Method for filtering fusion information of a vehicle and vehicle assisted driving control method

Abstract

The application provides a method for filtering fusion information of a vehicle, comprising: filtering fusion information of the vehicle to filter out a cross target in a stationary target determined based on the fusion information, wherein the fusion information is obtained based on fusion of sensing information about a target in front of the vehicle, and a filtering parameter is set based on a running condition of the vehicle. The application also provides a fusion information filtering device and a computer program product. According to the application, the fusion information of the vehicle is filtered by a cross target filter, and the parameter of the cross target filter is adjusted based on the running condition of the vehicle, so that the cross target contained in the stationary target determined based on the fusion information is filtered out, which lays a foundation for reliable implementation of various auxiliary driving functions of the vehicle, and further improves the safety and user experience of the auxiliary driving functions.

Description

Technical Field

[0001] This application relates to the field of vehicle driver assistance, and more particularly to a method for filtering and processing fused information of a vehicle, a sensing device, a method for vehicle driver assistance control, a driver assistance system, and a computer program product for implementing the steps of the method according to this application. Background Technology

[0002] With the development of vehicle technology, adaptive cruise control systems have been widely used in driving. These systems use onboard sensors to automatically track vehicles ahead and adjust the vehicle's speed and distance to maintain a safe distance. However, when a vehicle crosses in front of the driver, its longitudinal speed remains constant, often leading the driver to misinterpret it as a stationary object and trigger braking. This significantly impacts the user experience of adaptive cruise control in practical applications.

[0003] Therefore, there is room for improvement in the current adaptive cruise control system. Summary of the Invention

[0004] The purpose of this application is to provide a method for filtering and processing fused information of a vehicle, a sensing device, a method for assisted driving control of a vehicle, an assisted driving system, and a computer program product, to at least partially solve the problems in the prior art.

[0005] According to a first aspect of this application, a method for filtering fused information of vehicles is provided, the method comprising:

[0006] - The vehicle's fused information is filtered to remove cross-traversing targets among stationary targets determined based on the fused information, wherein the fused information is obtained based on sensing information about targets in front of the vehicle, and the filtering parameters are set based on the vehicle's operating conditions.

[0007] The core concept of this application includes at least the following: filtering vehicle fusion information through a cross-target filter, and adjusting the parameters of the cross-target filter based on the vehicle's operating conditions, thereby filtering out cross-targets included in stationary targets determined based on the fusion information. This lays the foundation for the reliable implementation of various driver assistance functions of the vehicle, thereby improving the safety and user experience of the driver assistance functions. These driver assistance functions may include, for example, adaptive cruise control, and / or traffic jam assist, automatic emergency braking, and / or forward cross-traffic area braking, etc.

[0008] According to a second aspect of this application, a sensing device is provided, wherein the sensing device may include:

[0009] - A sensing unit configured to acquire sensing information about targets in front of the vehicle;

[0010] as well as

[0011] - A control unit for performing the method according to this application.

[0012] According to a third aspect of this application, a method for assisted driving control of a vehicle is provided, the method comprising:

[0013] - The vehicle's drivability with respect to filtered stationary targets is evaluated based on the fused information processed by the method for filtering vehicle information according to this application, and the vehicle's driver assistance functions are controlled based on the drivability evaluation results.

[0014] According to a fourth aspect of this application, a driver assistance system is provided, the driver assistance system comprising:

[0015] -The sensing device according to this application; and

[0016] - An onboard control unit for performing the method for assisted driving control of a vehicle according to this application.

[0017] According to a fifth aspect of this application, a computer program product, such as a computer-readable program carrier, is provided, comprising computer program instructions that, when executed by a processor, implement the steps of the method according to this application. Attached Figure Description

[0018] The principles, features, and advantages of this application will be better understood below with reference to the accompanying drawings. The drawings include:

[0019] Figure 1 A flowchart illustrating a method for filtering fused information of vehicles according to an exemplary embodiment of this application is shown.

[0020] Figure 2 A schematic diagram of a driving scenario according to an exemplary embodiment of this application is shown;

[0021] Figure 3 A flowchart illustrating a method for filtering fused information of vehicles according to another exemplary embodiment of this application is shown;

[0022] Figure 4 A schematic block diagram of a sensing device according to an exemplary embodiment of this application is shown;

[0023] Figure 5 A flowchart illustrating a method for assisted driving control of a vehicle according to an exemplary embodiment of this application is shown; and

[0024] Figure 6 A schematic block diagram of an assisted driving system according to an exemplary embodiment of this application is shown. Detailed Implementation

[0025] To make the technical problems to be solved, the technical solutions, and the beneficial technical effects of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and several exemplary embodiments. It should be understood that the specific embodiments described herein are only for explaining this application and are not intended to limit the scope of protection of this application.

