Abnormal vehicle grading early warning method and device based on V2X trajectory prediction

Abstract

The invention provides an abnormal vehicle grading early warning method and device based on V2X trajectory prediction. According to the method, at least two braking strategies are defined according to the deceleration opportunity and / or deceleration of a vehicle; predicting the tracks of the vehicle when the vehicle executes the at least two braking strategies to form a predicted track of the vehicle; predicting the trajectory of the target vehicle to form a predicted trajectory of the target vehicle; evaluating whether the vehicle and the target vehicle have a collision risk according to the vehicle prediction trajectory and the target vehicle prediction trajectory; when the assessment result indicates that the first type of collision risk exists, first-level early warning is sent out to remind a driver to prepare for braking; and when the evaluation result indicates that the second type of collision risk exists, a second-level early warning signal is sent out to prompt the driver to immediately take a braking measure. According to the invention, based on the collision risk analysis model of the trajectory, a grading early warning scheme is provided, and the predictability of early warning and the safety of driver response are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle warning design, and specifically relates to a method and device for abnormal vehicle graded warning based on V2X trajectory prediction. Background Art

[0002] As an emerging communication technology, V2X has derived a series of warning functions. Compared with ADAS, although the warning function does not have the ability to assist or replace the driver for a short time, it has played a good supplementary role to the warning function of low-level assisted driving with its beyond-visual-range capability.

[0003] The Abnormal Vehicle Warning (AVW) function is a typical V2V (vehicle-to-vehicle) function, which is of great significance in reducing the risk of relative speed difference caused by abnormal target vehicles on highways. CSAE53 defines the typical use case of this function, but it still needs some extension definition and supplementation to be applied to mass production before it can reach the level of mass production.

[0004] On the one hand, it is difficult to achieve the ideal coverage of high-precision maps provided by the roadside in a short period of time. How to design functions that are compatible with both modes with and without map coverage is a difficult problem; on the other hand, how to provide accurate warnings on curves is also one of the difficulties. Most of the mass-produced models on the market currently follow the CSAE53 scenario definition, distinguish lanes based on maps and positioning, and implement standard warning functions. There is no distinction between the function definition and the corresponding expected results when there is a map and when there is no map, and there is no trajectory-based risk assessment model. The accuracy of the warning in the case of curves cannot be guaranteed. At the same time, the timing of the warning is not clearly defined, and the severity level of the warning cannot be distinguished.

[0005] Invention patent content

[0006] In view of the above problems, the present invention provides a method and device for abnormal vehicle graded warning based on V2X trajectory prediction, which proposes a graded warning scheme based on the collision risk analysis model of the trajectory, thereby improving the predictability of the warning and the safety of the driver's response. The purpose of the present invention is achieved through the following technical solutions:

[0007] The present invention first provides an abnormal vehicle classification warning method based on V2X trajectory prediction, which includes:

[0008] Define braking strategies: define at least two braking strategies based on the vehicle's deceleration timing and / or deceleration magnitude;

[0009] Collision risk assessment: predicting the trajectory of the vehicle when executing at least two of the above braking strategies respectively to form a predicted trajectory of the vehicle; predicting the trajectory of the target vehicle to form a predicted trajectory of the target vehicle; assessing whether there is a collision risk between the vehicle and the target vehicle based on the predicted trajectory of the vehicle and the predicted trajectory of the target vehicle;

[0010] Level 1 warning: When the evaluation results indicate that the vehicle will collide with the target vehicle when executing the first braking strategy, and will not collide with the target vehicle when executing the second braking strategy, that is, when there is a first-class collision risk, a level 1 warning signal (brake preparation signal) is issued to remind the driver to prepare for braking;

[0011] Level 2 warning: When the evaluation results indicate that the vehicle will collide with the target vehicle when executing the second braking strategy, that is, when there is a second-level collision risk, a level 2 warning signal is issued to prompt the driver to take braking measures immediately;

[0012] Among them, in the same evaluation, since the urgency of the first-level warning is less than that of the second-level warning, the driver is allowed a certain amount of free reaction and judgment time under the first-level warning. Usually, the driving range allowed by executing the first braking strategy is greater than the driving range allowed by executing the second braking strategy.

