Method for adaptive cruise control

Abstract

The application provides a method for adaptive cruise control, comprising: obtaining a first driving track of a first vehicle, the first driving track comprising a plurality of first driving track points; obtaining a second driving track of a second vehicle, the second driving track comprising a plurality of second driving track points, the second vehicle being a current tracking target of the adaptive cruise control of the first vehicle; determining, according to the plurality of first driving track points and the plurality of second driving track points, an average lateral distance value of the plurality of first driving track points relative to the second driving track and an overlapping degree of the first driving track and the second driving track; and determining a tracking strategy of the first vehicle according to the average lateral distance value and the overlapping degree. The method of the application can better determine the tracking strategy, improve the accuracy of vehicle control, and help the driving experience of the driver and the riding comfort of the passenger. In addition, the application also provides a control device and a computer program product associated with the direction.

Description

Technical Field

[0001] This application relates to a vehicle control method, and more specifically, to a method for Adaptive Cruise Control (ACC). Furthermore, this application also relates to control devices and computer program products associated with performing the method. Background Technology

[0002] In the field of vehicle control, the current market demands increasingly stringent ACC performance requirements, and the application scenarios for ACC functionality are becoming increasingly diverse. Currently, the algorithm logic used to select the ACC target in vehicle ACC functionality primarily relies on determining whether the ACC target (e.g., the vehicle ahead) and the vehicle itself (also known as the master vehicle or autonomous vehicle) are in the same lane. This makes the implementation of ACC functionality heavily dependent on acquiring lane line input. However, in certain real-world road conditions, such as at intersections or on roads without lane lines, ACC control often encounters problems. This makes the implementation of ACC functionality in certain situations a typical challenge faced by many Advanced Driving Assistance Systems (ADAS) suppliers today. Summary of the Invention

[0003] To address the aforementioned problems and achieve the aforementioned objectives, this application provides a target selection method for adaptive cruise control.

[0004] The method for adaptive cruise control includes: acquiring a first driving trajectory of a first vehicle, the first driving trajectory including a plurality of first driving trajectory points; acquiring a second driving trajectory of a second vehicle, the second driving trajectory including a plurality of second driving trajectory points, the second vehicle being the current tracking target of the adaptive cruise control of the first vehicle; determining, based on the plurality of first driving trajectory points and the plurality of second driving trajectory points, an average lateral distance value of the plurality of first driving trajectory points relative to the second driving trajectory and a degree of overlap between the first driving trajectory and the second driving trajectory; and determining a tracking strategy for the first vehicle based on the average lateral distance value and the degree of overlap.

[0005] This application also provides a control device related to the above-described method. The control device includes a processor unit and a storage unit configured to store computer program instructions configured to implement at least one of the above-described methods when executed.

[0006] Furthermore, this application provides a computer program product related to the above-described method. The computer program product includes computer program instructions configured to implement at least one of the above-described methods when executed.

[0007] The method for adaptive cruise control provided in this application can better determine the tracking strategy, improve the accuracy of vehicle control, and contribute to the driver's driving experience and the passenger's ride comfort. Attached Figure Description

[0008] The accompanying drawings are merely illustrative examples illustrating the technical solutions according to this application. The embodiments shown in the drawings are exemplary only and are not intended to limit the implementation of the subject matter of this application to these embodiments. In the drawings, similar or identical reference numerals are used to refer to similar or identical elements or components. In the drawings,

[0009] Figure 1 It is a schematic diagram illustrating a vehicle driving scenario.

[0010] Figure 2 yes Figure 1 A partially enlarged view of the scene shown.

[0011] Figure 3 This is a schematic diagram illustrating the requirements for obtaining vehicle trajectory points.

[0012] Figure 4 It is a diagram illustrating the relationship between variance and scaling factor.

[0013] Figure 5 This is a simplified flowchart of a method for adaptive cruise control according to this application. Detailed Implementation

[0014] The method for adaptive cruise control according to this application will now be described with reference to the accompanying drawings. It should be noted that although this application uses a vehicle as the subject of its illustrative description, it is not limited thereto and is equally applicable to other vehicles.