[0026] Figure 1 A flowchart illustrating a method for filtering fused information of vehicles according to an exemplary embodiment of this application is shown. The following exemplary embodiments describe the method according to this application in more detail.

[0027] like Figure 1 As shown, the method may include step S1. In step S1, the fused information of the vehicle may be filtered to filter out cross-traversing targets among stationary targets determined based on the fused information. The fused information is obtained based on the fusion of perception information about targets in front of the vehicle, and the filtering parameters may be set based on the vehicle's operating conditions. In the current embodiment of this application, the vehicle 1 is equipped with a sensing unit 11 for collecting sensing information about targets in front of the vehicle during vehicle operation. The type of sensor unit may include, for example, an onboard camera, millimeter-wave radar, and / or lidar, and the number of sensing units 11 may be one or more. For example, the vehicle 1 may be equipped with a left front millimeter-wave radar, and / or a right front millimeter-wave radar, and / or a forward-facing millimeter-wave radar, and / or a forward-facing camera, and / or a forward-facing lidar, etc. Figure 2 In the exemplary driving scenario diagram shown, the X-axis represents the longitudinal direction in which vehicle 1 travels along the driving lane, and vehicle 1 travels on the road at a longitudinal speed V. 1XMoving forward, the Y-axis represents the lateral direction of vehicle 1's movement along the lane, perpendicular to the longitudinal direction. During travel, vehicle 1 can, for example, acquire image information of targets within a predetermined distance range in front of it via its onboard camera, and / or acquire radar point cloud information of targets within the predetermined distance range in front of it via millimeter-wave radar, and / or acquire lidar point cloud information of targets within the predetermined distance range in front of it via lidar. By analyzing and processing the image information, and / or the radar point cloud information, and / or the lidar point cloud information, and fusing the analysis results, fused information about these targets can be obtained. Based on this fused information, the fused target in front of the vehicle can be determined. For example, by analyzing and fusing the acquired image information and radar point cloud information, a fused target 2 in front of vehicle 1 can be obtained. The fused target 2 may include, for example, a two-wheeled vehicle, a four-wheeled vehicle, and / or other multi-wheeled vehicles, etc. Figure 2 An example is shown of a four-wheeled vehicle 2. For instance, by analyzing and fusing the acquired image information and lidar point cloud information, another fused target 3 can be obtained in front of the vehicle. This fused target 3 may include pedestrians and / or obstacles, etc. Figure 2 The image shows a pedestrian 3 as an example.

[0028] Here, the fusion targets determined based on the fusion information can be classified, and the categories of fusion targets can include moving targets and stationary targets. Specifically, a WNJ kinematic model, i.e., a white noise model, can be established for the determined fusion targets based on the collected sensing information. This model is particularly suitable for describing the motion process of fusion targets with a traversing motion tendency. This fusion target mainly moves along the Y direction, and its relative speed with respect to vehicle 1 in the X direction is equal to the speed of vehicle 1. When calculating the motion probability of the fusion target based on the two-dimensional Gaussian distribution model, the input of the two-dimensional Gaussian distribution model is the speed of the fusion target in the X and Y directions. The mean speed of the fusion target with a traversing motion tendency is zero in both the X and Y directions, but the variance of the speed in the Y direction is greater than the variance of the speed in the X direction. Therefore, the motion probability of fusion targets with a traversing motion tendency, especially those with low-speed traversing motion, is low when determined by the WNJ kinematic model. Such fusion targets are usually classified as stationary targets. Here, the fused target object can be classified based on its historical motion probability data and current motion probability data. If the historical motion probability of the fused target object is less than a first probability threshold and its current motion probability is less than a second probability threshold, it means that the motion probability of the fused target object is low at the current moment and in the time period before and after the current moment. The fused target object can be classified as a stationary target object. Otherwise, the fused target object is classified as a moving target object.