[0013] Furthermore, the at least two braking strategies include:

[0014] The first braking strategy adopts a prediction model of first maintaining a constant speed and then braking (CSAE prediction model). Under this prediction model, it is assumed that the vehicle decelerates according to the first deceleration rate after traveling for the first necessary time; here, the first necessary time usually refers to the time reserved for the driver to react after receiving the warning signal.

[0015] The second braking strategy adopts a standard braking prediction model, under which it is assumed that the vehicle immediately decelerates at a first deceleration.

[0016] Furthermore, when the evaluation result is that there is a second type of collision risk, the critical deceleration when the vehicle and the target vehicle just do not collide is calculated, and when the absolute value of the critical deceleration is less than the absolute value of the second safety deceleration, an immediate braking signal in the second-level warning is issued to prompt the driver to take braking measures immediately; when the absolute value of the critical deceleration is greater than or equal to the absolute value of the second safety deceleration, an emergency braking signal in the second-level warning is issued to request the driver to immediately perform emergency braking. Preferably, the critical deceleration is measured by a safety braking prediction model.

[0017] Furthermore, the first deceleration is -1.5 m / s 2 ~-2.5m / s 2 ; Preferably, the first deceleration is -2m / s 2 ; The second safety deceleration is -2.5m / s2 ~-4m / s 2 ; Preferably, the second safety deceleration is -3m / s 2 .

[0018] Furthermore, the collision risk assessment method comprises the following steps:

[0019] At time point t in the time series 1 Linear interpolation is performed on the trajectory of the vehicle to obtain the predicted trajectory point P1 of the vehicle;

[0020] At the same time point t 1 Linear interpolation is performed on the target vehicle trajectory to obtain the target vehicle predicted trajectory point P2;

[0021] Construct risk area A1 for the predicted trajectory point P1 of the vehicle;

[0022] Determine whether the target vehicle predicted trajectory point P2 is located in the risk area A1. If the determination result is yes, it is determined that there is a collision risk.

[0023] If the judgment result is no, then move to the next time point t in the time series 2 Make judgment based on this.

[0024] Furthermore, the risk area A1 is constructed as a rectangular area with the predicted trajectory point P1 of the vehicle as a reference point, the length of the rectangular area is adapted to the length of two vehicle bodies, and the width of the rectangular area is adapted to the width of three lanes.

[0025] Furthermore, when the map is available, the predicted trajectory of the vehicle is predicted based on the actual road information provided by the map, and the actual road information includes the lane centerline, lane curvature, and lane intersection turning information. Preferably, the vehicle trajectory prediction uses the lane centerline as a reference to accurately predict the driving trajectory on a straight road or a curve; when passing through an intersection, the best guess of the driving route is made based on the allowed turns in the lane and the direction of the turn signal, so that the trajectory prediction can cross the intersection and fully detect the possibility of collision.

[0026] Furthermore, when the map is not available, the predicted trajectory of the vehicle is predicted based on the vehicle's real-time heading information. Preferably, the vehicle trajectory prediction uses the vehicle's real-time heading as a reference. At this time, it is assumed that the vehicle is traveling in a straight line along the heading direction, and the accuracy probability of trajectory prediction based on heading is improved by constraining the operating conditions at the functional level.

[0027] Preferably, regardless of whether there is a map or not, when performing collision risk assessment, the range of one lane to the left and right of the vehicle's trajectory is used as the lateral risk range, so that in both compatible modes, the risk of abnormal vehicles in the lane and adjacent lanes can be detected.

[0028] Furthermore, when the map is unavailable, whether to activate the warning function is optional, that is, the user is allowed to set it to prohibit activation without a map. At this time, the function will only work on roads with maps, improving the reliability of the warning to meet the needs of different users.