[0015] See Figure 1 The diagram illustrates a vehicle driving scenario. Although it shows a scenario where a first vehicle (in this example, a self-driving vehicle) follows a second vehicle (also known as the current tracking target or target, in this example, the preceding vehicle) making a right turn at an intersection, this application is not limited to this. For example, although the intersection is shown as a crossroads, the method of this application is equally applicable to other types of intersections, such as T-junctions, X-junctions, etc. Furthermore, this application is also applicable to other scenarios where lane line input cannot be detected, such as roads without lane line markings. Those skilled in the art can envision other applicable scenarios through the following description.

[0016] When ACC is enabled, in order to control the operation of the vehicle, the vehicle controller, such as the vehicle's central processing unit (CPU), can record the driving trajectory of the vehicle and the target during the vehicle's operation.

[0017] Below, we will first explain how to obtain the vehicle's driving trajectory (also known as the first driving trajectory).

[0018] In this application, it can be done at intervals of Δt. ego A single vehicle trajectory point is collected to obtain the vehicle's driving trajectory. This time interval Δt... ego It can be set according to the specific application or needs, for example, △t ego The time can be 100ms or 80ms. As an example, n vehicle trajectory points can be collected, where n can be any applicable positive integer, such as 20, 24, 30, etc.

[0019] Subsequently, trajectory point data is acquired for each of the n vehicle trajectory points. The corresponding trajectory point data for each vehicle trajectory point includes: the vehicle's longitudinal distance Dx. ego Horizontal distance of the vehicle Dy ego Vy, the variance of the lateral distance of the vehicle ego And obtain the vehicle's timestamp t for that vehicle trajectory point. ego .

[0020] Then, the vehicle reference point P was selected. ego (As an example, marked as) Figure 1 (Middle). This vehicle reference point can be any point on the vehicle structure; in this example, it is selected as the rear axle center. Using vehicle reference point P... ego Establish an xy coordinate system with the origin, where the x-axis represents the longitudinal direction of the vehicle, that is, the axis direction of the vehicle from the front to the rear, and the y-axis represents the lateral direction of the vehicle, that is, the left and right sides of the vehicle that are perpendicular to the axis direction. Figure 1 The x-axis and y-axis directions shown in the figure represent the longitudinal and lateral directions of the vehicle at the moment shown in the figure.

[0021] In this xy coordinate system, the relative position of each collected vehicle trajectory point to the vehicle reference point P is calculated. ego The longitudinal distance of the vehicle Dx ego (that is, along) Figure 2 The distance of the x-axis relative to the origin 0 (as shown) and the lateral distance Dy of the vehicle are shown. ego (that is, along) Figure 2 The distance of the y-axis relative to the origin 0 is shown in the diagram. Finally, based on the obtained (recorded) lateral distance Dy of the vehicle's trajectory points... egoCalculate the variance of the lateral distance Vy of the vehicle. ego Since calculating distance variance is a data processing method well known to those skilled in the art, this application will not elaborate on its calculation method.

[0022] For the target (the vehicle in front), it is also necessary to acquire its target trajectory points. Acquiring the target trajectory points can be done simultaneously with acquiring the vehicle's trajectory points, or sequentially or alternately. For example, a vehicle trajectory point and a target trajectory point can be acquired simultaneously at the same time. Alternatively, a target trajectory point can be acquired after acquiring a vehicle trajectory point. Those skilled in the art can conceive of other methods for acquiring trajectory points; therefore, this application does not limit this. Alternatively, the number of target trajectory points, m, and the number of vehicle trajectory points, n, can differ. In this application, m can also be any applicable positive integer, such as 40, 60, etc.

[0023] It is important to note that the number of these trajectory points is determined to ensure that the length of the obtained target trajectory (also known as the second driving trajectory) is sufficient to calculate the similarity to the vehicle's trajectory. Therefore, preferably, the duration of the recorded effective vehicle trajectory is kept within a certain range (e.g., more than 2 seconds), and the duration of the effective target trajectory also needs to be kept within a certain range (e.g., more than 5 seconds). This is because in this embodiment, the system's following interval is set to a maximum of 3 seconds, while collecting the vehicle trajectory requires an additional 2 seconds. It should be noted that the above data is illustrative only and not restrictive; those skilled in the art can select appropriate data values ​​based on specific application considerations.