[0029] The above-described classification method of the fused targets results in the static targets determined based on the fused information including not only completely stationary targets, but also targets that have a tendency to move laterally or are moving laterally at low speed. To filter out the fused targets from the static targets, the vehicle's fused information can be filtered using a fused target filter. In the context of this application, a fused target refers to a target moving in the lateral direction (i.e., the Y direction) in front of vehicle 1 or a target with a tendency to move laterally. Different filtering conditions can be set by adjusting the parameters of the fused target filter based on different characteristics of the fused targets to be filtered out. These filtering conditions can be set, for example, based on the vehicle 1's speed and / or the longitudinal relative distance between vehicle 1 and the static target, such as a longitudinal relative distance threshold, a lateral speed threshold, and / or a heading angle threshold. The fused target filtering process will be described in detail below, taking into account the filtering conditions set in different aspects.

[0030] For example, the longitudinal relative distance threshold for crossing the target filter can be set based on the travel speed of vehicle 1. The longitudinal relative distance threshold can be set to a predefined value, for example, a longitudinal relative distance threshold of 80m can be set when the travel speed of vehicle 1 is 60km / h. Optionally, the longitudinal relative distance threshold for crossing the target filter can also be adjusted based on the travel speed of vehicle 1, wherein the lower the travel speed of vehicle 1, the smaller the longitudinal relative distance threshold can be set. For example, in... Figure 2 In the driving scenario diagram shown, the longitudinal relative distance D between vehicle 1 and the fused target object 3 is... X3 If the distance is less than or equal to the longitudinal relative distance threshold, it means that the distance between vehicle 1 and the fusion target 3 is too close. Regardless of whether the fusion target 3 crosses the front of vehicle 1, it will affect the control of the vehicle 1's assisted driving function. Therefore, the fusion target 3 is not filtered, so that the fusion target 3 is exempt from the constraints of the filtering mechanism.

[0031] Conversely, at the longitudinal relative distance D between vehicle 1 and the fusion target 2... X2 If the distance exceeds the longitudinal relative distance threshold, the fused target 2 can be filtered using a cross-target filter. For example, the longitudinal relative distance D between vehicle 1 and the fused target 2 can be used as the filter. X2 Set the lateral velocity threshold for the cross-target filter. The lateral velocity threshold can be set to a predefined value, for example, the longitudinal relative distance D between vehicle 1 and the fused target 2. X2 When the distance is equal to 80m, a lateral velocity threshold of 3m / s can be set. Optionally, as vehicle 1 approaches the target object 2, the smaller the longitudinal relative distance between vehicle 1 and the target object 2, the larger the lateral velocity threshold can be set. If the lateral velocity of the target object 2 is greater than the lateral velocity threshold, the target object 2 can be filtered out as a crossing target; if the lateral velocity of the target object 2 is less than or equal to the lateral velocity threshold, the target object 2 can be left unfiltered, thus exempting it from the constraints of the filtering mechanism.

[0032] Furthermore, at the longitudinal relative distance D between vehicle 1 and the fusion target 2 X2 If the longitudinal relative distance is greater than the threshold, the longitudinal relative distance D between vehicle 1 and the fusion target 2 can also be used as the basis. X2 Set the heading angle threshold for the cross-target filter, such as Figure 2As shown, the heading angle α represents the angle between the direction of the velocity V2 of the fusion target 2 and the lateral direction. Considering that vehicles also travel laterally during a U-turn, and that the U-turn process takes much longer than the lateral movement process, the vehicle 1's driver assistance system needs to respond promptly to the fusion target making a U-turn (e.g., deceleration). Therefore, it is necessary to distinguish between vehicles making U-turns and those crossing the lateral direction. Since the heading angle of a vehicle during a U-turn is usually large, the heading angle can be used to determine whether the fusion target is making a U-turn or crossing the lateral direction. Specifically, the heading angle threshold of the lateral target filter can be set to a predefined value, for example, 15° at a longitudinal relative distance of 80m. When the heading angle α of the fusion target 2 is less than the heading angle threshold—for example, within the angle range of -15° to 15°—it means that the fusion target 2 is traveling in a direction close to the lateral direction, and the fusion target 2 can be filtered out as a lateral target. Optionally, as vehicle 1 moves closer to the fusion target 2, the smaller the longitudinal relative distance between vehicle 1 and the fusion target 2, the smaller the heading angle threshold can be set.

[0033] According to embodiments of this application, vehicle fusion information can be filtered by a cross-target filter, and the parameters of the cross-target filter can be adjusted based on the vehicle's operating conditions to filter out cross-targets included in stationary targets determined based on the fusion information, thus laying the foundation for the reliable implementation of various vehicle driver assistance functions.

