Detection method, device and equipment for road fence, vehicle and medium

Abstract

The invention relates to a detection method, device and equipment for a road fence, a vehicle and a medium. The method includes acquiring, by a radar of a vehicle, a plurality of detection points for the road fence. The method further comprises the step of determining the deviation degree of a connecting line of a first detection point and a second detection point in the multiple detection points relative to a predicted driving track of the vehicle. In addition, the method further comprises the step of determining that the static object corresponding to the first detection point is the object outside the road fence in response to the deviation degree meeting the deviation condition. Through the mode, the objects corresponding to the detection points can be accurately distinguished on the basis of the deviation degree of the connecting line of the detection points and the driving track of the vehicle, and objects outside the road fence are prevented from being recognized as the road fence, so that the recognition accuracy of the road fence is improved; therefore, the accuracy of the driving assistance system or the automatic driving system and the safety of the vehicle in the driving process are improved.

Description

Technical Field

[0001] Embodiments of the present disclosure generally relate to the field of intelligent driving, and specifically relate to a detection method, device, equipment, vehicle, and medium for road barriers. Background Art

[0002] As important components of the automobile driving assistance system, the adaptive cruise control system and the automatic emergency braking system have functional specifications that define the functional requirements of the vehicle braking to avoid collision and protect the vehicle's driving safety. That is, if there is a stationary target such as a vehicle or pedestrian in front of the vehicle's lane, and the vehicle and the surrounding environment meet the requirements of the functional specifications, the adaptive cruise control system and the automatic emergency braking system will control the vehicle to slow down and stop in order to maintain a safe distance from the target in front.

[0003] The main challenge for adaptive cruise control systems and automatic emergency braking systems is how to correctly process the road information around the vehicle detected by the vehicle sensors, while avoiding interference from objects around the road to the detection of sensors such as millimeter-wave radars, and ensuring the probability of triggering the adaptive cruise control system and automatic emergency braking system on the correct target. Therefore, how to process the road information detected by vehicle sensors is of great significance to the driving assistance system. Summary of the invention

[0004] The embodiments of the present disclosure propose a method, device, equipment, vehicle and medium for detecting road fences. In the embodiments of the present disclosure, the road fence ahead can be detected by a vehicle-mounted radar to obtain multiple detection points. Then, the embodiments of the present disclosure can select two detection points from the multiple detection points, and calculate the degree of deviation of the direction of the line connecting the two detection points from the predicted driving trajectory of the vehicle. Finally, in the embodiments of the present disclosure, the degree of deviation is judged according to the deviation condition, and when the degree of deviation meets the deviation condition, it is determined that one of the two detection points comes from a stationary object other than the road fence. In this way, it can be avoided that the detection point of the stationary object is mistakenly identified as the detection point of the road fence when the stationary object is too close to the road fence, thereby improving the accuracy of road fence recognition, and then improving the accuracy and safety of the driving assistance system.

[0005] In a first aspect of the present disclosure, a detection method for a road fence is provided. The method includes acquiring a plurality of detection points for the road fence by a radar of a vehicle. The method also includes determining a degree of deviation of a line connecting a first detection point and a second detection point among the plurality of detection points relative to a predicted driving trajectory of the vehicle. In addition, the method also includes determining, in response to the degree of deviation satisfying a deviation condition, that a stationary object corresponding to the first detection point is an object outside the road fence.

[0006] In a second aspect of the present disclosure, a detection device for a road fence is provided. The device includes a detection point acquisition module configured to acquire multiple detection points for the road fence through a radar of a vehicle. The device also includes a deviation degree determination module configured to determine the degree of deviation of a line connecting a first detection point and a second detection point among the multiple detection points relative to a predicted driving trajectory of the vehicle. In addition, the device also includes an object determination module configured to determine that the stationary object corresponding to the first detection point is an object outside the road fence in response to the deviation degree satisfying a deviation condition.

[0007] In a third aspect of the present disclosure, an electronic device is provided. The electronic device includes one or more processors; and a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement a detection method for a road fence. The method includes acquiring a plurality of detection points for the road fence through a radar of a vehicle. The method also includes determining a degree of deviation of a line connecting a first detection point and a second detection point among the plurality of detection points relative to a predicted driving trajectory of the vehicle. In addition, the method also includes determining, in response to the degree of deviation satisfying a deviation condition, that a stationary object corresponding to the first detection point is an object outside the road fence.