[0024] Furthermore, although the two time intervals for the vehicle and the target are set to be of equal duration here (e.g., Δt), ego =△t tar =100ms), but this application is not limited to this, and those skilled in the art can selectively set these two time intervals according to specific circumstances.

[0025] Subsequently, data was obtained for each target trajectory point. The corresponding trajectory point data for each target trajectory point includes: the target longitudinal distance Dx. tar Target lateral distance Dy tar Variance of target lateral distance Vy tar And obtain the target timestamp t of the target trajectory point. tar .

[0026] The trajectory point data is also obtained within the aforementioned xy coordinate system. In other words, the vehicle's reference point P is used. ego This eliminates the need to select a separate calculation reference point. Specifically, for each of the m collected target trajectory points, the vehicle's reference point P relative to the current time is calculated.ego Target longitudinal distance Dx tar (that is, along) Figure 2 The distance of the x-axis relative to the origin 0 (as shown) and the lateral distance Dy from the target are shown. tar (that is, along) Figure 2 The distance of the y-axis relative to the origin 0 is shown in the diagram. Finally, based on the target lateral distance Dy of all m target trajectory points obtained... tar Calculate the target lateral distance variance Vy tar .

[0027] It is worth noting that a key principle in the trajectory acquisition process described above is that the acquired vehicle trajectory should be covered by the acquired target trajectory in both the time and longitudinal spatial domains, thus ensuring an effective interaction between the two in the spatial-temporal domains. This trajectory coverage is specifically manifested in the fact that, along the vehicle's driving trajectory, the trajectory points are covered by the earliest vehicle timestamp t. 1ego The earliest corresponding vehicle trajectory point P 1ego It should not be located in the target trajectory point with the earliest target timestamp t 1tar The earliest corresponding target trajectory point P 1tar Behind it (e.g., in front of it or at least parallel to it along the longitudinal direction of the vehicle), and the vehicle trajectory point is the same as the latest vehicle timestamp t. 2ego The corresponding latest vehicle trajectory point P 2ego It should not be located in the target trajectory point with the latest target timestamp t 2tar The corresponding latest target trajectory point P 2tar In front of it (e.g., behind it or at least level with it along the longitudinal direction of the vehicle).

[0028] As an example, Figure 3 This is a simplified diagram illustrating the trajectory acquisition requirements, clearly showing the positional relationship between the earliest and latest trajectory points of the vehicle and the earliest and latest trajectory points of the target. Figure 3 In the diagram, the vehicle's reference point is taken as the origin 0 of the coordinate system. The horizontal axis represents time t, while the vertical axis represents the longitudinal distance Dx between the trajectory point and the vehicle's reference point.

[0029] Alternatively, valid trajectory points can be determined / selected from the acquired vehicle trajectory points. Valid trajectory points can be all acquired trajectory points, a subset of trajectory points excluding singularities in the measurement data, or trajectory points intentionally selected considering different factors or for different applications. In other words, for example, valid trajectory points can be determined before acquiring trajectory data, and then trajectory data can be acquired only for valid trajectory points; alternatively, valid trajectory points can be selected or determined based on the trajectory point data after acquisition; or valid trajectory points can be determined as needed based on specific application scenarios and design requirements.

[0030] It should be noted that determining / selecting valid trajectory points is not mandatory. This step can be omitted when permissible.

[0031] The following explanation uses the scenario where all acquired trajectory points are selected / determined as valid trajectory points as an example. In other words, all trajectory points mentioned in the following explanation are valid trajectory points.

[0032] In scenarios where the vehicle sensor does not detect lane line input or where the vehicle sensor determines that the target has entered a certain range of the intersection, if the above-mentioned principle for obtaining vehicle trajectory is met, namely, the longitudinal distance of the earliest target trajectory point (its distance to the vehicle reference point) is greater than the longitudinal distance of the earliest vehicle trajectory point (its distance to the vehicle reference point), then the filtering algorithm is used to accumulate and calculate the total lateral distance between the n obtained vehicle trajectory points and the target trajectory point.