[0034] Figure 3 A flowchart illustrating a method for filtering fused information of vehicles according to another exemplary embodiment of this application is shown. The following only describes the method in relation to... Figure 1 The differences between the embodiments shown are omitted for brevity, and the same steps will not be repeated.

[0035] like Figure 3 As shown, before filtering the vehicle's fusion information through the cross-target filter, the method may further include one or more steps S11 to S13 to filter the stationary target determined based on the fusion information from multiple aspects.

[0036] In step S11, a target existence probability filter can be used to filter the stationary targets determined based on the fused information, thereby removing targets with an existence probability lower than a third probability threshold. Considering that misjudgments of fused targets may occur when fusing sensing information collected by sensing units 11 such as vehicle-mounted cameras, millimeter-wave radar, and / or lidar, the existence probability of each stationary target can be evaluated based on the sensing information collected by each sensing unit 11, and a target existence probability filter can be used to remove stationary targets with a low existence probability.

[0037] In step S12, a target object classification probability filter can be used to filter the stationary targets determined based on the fusion information, in order to filter out targets from the stationary targets 2 whose probability of being classified as motor vehicles is lower than the fourth probability threshold. Since the fused targets include motor vehicles, pedestrians and / or obstacles, the subsequent cross-traffic target filter will only filter motor vehicles—which include, for example, two-wheeled vehicles, four-wheeled vehicles and / or other multi-wheeled vehicles—therefore, it is necessary to filter out targets from the stationary targets whose probability of being classified as motor vehicles is lower than the fourth probability threshold, such as pedestrians 3.

[0038] In step S13, a target lateral stability filter can be used to filter stationary targets determined based on the fusion information, thereby removing targets with a lateral stability probability lower than the fifth probability threshold. In the radar point cloud output by millimeter-wave radar and / or lidar detecting targets with high reflectivity, in addition to imaging at the target's true location, a false image of similar shape and size, known as a "ghost image," may form at other locations. The position of this false image constantly changes. Considering that the lateral distance between vehicle 1 and the target's true location remains essentially constant during the approach to the target, the lateral position variance of the true image in the radar point cloud is small, while the lateral position variance of the false image is large. Therefore, the target lateral stability filter can filter out radar point cloud data with large lateral position variance as false images, thus eliminating the impact of such false images on the fusion information. Through the fusion processing of these filters and the cross-target filter, the robustness and reliability of the target filtering mechanism can be effectively improved.

[0039] Figure 4 A schematic block diagram of a sensing device 10 according to an exemplary embodiment of this application is shown.

[0040] like Figure 4 As shown, the sensing device 10 may include:

[0041] - A sensing unit 11, configured to acquire sensing information of targets in front of the vehicle, wherein the sensing unit 11 includes, for example, an onboard camera, millimeter-wave radar, and / or lidar; and

[0042] - Control unit 12 is used to execute the method for filtering and processing vehicle fusion information according to this application.

[0043] Optionally, the control unit 12 may be configured as a domain controller, which includes at least one processor and a memory, in which program instructions executable by the at least one processor are stored, which, when executed by the at least one processor, implement the method for filtering and processing fused information of a vehicle according to this application.

[0044] Optionally, the sensing unit 11 and the control unit 12 may also be integrated into a camera and / or millimeter-wave radar and / or lidar, thereby enabling the execution of the method for filtering and processing vehicle fusion information according to this application via the camera and / or millimeter-wave radar and / or lidar.

[0045] Next, combined Figure 5 The exemplary embodiments illustrate in detail the application of this fusion information filtering processing method in the driver assistance functions of vehicles. Figure 5 A flowchart illustrating a method for assisted driving control of a vehicle according to another exemplary embodiment of this application is shown. The following only describes... Figure 1 The differences between the embodiments shown are omitted for brevity, and the same steps will not be repeated.