[0008] In a fourth aspect of the present disclosure, a vehicle is provided, comprising a radar and the electronic device provided in the third aspect of the present disclosure.

[0009] In a fifth aspect of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer-executable instructions, wherein the computer-executable instructions are executed by a processor to implement the method provided according to the first aspect of the present disclosure.

[0010] It should be understood that the contents described in the summary of the invention are not intended to limit the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:

[0012] Figure 1 A schematic diagram illustrating an example environment in which certain embodiments of the present disclosure may be implemented;

[0013] Figure 2A flow chart showing a detection method for road barriers according to certain embodiments of the present disclosure;

[0014] Figure 3 A schematic diagram showing a scenario of determining a deviation degree based on a self-vehicle coordinate system according to certain embodiments of the present disclosure;

[0015] Figure 4 A schematic diagram showing another scenario of determining a deviation degree based on a self-vehicle coordinate system according to certain embodiments of the present disclosure;

[0016] Figure 5 A schematic diagram showing a road barrier fitting result under a straight road condition according to some embodiments of the present disclosure;

[0017] Figure 6 A schematic diagram showing a road barrier fitting result under a curved road condition according to some embodiments of the present disclosure;

[0018] Figure 7 A flowchart showing another detection method for road barriers according to certain embodiments of the present disclosure;

[0019] Figure 8 A block diagram showing a detection device for road barriers according to some embodiments of the present disclosure; and

[0020] Fig. 9 A block diagram of an electronic device is shown in which certain embodiments of the present disclosure may be implemented.

[0021] Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. DETAILED DESCRIPTION

[0022] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0023] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0024] The vehicle's adaptive cruise control system, automatic emergency braking system and other driving assistance systems, as well as the vehicle's automatic driving system, are the main components of L0-L2 assisted driving functions. Their functional specifications define the functional requirements of automatic braking to avoid collisions and protect the vehicle's driving safety. That is, if there are stationary vehicles or pedestrians in front of the lane, and factors such as the vehicle and the surrounding environment meet the requirements of the functional specifications, the driving assistance system or the automatic driving system needs to control the vehicle to slow down and stop until it maintains a safe distance from the stationary object in front.

[0025] For the driving assistance system or the autonomous driving system, the corresponding challenge is how to correctly process the road information around the vehicle detected by the radar sensor, while avoiding the interference of objects around the road on the radar sensor to reduce the false trigger rate, and at the same time ensuring the trigger rate of the correct identification object of the assisted driving system or the autonomous driving system. In the related technology, millimeter-wave radar is generally used to obtain road information. However, there is often interference with millimeter-wave radar detection in actual road scenes, making it easy for stationary objects next to the road fence to be identified as part of the road fence. This erroneous recognition result will cause the assisted driving system or the autonomous driving system to fail to successfully avoid collisions, making the vehicle less safe during driving.

[0026] To this end, a detection method for road barriers is provided in an embodiment of the present disclosure. First, the road barriers ahead are detected by a vehicle-mounted radar to obtain multiple detection points. Then, the embodiment of the present disclosure can select two detection points from the multiple detection points, and calculate the degree of deviation of the direction of the line connecting the two detection points from the predicted driving trajectory of the vehicle. Finally, in the embodiment of the present disclosure, the degree of deviation is judged according to the deviation condition, and when the degree of deviation meets the deviation condition, it is determined that one of the two detection points comes from a stationary object other than the road barrier. In this way, it can be avoided that the detection point of the stationary object is mistakenly identified as the detection point of the road barrier when the stationary object is too close to the road barrier, thereby improving the accuracy of road barrier recognition, and further improving the accuracy and safety of the driving assistance system or the automatic driving system.

[0027] Figure 1 1 is a schematic diagram of an example environment 100 in which certain embodiments of the present disclosure may be implemented. Figure 1As shown, in the example environment 100, the road 102 includes a lane left edge line 102-1A, a lane right edge line 102-1B, a reverse lane 102-2A, a forward lane 102-2B, a lane center line 102-3, a reverse driving sign 102-4A, and a forward driving sign 102-4B. In addition, the road 102 may also include a lane dividing line (not shown) for separating lanes in the same direction. A road fence 104 is arranged on the right side of the road 102. The road fence 104 is arranged parallel to the lane right edge line 102-1B, and the road fence 104 may be close to the lane right edge line 102-1B. In the forward lane 102-2B, the vehicle 106 is a stationary vehicle parked on the side of the road, and the vehicle 108 is a self-vehicle that is moving.