[0029] The present invention further provides an abnormal vehicle classification warning device based on V2X trajectory prediction, which includes:

[0030] A braking strategy module, used to define and store multiple braking strategies;

[0031] The trajectory prediction module is used to generate the predicted trajectory of the vehicle and the predicted trajectory of the target vehicle based on the collected driving parameter information;

[0032] The collision analysis module determines whether there is a collision risk between the host vehicle and the target vehicle based on the predicted trajectories of the host vehicle and the target vehicle;

[0033] A first-level warning module, which issues a first-level warning signal according to the judgment of the first-class collision risk given by the collision analysis module;

[0034] The secondary warning module sends out a secondary warning signal according to the judgment of the second type of collision risk given by the collision analysis module.

[0035] Furthermore, the trajectory prediction module has a map availability judgment unit, a map call prediction unit and a map-free prediction unit:

[0036] A map availability determination unit, used to determine whether there is currently available map information;

[0037] A map call prediction unit is used to predict the trajectories of the host vehicle and the target vehicle based on the actual road information provided by the map;

[0038] The map-free prediction unit is used to predict the trajectories of the vehicle and the target vehicle based on the real-time heading information of the vehicle in the absence of a map.

[0039] The beneficial effects of the present invention are:

[0040] (1) The warning function of the present invention supports providing different levels of warning output according to the degree of collision risk, giving the driver a certain reaction time and different emergency indications, thereby improving the user experience.

[0041] (2) This scheme predicts multiple trajectories for the vehicle, among which the risk analysis is performed on the trajectory of first constant speed and then uniform deceleration and the trajectory of the distant vehicle, so as to determine the timing of starting the warning. This is a warning model based on safe distance. Compared with the TTC warning model, it is safer and can ensure the distance required for parking.

[0042] (3) When there is no map, this solution recommends using the vehicle heading as a reference for trajectory prediction. At the same time, by constraining the function operating conditions (vehicle speed range, yaw rate range) to improve the heading-based trajectory prediction accuracy, minimize dependence on external systems, conform to the current technical level, and help accelerate the mass production of V2X functions. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 FIG. 1 is a flow chart of an abnormal vehicle classification warning method according to Embodiment 1 of the present invention;

[0044] Figure 2 FIG. 2 is a schematic diagram of collision risk assessment in Embodiment 2 of the present invention;

[0045] Figure 3 It is a schematic diagram of the abnormal vehicle classification warning device in Embodiment 3 of the present invention;

[0046] Figure 4 Shown is a schematic diagram of the functional state machine in Example 3 of the present invention. DETAILED DESCRIPTION

[0047] The technical scheme of the preferred embodiment of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] Embodiment 1:

[0049] This embodiment provides a method for abnormal vehicle classification warning based on V2X trajectory prediction, such as Figure 1 As shown, the method specifically comprises the following steps:

[0050] Step S100-Define braking strategy: define at least two braking strategies according to the deceleration timing and / or deceleration magnitude of the vehicle; in a preferred embodiment, the at least two braking strategies include:

[0051] The first braking strategy adopts a prediction model of constant speed first and then braking (CSAE prediction model). Under this prediction model, the vehicle is allowed to travel for the first necessary time (i.e., the reaction time reserved for the driver) and then follow the first deceleration, such as -2m / s 2 deceleration, decelerate;

[0052] The second braking strategy uses a standard braking prediction model. Under the standard braking prediction model, the vehicle immediately decelerates at a first speed, such as -2 m / s. 2 deceleration rate to decelerate.

[0053] Step S200-collision risk assessment:

[0054] Step S210 - vehicle trajectory prediction: predicting the trajectory of the vehicle when executing the above-mentioned at least two braking strategies respectively to form a predicted trajectory of the vehicle;

[0055] Step S220 - target vehicle trajectory prediction: predicting the trajectory of the target vehicle to form a predicted trajectory of the target vehicle;

[0056] Step S230-collision risk analysis: assessing whether there is a collision risk between the host vehicle and the target vehicle based on the predicted trajectory of the host vehicle and the predicted trajectory of the target vehicle;