[0033] Specifically, in the above xy coordinate system, for each of the n vehicle trajectory points, the longitudinal distance dx between it and each target trajectory point is calculated according to the following formula (1):

[0034] dx = |Dx ego - Dx tar | (1)

[0035] In the formula, dx represents the magnitude of the longitudinal distance between the vehicle trajectory point and the target trajectory point;

[0036] Dx ego Represents the longitudinal distance of the vehicle's trajectory points; and

[0037] Dx tar This represents the longitudinal distance of the target trajectory point.

[0038] Subsequently, for each vehicle trajectory point, a matching trajectory point is determined among the target trajectory points. The point with the smallest longitudinal distance value (dx) between the target longitudinal distance and the vehicle's longitudinal distance is the matching trajectory point. Therefore, a one-to-one correspondence is established between the matching trajectory point and the corresponding vehicle trajectory point.

[0039] Then, for a pair of matching vehicle trajectory points and their matching trajectory points, the lateral distance value Dy between the points is calculated according to the following formula (2):

[0040] Dy = | Dy ego - Dy tar | (2)

[0041] In the formula, Dy represents the lateral distance between the vehicle trajectory point and its corresponding matching trajectory point;

[0042] Dy ego This represents the lateral distance of the vehicle's trajectory points; and

[0043] Dy tar This indicates the target lateral distance of the corresponding matched trajectory point.

[0044] Therefore, for n vehicle trajectory points and their corresponding matching points, a total of n lateral distance values ​​between points can be obtained.

[0045] Although the above description states that obtaining the vehicle trajectory point data precedes obtaining the target trajectory point data, those skilled in the art should understand that this application does not restrict the order of data acquisition; it can be performed in reverse order or simultaneously.

[0046] The calculation of the lateral distance variance is described below. Similarly, this application does not limit the order in which the lateral distance variance is obtained and the lateral distances between the points are obtained. Depending on the needs or settings, the steps of obtaining both can be performed simultaneously or sequentially.

[0047] As described above, the variance of the lateral distance Vy of the vehicle is obtained for each vehicle trajectory point. ego Furthermore, the variance of the target lateral distance Vy is obtained for each target trajectory point. tar This application also makes no restriction on the order in which the two were obtained.

[0048] Then, for each vehicle trajectory point, the sum of its lateral distance variance and the target lateral distance variance of the corresponding matching trajectory point is calculated according to the following formula.

[0049] TVy = Vy ego + Vy tar (3)

[0050] In the formula, TVy represents the total lateral distance variance between the vehicle trajectory point and the corresponding matched trajectory point;

[0051] Vy ego The variance of the lateral distance between the vehicle's trajectory points and the vehicle's trajectory points; and

[0052] Vy tar This represents the variance of the target lateral distance for the corresponding matched trajectory points.

[0053] Based on the total lateral distance variance TVy obtained for each vehicle trajectory point, query the variance-scale factor relationship pre-stored in the vehicle storage system. Figure 4 (A schematic diagram illustrating this relationship is shown as an example.) This relationship can be obtained through computer simulation or experimentation. Therefore, for each vehicle trajectory point, a scaling factor R is determined based on its total lateral distance variance TVy and the aforementioned relationship. Subsequently, the total lateral distance value TDy between the vehicle and the target is calculated using a filtering algorithm according to the following formula:

[0054] TDy=Dy1×R1+…△Dy n ×R n

[0055] In the formula, TDy represents the total lateral distance between the vehicle and the target, which is used to characterize the overall consideration of the lateral distance between each vehicle trajectory point and the corresponding matching trajectory point.

[0056] Dy1 indicates that it corresponds to the latest timestamp. Figure 1 and Figure 2 The magnitude of the lateral distance between points obtained from the latest trajectory point of the vehicle (the closest point to the rear of the vehicle);

[0057] R1 represents the scaling factor obtained for the latest vehicle trajectory point;

[0058] Dy n This refers to the one corresponding to the earliest timestamp. Figure 1 and Figure 2 The lateral distance between points obtained from the earliest trajectory point of the vehicle (the furthest point from the rear of the vehicle) is shown in the figure.