[0028] Continue to refer Figure 1 , an electronic device 110 is provided in the vehicle 108, and a radar 112 and a control system 114 are provided in the electronic device 110. The radar 112 detects the road fence to collect detection points, and sends the detection points to the control system 114. The control system 114 first executes the acquisition of detection points 116 to receive the detection points sent by the radar 112. Then, the control system 114 executes the deviation degree determination 118 to calculate the deviation degree of the connection line of two of the detection points to the predicted driving trajectory of the vehicle 108. Finally, the control system 114 executes the object determination 120 to determine whether the deviation degree meets the deviation condition, and when the deviation condition is met, the stationary object corresponding to one of the two detection points is determined as an object outside the road fence.

[0029] The present disclosure calculates the degree of deviation of the line connecting two detection points relative to the predicted driving trajectory of the vehicle, and performs a deviation condition judgment on the degree of deviation to determine the object category of the detection point, thereby avoiding the detection point of the stationary object from being mistakenly identified as the detection point of the road fence when the stationary object is too close to the road fence, thereby improving the accuracy of road fence recognition, and improving the accuracy and safety of the driving assistance system. In addition, in some embodiments, the sum of the distances between one detection point and the other two detection points can be used to determine whether the object corresponding to the one detection point is too far from the road fence, and the distance judgment result can be obtained, and then the deviation judgment result and the distance judgment result of the one detection point are combined to determine whether the object corresponding to the one detection point belongs to the road fence, thereby further improving the accuracy of road fence recognition. It should be understood that the architecture and functions in the example environment 100 are described only for exemplary purposes, and do not imply any limitation on the scope of the present disclosure. The embodiments of the present disclosure can also be applied to other environments with different structures and / or functions.

[0030] The following will combine Figures 2 to 9The process according to the embodiment of the present disclosure is described in detail. For ease of understanding, the specific data mentioned in the following description are exemplary and are not intended to limit the scope of protection of the present disclosure. It is understood that the embodiments described below may also include additional actions not shown and / or the actions shown may be omitted, and the scope of the present disclosure is not limited in this respect.

[0031] Figure 2 FIG. 2 is a flow chart showing a detection method 200 for road barriers according to some embodiments of the present disclosure. The detection method 200 may be composed of Figure 1 The control system 114 is coupled to or communicates with the radar 112. Figure 2 In block 202, a plurality of detection points for the road fence 104 are acquired by the radar 112 of the vehicle. In some embodiments, the detection points of the road fence 104 can be acquired by a millimeter-wave radar arranged on the vehicle. In an actual road scene, if the road fence 104 and the stationary object next to it are made of different materials, for example, the road fence 104 is made of metal, and the stationary object, i.e., the stationary object next to the road fence 104, is made of non-metallic material, the road fence 104 and the stationary object can be distinguished according to the properties of the detection points acquired by the millimeter-wave radar. However, in the case where the two materials are the same, for example, when the stationary object is a vehicle, since the materials of the road fence 104 and the vehicle are both metal, the properties of the detection points reflected by the two are similar and need to be further distinguished.

[0032] In some embodiments, after acquiring multiple detection points for the road barrier 104 through the radar 112, a clustering algorithm may be used to process these detection points, remove detection points with low correlation with the road barrier 104, and use detection points with high correlation with the road barrier 104 as detection points from the road barrier 104 and perform fitting, thereby obtaining a fitting curve or fitting polyline of the road barrier 104. However, when a stationary object is too close to the road barrier 104, or the attributes of the detection points from the stationary object are similar to those of the detection points from the road barrier 104, the detection points from the stationary object may be identified as detection points with high correlation with the road barrier 104, thereby causing the stationary object next to the road barrier 104 to be associated with the fitting curve or fitting polyline, resulting in erroneous recognition and fitting of the road barrier 104. In this regard, the solution in some embodiments of the present disclosure may be further combined to determine the object corresponding to the detection point.