[0057] Step S300-Level 1 warning: When the evaluation result indicates that the vehicle will collide with the target vehicle when executing the first braking strategy, and will not collide with the target vehicle when executing the second braking strategy, that is, when there is a first-class collision risk, a Level 1 warning (brake preparation signal) is issued to remind the driver to prepare for braking;

[0058] Step S400-Secondary Warning: When the evaluation result indicates that the vehicle will collide with the target vehicle when executing the second braking strategy, that is, when there is a second type of collision risk, a secondary warning signal is issued to prompt the driver to take braking measures immediately; in a preferred embodiment, step S400 includes:

[0059] Step S410 - calculating the critical deceleration: calculating the critical deceleration when the host vehicle and the target vehicle just do not collide by using the safety braking prediction model;

[0060] Step S420 - Risk level determination: comparing the critical deceleration and the second safety deceleration;

[0061] Step S430-immediate braking warning: when the absolute value of the critical deceleration is less than the absolute value of the second safety deceleration, an immediate braking signal in the second-level warning is issued to prompt the driver to take braking measures immediately;

[0062] Step S440 - emergency braking warning: when the absolute value of the critical deceleration is greater than the absolute value of the second safety deceleration, an emergency braking signal in the second-level warning is issued to request the driver to perform emergency braking immediately.

[0063] Among them, in the same evaluation, since the urgency of the first-level warning is less than that of the second-level warning, the driver is allowed a certain amount of free reaction and judgment time under the first-level warning. Usually, the deceleration requirement of the first braking strategy is less than the deceleration requirement of the second braking strategy, or the driving range allowed by the first braking strategy is greater than the driving range allowed by the second braking strategy. In a preferred embodiment, the first deceleration is -1.5m / s 2 ~-2.5m / s2 ; Preferably, the first deceleration is -2m / s 2 ; The second safety deceleration is -2.5m / s 2 ~-4m / s 2 ; Preferably, the second safety deceleration is -3m / s 2 .

[0064] Embodiment 2:

[0065] This embodiment specifically provides a method for completing step S200 in Embodiment 1, that is, a method for evaluating collision risk, which includes the following steps:

[0066] Step S210: Predict the trajectory of the vehicle when executing the above-mentioned at least two braking strategies respectively to form a predicted trajectory of the vehicle; in a preferred embodiment, the trajectory prediction time should be greater than 8s, and the trajectory should be a sequence of trajectory points, and the trajectory point format is: (t, px, py, vx, vy, ax, ay, s), where t: the time corresponding to the trajectory point, px, vx, ax: x-axis coordinates, speed, acceleration, py, vy, ay: y-axis coordinates, speed, acceleration, s: the cumulative distance from the predicted start time to the current point. This step S210 specifically includes: at a given reference time sequence T at time point t 1 On the top, the CSAE trajectory of the vehicle and the standard braking trajectory of the vehicle are interpolated to obtain the predicted trajectory point P1 of the vehicle;

[0067] Step S220: predicting the trajectory of the target vehicle to form a predicted trajectory of the target vehicle;

[0068] The specific implementation method is as follows: at the same time point t 1 Linear interpolation is performed on the target vehicle trajectory to obtain the target vehicle predicted trajectory point P2;

[0069] Step S230: Evaluate whether there is a collision risk between the host vehicle and the target vehicle based on the predicted trajectory of the host vehicle and the predicted trajectory of the target vehicle. The specific process is as follows:

[0070] Step S231: constructing a risk area A1 for the interpolation point P1 on the CSAE predicted trajectory and the standard braking predicted trajectory of the vehicle;

[0071] Step S232: determining whether the target vehicle predicted trajectory point P2 is located within the risk area A1;

[0072] Step S233: If the judgment result is yes, it is determined that there is a corresponding collision risk, and the collision point time and the remaining distance of the vehicle are returned;

[0073] Step S234: If the judgment result is no, then move to the next time point t in the time series. 2Make judgments until t n If no collision is found beyond a certain trajectory range, the system returns to no collision risk.