[0046] like Figure 5 As shown, the method may further include step S1'. In step S1', the drivability of the vehicle with respect to the filtered stationary target can be evaluated based on the fused information processed according to the method for filtering vehicle information of this application, and the vehicle's assisted driving function can be controlled based on the drivability evaluation result. In the sense of this application, drivability represents whether vehicle 1 can safely drive through the current location of the stationary target while maintaining its current operating conditions. Assuming vehicle 1 continues to drive in its current lane, the probability of vehicle 1 colliding with the stationary target can be evaluated based on the position information of vehicle 1's current lane and the stationary target, thereby evaluating vehicle 1's drivability with respect to the filtered stationary target. Considering that vehicle 1 may change lanes according to the planned driving route, the planned driving route of vehicle 1 can also be obtained from vehicle 1's navigation system, and the probability of vehicle 1 colliding with the stationary target can be evaluated based on vehicle 1's planned driving route and the position information of the stationary target, thereby evaluating vehicle 1's drivability with respect to the filtered stationary target.

[0047] For example, the adaptive cruise control function of a vehicle can be controlled based on the results of a trafficability assessment. A vehicle 1 with adaptive cruise control activated can automatically drive on the road without driver intervention based on traffic conditions ahead—including the speed of the vehicle in front, the distance to the vehicle in front, road information, etc. For instance, if there is a vehicle in front of vehicle 1 at a predetermined distance, vehicle 1 automatically follows the vehicle in front and maintains a set safe following distance; if there is no vehicle in front of vehicle 1 at a predetermined distance, vehicle 1 automatically drives along the navigation path at a set cruise speed based on road information—including road topology information, speed limit information, ramp information, etc. If a cross-traffic object in front of vehicle 1 is sufficiently far from vehicle 1, has a sufficiently small heading angle, and a sufficiently high lateral speed, this means that by the time vehicle 1 reaches the current location of the stationary object, the cross-traffic object has already completed its crossing and left the current lane. Since these cross-traffic objects have been filtered out from the stationary object selection criteria, vehicle 1's passability with respect to the filtered stationary object is determined to be positive. Therefore, vehicle 1 can safely pass the current location of the stationary object. This allows vehicle 1 to maintain its current operating conditions and continue driving, minimizing the impact of the cross-traffic object on the adaptive cruise control function, especially preventing accidental braking caused by the cross-traffic object. Conversely, if vehicle 1's passability with respect to the filtered stationary object is negative, vehicle 1 cannot safely pass the current location of the stationary object. This allows adjustment of vehicle 1's operating parameters, such as reducing vehicle 1's speed and / or controlling vehicle 1 to perform lane changes, to minimize the risk of traffic accidents between vehicle 1 and the stationary object. In this way, the impact of crossing an object on the vehicle's adaptive cruise control can be minimized, avoiding accidental braking caused by crossing an object, and effectively improving the reliability of the adaptive cruise control and the user experience.

[0048] For example, the traffic jam assist function of a vehicle can be controlled based on the assessment results of its accessibility. When traffic congestion occurs on the road the vehicle is traveling on, vehicle 1 with traffic jam assist activated can automatically control the vehicle's start and stop, and follow the vehicle in front while maintaining a safe following distance, without requiring the driver to frequently operate the accelerator and brake pedals. For targets within a predetermined distance in front of vehicle 1, the fusion information of the vehicle according to the method of this application can be executed to filter out cross-traversing targets among the stationary targets determined based on the fusion information, and the accessibility of the vehicle with respect to the filtered stationary targets can be assessed. If the accessibility of vehicle 1 with respect to the filtered stationary targets is determined to be positive, it means that vehicle 1 can safely drive through the current location of all stationary targets, thereby eliminating the need for additional adjustments to the operating parameters of vehicle 1, minimizing the impact of cross-traversing targets on the control of vehicle 1's traffic jam assist function, and especially avoiding accidental braking of vehicle 1 caused by cross-traversing targets. Conversely, if the passability of vehicle 1 with respect to the filtered stationary target is negative, then vehicle 1 cannot safely drive through the current location of the stationary target. This allows adjustment of vehicle 1's operating parameters, such as reducing vehicle 1's speed and / or controlling vehicle 1 to perform lane-changing operations, in order to avoid collisions as much as possible, thereby improving the safety and user experience of the traffic jam assist function.