[0059] R n This represents the scaling factor used to determine the earliest vehicle trajectory point; and

[0060] n represents the number of vehicle trajectory points obtained.

[0061] Based on the obtained total lateral distance value, the average lateral distance value ADy is calculated according to the following formula:

[0062] ADy = TDy / n,

[0063] In the formula, ADy represents the average lateral distance between the vehicle and the target, which is used to characterize the similarity between the vehicle trajectory and the target trajectory;

[0064] TDy represents the total lateral distance between the vehicle and the target; and

[0065] n represents the number of vehicle trajectory points obtained.

[0066] The following describes how the degree of vehicle overlap is determined. The degree of overlap can include the overlap rate.

[0067] The overlap rate between the vehicle and the target is calculated using the following formula:

[0068] OverRate = 1 – Dy t / (0.5×(W ego +W tar )),

[0069] In the formula, OverRate represents the overlap rate between the vehicle and the target;

[0070] Dy t This represents the reference distance value between the vehicle and the target (i.e., the distance between the vehicle's reference point and the target).

[0071] (The lateral distance between target reference points);

[0072] W ego Indicates the width of the vehicle; and

[0073] W tar Indicates the width of the target vehicle.

[0074] As an example, the rear axle center of the vehicle can be selected as the vehicle's reference point, and the rear axle center of the target vehicle can be selected as the target reference point. However, as those skilled in the art will understand, this application is not limited to this. Depending on actual needs, the corresponding positions of the vehicle and the preceding vehicle can be selected as their respective reference points for calculation.

[0075] Based on the above description, this application provides a method for adaptive cruise control. The following section combines... Figure 5 The flowchart shown below provides a brief summary of one embodiment of the method for adaptive cruise control according to this application.

[0076] S1: Record the vehicle trajectory and target trajectory by obtaining the vehicle trajectory points and target trajectory points respectively, and obtain trajectory point data for each vehicle trajectory point and each target trajectory point. The vehicle trajectory point data includes the vehicle's longitudinal distance, lateral distance, lateral distance variance, and vehicle timestamp. The target trajectory point data includes the target's longitudinal distance, lateral distance, lateral distance variance, and target timestamp.

[0077] S2: Determine whether the longitudinal distance of the earliest target trajectory point is less than the longitudinal distance of the earliest vehicle trajectory point.

[0078] If the determination result in step S2 is yes, proceed to step S3: For each vehicle trajectory point, determine the matching trajectory point among the target trajectory points, where the longitudinal distance between the vehicle trajectory point and the corresponding matching trajectory point is the shortest, and then calculate the lateral distance value Dy between the vehicle trajectory point and the corresponding matching trajectory point. If the determination result is no, return to step S1.

[0079] S4: Using a filtering algorithm, based on the lateral distance value Dy between each vehicle trajectory point and the corresponding matching trajectory point, obtain the total lateral distance value TDy and the average lateral distance value ADy between the vehicle and the target.

[0080] S5: Calculate the overlap rate (OverRate) between the vehicle and the target.

[0081] S6: Determine whether OverRate > the first threshold and ADy < the second threshold. In the above example, the first threshold can be set to 0.1 and the second threshold can be set to 1m, but this application does not limit this. The first threshold and the second threshold can be set separately based on the specific application scenario.

[0082] If the determination result in S6 is yes, proceed to step S7: determine the tracking strategy to keep the current target as the tracking target. If the determination result is no, return to execute step S1 and / or determine the tracking strategy to release the current target.

[0083] Thus, a workflow for the method of this application is completed. Using the method for adaptive cruise control according to this application, stable tracking can be ensured throughout the turning process.

[0084] Furthermore, the method for adaptive cruise control in this application may further include a determination process, namely, determining whether to implement the above-mentioned step S6, or even S1-S7, only when the target is within a certain range of the intersection or when no lane line input is obtained.