[0074] In a preferred embodiment, the risk area A1 is constructed as a rectangular area with the predicted trajectory point P1 of the vehicle as a reference point, the length of the rectangular area is adapted to the length of two vehicle bodies, and the width of the rectangular area is adapted to the width of three lanes. Figure 2 As shown, the risk area A1 involves multiple lanes. As long as the target vehicle is located in the risk area A1, a warning signal is issued (it may be the first warning signal or the second warning signal depending on the collision risk). When the target vehicle is too close, a second warning signal is issued.

[0075] In a preferred embodiment, when making a prediction, it is necessary to first determine whether map information can be called:

[0076] When the map is available, the predicted trajectory of the vehicle is predicted based on the actual road information provided by the map, which includes the lane centerline, lane curvature, and lane intersection turning information. Preferably, the vehicle trajectory prediction uses the lane centerline as a reference to accurately predict the driving trajectory on a straight road or a curve; when passing an intersection, the best guess for the driving route is made based on the allowed turns in the lane and the direction of the turn signal, so that the trajectory prediction can cross the intersection and fully detect the possibility of collision.

[0077] When the map is unavailable, the predicted trajectory of the vehicle is predicted based on the vehicle's real-time heading information. Preferably, the vehicle trajectory prediction uses the vehicle's real-time heading as a reference. At this time, it is assumed that the vehicle is traveling in a straight line along the heading direction. At the functional level, the accuracy probability of trajectory prediction based on heading is improved by constraining the operating conditions.

[0078] In a preferred embodiment, regardless of whether there is a map or not, when performing collision risk assessment, the range of one lane to the left and right of the vehicle trajectory is used as the lateral risk range, so that in both compatible modes, abnormal vehicle risks in the lane and adjacent lanes can be detected.

[0079] In a preferred embodiment, when the map is unavailable, whether to activate the warning function is optional, that is, the user is allowed to set it to prohibit activation without a map. At this time, the function will only work on roads with maps, improving the reliability of the warning to meet the needs of different users.

[0080] Embodiment 3:

[0081] This embodiment provides a device for implementing the methods of Embodiment 1 and Embodiment 2, namely, an abnormal vehicle classification warning device based on V2X trajectory prediction, such as Figure 3 As shown, the device comprises:

[0082] A braking strategy module M1, used for defining and storing multiple braking strategies;

[0083] The trajectory prediction module M2 is used to generate the predicted trajectory of the vehicle according to the multiple braking strategies set in the braking strategy module M1 and the collected driving parameter information of the vehicle, and to generate the predicted trajectory of the target vehicle according to the driving speed and road condition information of the target vehicle;

[0084] The collision analysis module M5 determines whether there is a collision risk between the host vehicle and the target vehicle according to the predicted trajectory of the host vehicle and the predicted trajectory of the target vehicle generated by the trajectory prediction module M2;

[0085] The first-level warning module M3 receives the judgment signal of the first-class collision risk from the collision analysis module M5 and triggers the first-level warning signal;

[0086] The secondary warning module M4 receives the judgment of the second type of collision risk given by the collision analysis module M5 and triggers the secondary warning signal.

[0087] In a preferred embodiment, the trajectory prediction module comprises a map availability judgment unit, a map call prediction unit and a map-free prediction unit:

[0088] A map availability determination unit U21 is used to determine whether there is currently available map information;

[0089] The map calling prediction unit U22 is used to predict the trajectory of the host vehicle and the target vehicle based on the actual road information provided by the map;

[0090] The map-free prediction unit U23 is used to predict the trajectories of the host vehicle and the target vehicle based on the real-time heading information of the vehicle in the absence of a map.

[0091] In a preferred embodiment, the secondary warning module is provided with a critical deceleration calculation unit U41, a risk level judgment unit U42, an immediate braking warning unit U43 and an emergency braking warning unit U44: the critical deceleration calculation unit U41 is used to calculate the critical deceleration when the vehicle just does not collide with the target vehicle through a safety braking prediction model; the risk level judgment unit U42 is used to output the judgment result by judging the critical deceleration and the pre-stored second safety deceleration; the immediate braking warning unit U43 is used to issue an immediate braking signal in the secondary warning when the absolute value of the critical deceleration is less than the absolute value of the second safety deceleration, prompting the driver to take braking measures immediately; the emergency braking warning unit U44 is used to issue an emergency braking signal in the secondary warning when the absolute value of the critical deceleration is greater than the absolute value of the second safety deceleration, requesting the driver to perform emergency braking immediately.