[0049] For example, the vehicle's automatic emergency braking function can be controlled based on the results of a passability assessment. When the automatic emergency braking function is activated, vehicle 1 will issue a warning to the driver upon detecting a potential collision risk ahead and will automatically perform emergency braking if necessary. Similarly, for targets within a predetermined distance ahead of vehicle 1, the vehicle's fused information can be filtered to remove intervening targets among the stationary targets identified based on the fused information, and the vehicle's passability with respect to the filtered stationary targets can be assessed. If the passability of vehicle 1 with respect to the filtered stationary targets is determined to be positive, it means that vehicle 1 can safely pass through the current locations of all stationary targets. Therefore, no additional adjustments to the vehicle 1's operating parameters are required to minimize the impact of intervening targets on the control of the vehicle 1's automatic emergency braking function, especially to avoid false warnings and / or false braking caused by intervening targets. Conversely, if the passability of vehicle 1 with respect to the filtered stationary target is negative, then vehicle 1 cannot safely drive through the current location of the stationary target. This allows adjustment of vehicle 1's operating parameters, such as sending a warning message about the risk of collision to prompt the driver to take braking or evasive action. If the driver fails to take effective action in time, the vehicle can automatically brake to avoid a collision, thereby improving the safety and user experience of the automatic emergency braking function.

[0050] For example, the vehicle's forward cross-traffic braking function can be controlled based on the passability assessment results. When the forward cross-traffic braking function is activated, vehicle 1 monitors other road users approaching it laterally from the front in real time. Upon detecting a potential collision risk ahead, it issues a warning to the driver and automatically performs braking if necessary. Similarly, for targets within a predetermined distance in front of vehicle 1, the vehicle's fused information can be filtered to remove cross-traffic targets among the stationary targets identified based on the fused information, and the vehicle's passability with respect to the filtered stationary targets can be assessed. If the passability of vehicle 1 with respect to the filtered stationary targets is determined to be positive, it means that vehicle 1 can safely pass through the current location of all fused targets. Therefore, no additional adjustments to vehicle 1's operating parameters are required to minimize the impact of cross-traffic targets on the control of the vehicle's forward cross-traffic braking function, especially avoiding false warnings and / or false braking caused by cross-traffic targets. Conversely, if the passability of vehicle 1 with respect to the filtered stationary target is negative, vehicle 1 cannot safely drive through the current location of the partially fused target. This allows adjustment of vehicle 1's operating parameters, such as sending a warning message about collision risk to prompt the driver to take braking or evasive action. If the driver fails to take timely and effective measures, the vehicle can automatically brake to avoid a collision, thereby improving the safety and user experience of the braking function in the forward intersection area.

[0051] In addition, it should be noted that the step numbers described herein do not necessarily represent the order of steps, but are merely a reference numeral. The order may be changed depending on the specific circumstances, as long as the technical objective of this application can be achieved.

[0052] Figure 6 A schematic block diagram of an assisted driving system 100 according to an exemplary embodiment of this application is shown.

[0053] like Figure 6 As shown, the driver assistance system 100 may include:

[0054] -The sensing device 10; and

[0055] - Vehicle control unit 110, which is used to perform the method for assisted driving control of a vehicle according to the present application.

[0056] Optionally, the vehicle control unit 110 may be configured as a domain controller, which includes at least one processor and a memory, in which program instructions executable by the at least one processor are stored, which, when executed by the at least one processor, implement the method for assisted driving control of a vehicle according to this application.

[0057] It should be understood that the terms “first,” “second,” “third,” etc., used in this document are for descriptive purposes only and should not be construed as indicating or implying relative importance, nor should they be construed as implicitly specifying the number of technical features indicated.

[0058] If an embodiment includes an "and / or" association between a first feature and a second feature, it should be interpreted as follows: according to one implementation, the embodiment has not only the first feature but also the second feature; according to another implementation, the embodiment has either only the first feature or only the second feature.

[0059] Although specific embodiments have been described above, these embodiments are not intended to limit the scope of this application, even when only a single embodiment is described with respect to a particular feature. The feature examples provided in this application are intended for illustrative purposes and not for limitation, unless otherwise stated. In practice, multiple features may be combined with each other as needed and where technically feasible. Various substitutions, modifications, and alterations are also conceived without departing from the spirit and scope of this application.

Claims

1. A method for filtering and processing fused information about vehicles, the method comprising: The vehicle's fused information is filtered to remove cross-traversing targets among stationary targets identified based on the fused information, wherein the fused information is obtained based on sensing information about targets in front of the vehicle, and the filtering parameters are set based on the vehicle's operating conditions.

2. The method according to claim 1, wherein, A longitudinal relative distance threshold for the cross-target filter is set based on the vehicle's speed. When the longitudinal relative distance between the vehicle and the stationary target is greater than the longitudinal relative distance threshold, the stationary target is filtered. The cross-target filter filters the vehicle's fused information to remove cross-targets from the stationary targets determined based on the fused information.