[0085] As previously stated, the solution of this application is not limited to use in vehicles, but is applicable to a variety of suitable vehicles. Therefore, the solution of this application fundamentally includes: acquiring a first driving trajectory of a first vehicle, the first driving trajectory including a plurality of first driving trajectory points; acquiring a second driving trajectory of a second vehicle, the second driving trajectory including a plurality of second driving trajectory points, the second vehicle being the current tracking target of the adaptive cruise control of the first vehicle; determining, based on the plurality of first driving trajectory points and the plurality of second driving trajectory points, an average lateral distance value of the plurality of first driving trajectory points relative to the second driving trajectory and a degree of overlap between the first driving trajectory and the second driving trajectory; and determining a tracking strategy for the first vehicle based on the average lateral distance value and the degree of overlap.

[0086] The tracking strategy is determined using the method described above only when the second vehicle is detected to be within a certain range (e.g., a first range) of the intersection or when no lane line input is detected. This reduces the computational load. The application scenarios presented herein are merely exemplary and not limiting. Those skilled in the art can conceive of other applicable application scenarios based on practical considerations.

[0087] In one embodiment of this application, the tracking strategy includes: maintaining the second vehicle as the tracking target of the first vehicle when a threshold condition related to the average lateral distance value and the overlap ratio is met; and releasing the current tracking target when the threshold condition is not met. As an example, the threshold condition is that the average lateral distance value is less than a first threshold and the overlap ratio is greater than a second threshold.

[0088] In the above method, the degree of overlap includes an overlap rate, which is associated with the width of the first vehicle, the width of the second vehicle, and a reference distance between the first and second vehicles. The overlap rate is an important parameter for vehicle control, particularly in vehicle crash testing, regarding the degree of overlap between the impacted portion of the vehicle structure and an obstacle. In the method of this application, the adoption of this overlap rate is primarily aimed at considering driving safety between the two vehicles.

[0089] In the method of this application, for different vehicles, some or all of the multiple trajectory points can be considered as valid trajectory points, and trajectory point data is then obtained for each valid trajectory point. For the first vehicle, its first trajectory point data includes a first longitudinal distance, a first lateral distance, a first lateral distance variance, and a first timestamp. For the second vehicle, its second trajectory point data includes a second longitudinal distance, a second lateral distance, a second lateral distance variance, and a second timestamp. Meanwhile, to ensure the validity of subsequent data calculation and processing, the range of the second timestamp should not be less than the range of the first timestamp.

[0090] The relationship between these two timestamp ranges is determined by considering the following parameter: the difference between the first longitudinal distance of the first driving trajectory point with the earliest first timestamp and the second longitudinal distance of the second driving trajectory point with the earliest second timestamp. Therefore, only when the difference is negative is the requirement met.

[0091] When the difference is negative, based on the first longitudinal distance between the plurality of first driving trajectory points and the second longitudinal distance between the plurality of second driving trajectory points, a matching trajectory point is determined for each valid trajectory point among the plurality of first driving trajectory points from the plurality of second driving trajectory points, wherein the difference between the second longitudinal distance of the matching trajectory point corresponding to a valid trajectory point and the first longitudinal distance of the valid trajectory point is the smallest.

[0092] Then, based on each valid trajectory point among the plurality of first driving trajectory points and the corresponding matching trajectory point, the lateral distance value between the points is calculated.

[0093] In the method of this application, the average lateral distance value used is associated with the lateral distance between each valid trajectory point among the plurality of first travel trajectory points of the first vehicle, a scaling factor, and the number of the plurality of first travel trajectory points. The scaling factor is determined based on the total variance of the lateral distance and a predetermined correspondence between variance and scaling factor. The introduction of the scaling factor aims to take into account the degree of influence of each data point on the final data, thereby making the acquisition of the final data more rational.

[0094] Furthermore, this application also relates to a control device comprising a processor unit and a storage unit. The storage unit is configured to store computer program instructions configured to, when executed, implement any of the aforementioned methods for adaptive cruise control.

[0095] Furthermore, this application also relates to a computer program product comprising computer program instructions configured to implement any of the methods described above when executed.

