceleration increase time t2 is 0.2s, and the safety distance d0 at rest is 3m.

[0091] Based on the above conditions, this device selects a minimum safe distance model between vehicles. Assuming that the vehicles are all in a relatively stable driving motion state and using the safety warning model, the earliest and latest safety warning distances can be known. The calculation method of the minimum safety warning is as follows:

[0092]

[0093] When the calculated actual path distance Dist of the vehicle is between the maximum and minimum safe distances, the forward collision warning can be triggered.

[0094] This device is not limited to the forward safety collision warning scenario and is also applicable to other scenarios, such as the following of a convoy when turning, the abnormality of the vehicle in front in a curve, etc. Scenarios where corresponding judgments and decisions can be made based on the trajectory of the vehicle in front.

[0095] Through the real-time information interaction between vehicles in the vehicle-to-vehicle collaborative environment, when the vehicle is about to enter the curve scenario, the historical trajectory point information of the vehicle in front is obtained. After calculation, storage and broadcast, the vehicle can perform map matching on the historical trajectory information of the vehicle in front, judge the position relationship with the vehicle in front, and provide relevant driving warning services for the safety of the vehicle.

[0096] The device of the present invention can provide stable trajectory tracking services for vehicles in both straight and curved road scenarios, without relying on map information to solve the safety collision warning in curved road scenarios, and has good applicability and popularization. The judgment of the positional relationship between vehicles is accurate and has a high precision, and it also has good practicability under sub-meter positioning conditions.

[0097] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for matching the historical trajectory of the vehicle in front in a V2V environment, characterized in that, It includes the following steps: S1. The leading vehicle obtains the actual historical trajectory of the vehicle in front of it and calculates its actual historical trajectory, which includes the following steps: S101, Initialize parameters and select a starting point in the actual historical trajectory of the vehicle , the previous point , and the next point ; S102, calculate and the chord length distance between points, if it is greater than the threshold, directly enter S104; S103. Calculate the vertical distance ErrorDist from the actual trajectory point to the chord of the historical trajectory sampling point. S104, if the ErrorDist is greater than the allowed error range or the chord length distance is greater than the threshold, then store the point , and update , and the coordinates of three points: , , , then enter S405; If ErrorDist is less than the allowed error range, update and point coordinates: , ; S105. Calculate the cumulative distance totalDist between each actual trajectory point. If the cumulative distance is greater than the preset threshold, delete the last point until totalDist is less than the preset threshold. If the number of stored points exceeds the storage length, delete the last historical trajectory point to ensure that the stored historical trajectory always remains at the target length, and then enter S102 to calculate the actual historical trajectory of the leading vehicle in a loop. S2. Perform path matching between the historical trajectories of the host vehicle and the vehicle in front. The matching steps are as follows: S201. Screen out the vehicle in front of the host vehicle's HV. S202. Select the point closest to the host vehicle from the historical trajectory of the leading vehicle, calculate the direction angle between this point (lat1, lon1) and the host vehicle's GPS point (lat2, lon2), and judge the relationship between the vehicle's heading angle and the azimuth angle courseDegree. The calculation method of the direction angle is as follows: ； If , then ; If the calculated nearest trajectory point is a point and the calculated azimuth angle is , then judge the relationship between the azimuth angle of the point and the heading angle: If , it indicates that the vehicle point is located behind the vehicle, then select point as the front point of the vehicle HV; If , it indicates that the vehicle point is in front of the vehicle, then select point as the point behind the vehicle HV; S203, calculate the azimuth angle with the HVGPS point : If , it indicates that this point is in front of the vehicle, which conforms to the speculation in S202; If , it indicates that no valid points are obtained, and this calculation is exited; S204, Determine whether the direction angles of the front and rear points are consistent with the vehicle's heading angle; S205, calculate the perpendicular distance from the GPS point of this vehicle to the trajectory point of the vehicle in front. and the point on the line connecting the points. If the perpendicular distance is less than the threshold, it indicates that the vehicle has successfully matched the trajectory of the vehicle in front; if the trajectory of the vehicle in front is not matched, the matching fails. S3. After the host vehicle matches the actual historical trajectory of the vehicle in front, calculate the actual distance from the historical trajectory of the leading vehicle to the true position of the vehicle in front. Based on this actual distance and the earliest and latest safety warning distances calculated according to GB / T33577 for the host vehicle, when the actual path distance Dist is between the maximum and minimum safety distances, trigger a forward collision warning.

2. The method according to claim 1, characterized in that, In step S102, the calculation method of the chord length distance ChordLength is as follows: Each trajectory point is represented by its corresponding latitude and longitude. For the three selected trajectory points P1(lon1, lat1), lon1 and lat1 represent its longitude and latitude, and the same applies to P2 and P3, all in radians. The radius of the Earth's equator is ; The calculation formula for the actual distance of the chord is as follows: Among them 。 3. The method according to claim 1, characterized in that, In step S103, for the calculation of ErrorDist, first convert the latitude and longitude coordinates into plane coordinates through the Mercator latitude and longitude coordinate conversion formula, and then calculate the distance from the point to the line through the coordinates of three points using Heron's formula. The calculation method is as follows: Mercator conversion formula: ； ； Among them, is the longitude value, is the latitude value, is the equatorial ellipsoid radius of the earth, with the unit of m; Heron's formula: ； ； Among them, is and the chord length distance between points, is and the distance of the chord length between points, is and the chord length distance between points.

4. An apparatus for matching the historical trajectory of the vehicle in front in a curve based on the method described in claim 1, comprising a vehicle on-board unit OBU, characterized in that, It includes: Data acquisition module, including: GNSS high-precision positioning module, which can collect lane-level high-precision positioning information; status information acquisition module, used to obtain its own status information in real time, including vehicle latitude and longitude, speed, and heading angle information; vehicle controller area network module, which obtains the vehicle's current brake information, steering wheel angle information, anti-lock braking system information, electronic stability program information of the vehicle body, and traction controller information in real time. Communication module, used to send and receive the vehicle's own status data collected through the V2V method to other nearby vehicles. Message center module, used to parse the received information of the remote vehicle, the information collected by the vehicle's own sensors, and the messages to be sent after being processed by the intelligent analysis module. Intelligent analysis module, used to analyze and store its own vehicle information, make decisions on the information of surrounding vehicles, and send the warning strategy to the message center module. APP module, including a warning reminder module, used to display and broadcast the warning information sent by the algorithm module, and a human-machine interaction module, used to display the basic status information of the vehicle.

Images