3. The method according to claim 2, wherein the longitudinal relative distance threshold is set to a pre-given value; and / or The lower the vehicle speed, the smaller the longitudinal relative distance threshold is set.

4. The method according to claim 2, wherein, When the longitudinal relative distance between the vehicle and the stationary target is greater than the longitudinal relative distance threshold, the lateral velocity threshold of the cross-target filter is set based on the longitudinal relative distance between the vehicle and the stationary target. When the lateral velocity of the stationary target is greater than the lateral velocity threshold, the stationary target is filtered out as a cross-target.

5. The method according to claim 4, wherein, The lateral velocity threshold for crossing the target filter is set to a pre-defined value; and / or The smaller the longitudinal relative distance between the vehicle and the stationary target, the larger the lateral velocity threshold is set.

6. The method according to claim 2, wherein, When the longitudinal relative distance between the vehicle and the stationary target is greater than the longitudinal relative distance threshold, the heading angle threshold of the cross-target filter is set based on the longitudinal relative distance between the vehicle and the stationary target. When the heading angle of the stationary target is less than the heading angle threshold, the stationary target is filtered out as a cross-target.

7. The method according to claim 6, wherein, The heading angle threshold for the cross-target filter is set to a predefined value; and / or The smaller the longitudinal relative distance between the vehicle and the stationary target, the smaller the heading angle threshold is set.

8. The method according to any one of claims 1 to 7, wherein, The method further includes: The fused target objects determined based on the fused information are classified, wherein the categories of the fused target objects include moving target objects and stationary target objects.

9. The method according to claim 8, wherein, The fused target object is classified based on its historical motion probability data and current motion probability data. Specifically, if the historical motion probability of the fused target object is less than a first probability threshold and its current motion probability is less than a second probability threshold, the fused target object is classified as a stationary target object.

10. The method according to any one of claims 1 to 7, wherein, The method further includes: The stationary targets identified based on the fused information are filtered using a target presence probability filter to remove targets with a probability of presence lower than a third probability threshold; and / or A target object classification probability filter is used to filter stationary targets determined based on the fused information, in order to remove targets whose probability of being classified as motor vehicles is lower than a fourth probability threshold. The motor vehicles include two-wheeled vehicles, four-wheeled vehicles, and / or other multi-wheeled vehicles; and / or The stationary targets determined based on the fusion information are filtered by a target object lateral stability filter to remove targets whose lateral stability probability is lower than the fifth probability threshold.

11. A sensing device (10), wherein, The sensing device (10) includes: Sensing unit (11), configured to acquire sensing information about a target object in front of the vehicle; and Control unit (12) for performing the method according to any one of claims 1 to 10.

12. The sensing device (10) according to claim 11, wherein, The control unit (12) is configured as a domain controller; and / or The sensing unit (11) and the control unit (12) are integrated into the camera and / or millimeter-wave radar and / or lidar.

13. A method for assisted driving control of a vehicle, the method comprising: The vehicle's drivability with respect to filtered stationary targets is evaluated based on the fused information processed by the method according to any one of claims 1 to 10, and the vehicle's driver assistance functions are controlled based on the drivability evaluation results.

14. The method according to claim 13, wherein, Based on the assessment results of accessibility, the vehicle's adaptive cruise control and / or traffic jam assist functions are controlled, particularly when the accessibility is negative, by reducing the vehicle's speed and / or controlling the vehicle to perform lane change operations.

15. The method according to claim 13, wherein, Based on the results of the accessibility assessment, the vehicle's automatic emergency braking function and / or forward intersection braking function are controlled, wherein, especially when the accessibility is negative, warning information about the risk of collision is sent and / or the vehicle's braking operation is performed.

16. The method according to any one of claims 13 to 15, wherein, The vehicle's passability with respect to filtered stationary targets is assessed based on the vehicle's current lane and the location information of stationary targets. and / or The vehicle's passability with respect to filtered stationary targets is assessed based on the vehicle's planned driving route and the location information of stationary targets.

17. A driver assistance system (100), said driver assistance system (100) include: The sensing device (10) according to claim 11 or 12; and The vehicle control unit (110) is used to perform the method according to any one of claims 13 to 16.

18. The driver assistance system (100) according to claim 17, wherein, The vehicle control unit (110) is configured as a domain controller.

19. A computer program product, such as a computer-readable program carrier, comprising computer program instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 10 and 13 to 16.

Images