[0096] Although several embodiments of this application have been described with reference to the accompanying drawings, as will be understood by those skilled in the art, various modifications can be made to the above embodiments without departing from the scope defined by the appended claims. The above embodiments are provided merely as examples to illustrate the technical solutions of this application and are not intended to limit the scope of protection of this application. Features or elements described in one embodiment may be incorporated into another embodiment unless they contradict existing features or elements in another embodiment. Furthermore, the specific wording of features and the possible use of reference numerals in the appended claims are not intended to limit the scope of protection claimed.

Claims

1. A method for adaptive cruise control, comprising: Obtain the first driving trajectory of the first vehicle, the first driving trajectory including multiple first driving trajectory points; The second driving trajectory of the second vehicle is obtained, the second driving trajectory includes multiple second driving trajectory points, and the second vehicle is the current tracking target of the adaptive cruise control of the first vehicle; Based on the plurality of first driving trajectory points and the plurality of second driving trajectory points, determine the average lateral distance value of the plurality of first driving trajectory points relative to the second driving trajectory and the degree of overlap between the first driving trajectory and the second driving trajectory; as well as When the second vehicle is within the first range of the fork in the road or when no lane line input is detected, the tracking strategy of the first vehicle is determined based on the average lateral distance value and the degree of overlap, so as to determine whether to keep the second vehicle as the tracking target of the first vehicle.

2. The method according to claim 1, wherein, The tracking strategy includes: Under the condition that the threshold conditions related to the average lateral distance value and the degree of overlap are met, the second vehicle is maintained as the tracking target of the first vehicle; and If the threshold condition is not met, release the currently tracked target.

3. The method according to claim 2, wherein, The degree of overlap includes an overlap rate, which is associated with the width of the first vehicle, the width of the second vehicle, and a reference distance between the first vehicle and the second vehicle.

4. The method according to claim 3, wherein, The threshold conditions are that the average lateral distance value is less than a first threshold and the overlap rate is greater than a second threshold.

5. The method according to any one of claims 1-3, wherein, The method includes: For each valid trajectory point among the plurality of first driving trajectory points, first trajectory point data is obtained, wherein the first trajectory point data includes a first longitudinal distance, a first lateral distance, a first lateral distance variance, and a first timestamp; For each valid trajectory point among the plurality of second driving trajectory points, second trajectory point data is obtained. The second trajectory point data includes a second longitudinal distance, a second lateral distance, a second lateral distance variance, and a second timestamp. Furthermore, the range of the second timestamp is greater than or equal to the range of the first timestamp.

6. The method according to claim 5, wherein, The method includes: Determine the difference between the first longitudinal distance of the first driving trajectory point with the earliest first timestamp and the second longitudinal distance of the second driving trajectory point with the earliest second timestamp.

7. The method according to claim 6, wherein, The method includes: When the difference is negative, based on the first longitudinal distance between the plurality of first driving trajectory points and the second longitudinal distance between the plurality of second driving trajectory points, a matching trajectory point is determined for each valid trajectory point among the plurality of first driving trajectory points from the plurality of second driving trajectory points, wherein the difference between the second longitudinal distance of the matching trajectory point corresponding to a valid trajectory point and the first longitudinal distance of the valid trajectory point is the smallest.

8. The method according to claim 7, wherein, The method includes: Based on each valid trajectory point among the plurality of first driving trajectory points and the corresponding matching trajectory point, the lateral distance value between the points is calculated.

9. The method according to any one of claims 1-4, wherein, The average lateral distance value is associated with the lateral distance between each valid trajectory point among the plurality of first driving trajectory points of the first vehicle, the scaling factor, and the number of the plurality of first driving trajectory points; as well as The scaling factor is determined based on the total variance of the lateral distance and the predetermined correspondence between variance and scaling factor.

10. A control device, comprising: Processor unit; and A storage unit configured to store computer program instructions configured to, when executed, implement the method according to any one of claims 1-9.

11. A computer program product comprising computer program instructions configured to, when executed, implement the method according to any one of claims 1-9.

Images