[0092] In a specific implementation, the state machine of the hierarchical warning device is as follows: Figure 4 As shown, the function jump mode of the state machine is as follows:

[0093] Main function status description:

[0094] Standby: The function is turned on but not activated and is in waiting state;

[0095] Activation 1: Sub-function 1 is activated, corresponding to the first-level warning module;

[0096] Activation 2: Sub-function 2 is activated, corresponding to the secondary warning module;

[0097] Activation 3: Sub-function 3 is activated, corresponding to the special warning module;

[0098] Silent: The function does not need to be activated at certain times

[0099] Main state machine jump conditions:

[0100] Triggering jump 12 of the first-level warning module: standby jump activation 1. This sub-function 1 is activated when all of the following conditions are met at the same time: (1) the main vehicle speed is 5-120 kph when a map is available; or: the main vehicle (this vehicle) speed is 60-120 kph when no map is available and no map activation is turned on; (2) the speed difference with the nearest dangerous vehicle (target vehicle) is greater than 40 km / h; (3) there is a first-category collision risk with the nearest dangerous vehicle, but there is no second-category collision risk;

[0101] Triggering the No. 13 jump of the secondary warning module: Standby jump activation 2. Sub-function 2 is activated when all of the following conditions are met at the same time: (1) The main vehicle speed is 5-120 kph when there is a map, or: the main vehicle speed is 60-120 kph when there is no map and the no-map activation is turned on; (2) The speed difference with the nearest dangerous vehicle is greater than 40 km / h, and there is a second-category collision risk with the nearest dangerous vehicle;

[0102] Jump No. 15 based on driver behavior: Activate jump silence. Automatic jump condition: The driver accelerates.

[0103] Jump No. 16 based on constant speed driving: Silent jump standby. Automatic jump conditions are: (1) The driver does not accelerate (cancel acceleration); (2) The current state times out (e.g., more than 2 seconds).

[0104] Jump No. 17 based on abnormal status of the target vehicle (nearest dangerous vehicle): Standby jump activation 3. Automatic jump conditions are: (1) the main vehicle speed is 5-120 kph when a map is available, or: when no map is available, the main vehicle speed is 60-120 kph and no map activation is turned on; (2) the nearest dangerous vehicle has its hazard lights on; (3) there is a first-category collision risk with the nearest dangerous vehicle.

[0105] Jump No. 18 based on risk elimination: Activate Jump Standby No. 3. Automatic jump condition is: current state timeout (e.g., more than 1s).

[0106] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made in form and details without departing from the scope defined by the claims of the present invention. The dimensions described in the drawings and embodiments are not related to the specific objects and are not used to limit the protection scope of the present invention. The dimensions of the objects can be selected and changed according to actual needs.

Claims

1. A method for abnormal vehicle classification warning based on V2X trajectory prediction, comprising: Define braking strategies: define at least two braking strategies based on the vehicle's deceleration timing and / or deceleration magnitude; Collision risk assessment: predicting the trajectory of the vehicle when executing at least two of the above braking strategies respectively to form a predicted trajectory of the vehicle; predicting the trajectory of the target vehicle to form a predicted trajectory of the target vehicle; assessing whether there is a collision risk between the vehicle and the target vehicle based on the predicted trajectory of the vehicle and the predicted trajectory of the target vehicle; Level 1 warning: When the evaluation results indicate that the vehicle will collide with the target vehicle when executing the first braking strategy, but will not collide with the target vehicle when executing the second braking strategy, that is, when there is a first-class collision risk, a level 1 warning is issued; Level 2 warning: When the evaluation results indicate that the vehicle will collide with the target vehicle when executing the second braking strategy, that is, when there is a second-level collision risk, a level 2 warning is issued; In the same evaluation, the driving range allowed by executing the first braking strategy is greater than the driving range allowed by executing the second braking strategy.

2. According to claim 1, a method for abnormal vehicle classification warning based on V2X trajectory prediction is characterized in that: The at least two braking strategies include: A first braking strategy allows the vehicle to decelerate according to a first deceleration rate after traveling for a first necessary time; In the second braking strategy, the vehicle immediately decelerates at the first deceleration rate.

3. The abnormal vehicle classification warning method based on V2X trajectory prediction according to claim 1 is characterized in that: When the assessment result is that there is a second-class collision risk, the critical deceleration at which the vehicle just does not collide with the target vehicle is calculated. When the absolute value of the critical deceleration is less than the absolute value of the second safety deceleration, the immediate braking signal in the second-level warning is issued; when the absolute value of the critical deceleration is greater than or equal to the absolute value of the second safety deceleration, the emergency braking signal in the second-level warning is issued.

4. The abnormal vehicle classification warning method based on V2X trajectory prediction according to claim 3 is characterized in that: The first deceleration is -1.5 m / s 2 ~-2.5m / s 2 ; Preferably, the first deceleration is -2m / s 2 ; The second safety deceleration is -2.5m / s 2 ~-4m / s 2 ; Preferably, the second safety deceleration is -3m / s 2 .

5. A V2X-based abnormal vehicle classification warning method according to any one of claims 1 to 4, characterized in that: The collision risk assessment method comprises the following steps: Perform linear interpolation on the vehicle trajectory at time point t1 of the time series to obtain the predicted trajectory point P1 of the vehicle; At the same time point t1, linear interpolation is performed on the target vehicle trajectory to obtain the target vehicle predicted trajectory point P2; Construct risk area A1 for the predicted trajectory point P1 of the vehicle; Determine whether the target vehicle predicted trajectory point P2 is located in the risk area A1. If the determination result is yes, it is determined that there is a collision risk. If the judgment result is no, then move to the next time point t2 of the time series for judgment.

6. The abnormal vehicle classification warning method based on V2X trajectory prediction according to claim 5 is characterized in that: The risk area A1 is constructed as a rectangular area with the predicted trajectory point P1 of the vehicle as a reference point. The length of the rectangular area is adapted to the length of two vehicle bodies, and the width of the rectangular area is adapted to the width of three lanes.

7. A V2X-based abnormal vehicle classification warning method according to any one of claims 1 to 6, characterized in that: When the map is available, the predicted trajectory of the vehicle is predicted based on the actual road information provided by the map, and the actual road information includes the lane centerline, lane curvature, and lane intersection turning information.

8. The abnormal vehicle classification warning method based on V2X trajectory prediction according to any one of claims 1 to 6 is characterized in that: When the map is unavailable, the predicted trajectory of the vehicle is obtained based on the vehicle's real-time heading information.

9. An abnormal vehicle classification warning device based on V2X trajectory prediction, characterized in that: include: A braking strategy module, used to define and store multiple braking strategies; The trajectory prediction module is used to generate the predicted trajectory of the vehicle and the predicted trajectory of the target vehicle according to the collected driving parameter information; The collision analysis module determines whether there is a collision risk between the host vehicle and the target vehicle based on the predicted trajectories of the host vehicle and the target vehicle; A first-level warning module, which issues a first-level warning signal according to the judgment of the first-class collision risk given by the collision analysis module; The secondary warning module sends out a secondary warning signal according to the judgment of the second type of collision risk given by the collision analysis module.

10. The abnormal vehicle classification warning device based on V2X trajectory prediction according to claim 9 is characterized in that: The trajectory prediction module has a map availability judgment unit, a map call prediction unit and a map-free prediction unit: A map availability determination unit, used to determine whether there is currently available map information; A map call prediction unit is used to predict the trajectories of the host vehicle and the target vehicle based on the actual road information provided by the map; The map-free prediction unit is used to predict the trajectories of the vehicle and the target vehicle based on the real-time heading information of the vehicle in the absence of a map.

Images