[0033] In block 204, the degree of deviation of the line connecting the first detection point and the second detection point among the plurality of detection points relative to the predicted driving trajectory of the vehicle is determined. In some embodiments, the first detection point and the second detection point may be adjacent points among the plurality of detection points. Alternatively, the first detection point and the second detection point may also be non-adjacent points. It is understood that the driving trajectory of the vehicle 108 is generally parallel to the road fence 104. If the first detection point and the second detection point are both derived from the road fence 104, the direction of the line connecting the first detection point and the second detection point can reflect the direction of the road fence 104, so the direction of the line should also be close to parallel with the driving trajectory, that is, the degree of deviation between the two is small; on the contrary, if the first detection point is derived from a stationary object beside the road fence 104, and the second detection point is derived from the road fence 104, the direction of the line connecting the first detection point and the second detection point cannot reflect the direction of the road fence 104, so the direction of the line will not be close to parallel with the driving trajectory, that is, the degree of deviation between the two is large. It should be noted that the first detection point and the second detection point in the present disclosure are only used to distinguish and conveniently describe two detection points among the multiple detection points, and do not limit the collection order of the two detection points.

[0034] In some embodiments, the predicted driving trajectory may refer to the driving trajectory that the ego vehicle 108 will pass through, and the predicted driving trajectory is calculated through the driving trajectory that the ego vehicle 108 has already traveled (i.e., the historical driving trajectory in this article) and the heading angle, so as to obtain the predicted driving trajectory. Further, in order to facilitate the observation and calculation of the degree of deviation, the predicted driving trajectory can also be widened to the left and right sides by a certain distance to obtain the ego vehicle driving trajectory model, for example, widened to the left and right sides by 1.75 meters to 1.85 meters respectively, and one of the ego vehicle driving trajectories closer to the road fence 104 in the ego vehicle driving trajectory model is selected to calculate the degree of deviation. It can be understood that the method of calculating the predicted driving trajectory through the driving trajectory that has been traveled and the heading angle does not need to rely on the real-time detection of the road environment, and can avoid the interference of the predicted driving trajectory caused by inaccurate road information due to factors such as illumination, occlusion, interference, etc., thereby improving the accuracy of the predicted driving trajectory.

[0035] Alternatively or additionally, since the driving direction of the ego vehicle 108 is generally consistent with the lane line of the road, the predicted driving trajectory can also be determined according to the lane line of the driving lane of the ego vehicle 108, for example, based on Figure 1The predicted driving trajectory can be determined by the left edge line 102-1A, the right edge line 102-1B, and the center line 102-3 of the lane shown in the figure. The left edge line 102-1A, the right edge line 102-1B, and the center line 102-3 of the lane can also be directly determined as the predicted driving trajectory. It can be understood that the method of directly determining the predicted driving trajectory through the lane line can reduce the amount of calculation, thereby improving the calculation speed of the predicted driving trajectory. Alternatively or optionally, the predicted driving trajectory can also be calculated in combination with the driving trajectory, heading angle, and lane line of the driving lane of the ego vehicle 108 that has been traveled. For example, different predicted driving trajectory calculation methods are selected according to different scenes such as lighting, occlusion, and equipment computing power, or the corresponding predicted driving trajectories are obtained according to two calculation methods respectively and the results of the two predicted driving trajectories are merged.

[0036] In box 206, it is determined whether the degree of deviation satisfies the deviation condition. In some embodiments, the deviation condition can be set as a threshold condition, and then it can be determined whether the degree of deviation reaches a certain deviation threshold. In some embodiments, the degree of deviation can be the angle formed between the line connecting the two detection points and the tangent of the driving track. Alternatively or additionally, the degree of deviation can be the angle formed between the line connecting the two detection points and the line connecting the two track points on the driving track. In box 208, if the degree of deviation satisfies the deviation condition, it is determined that the stationary object corresponding to the first detection point is an object outside the road fence 104. In some embodiments, if the first detection point and the second detection point are both derived from the road fence 104, the corresponding degree of deviation is small; on the contrary, if the first detection point is derived from a stationary object outside the road fence 104, and the second detection point is derived from the road fence 104, the corresponding degree of deviation is large. Therefore, when the degree of deviation reaches a certain deviation threshold, it can be determined that the first detection point is derived from a stationary object outside the road fence 104.

[0037] In some embodiments, the collected multiple detection points are judged respectively to obtain a group of detection points (which may be referred to as a group of target detection points) from the road barrier 104, and the group of detection points from the road barrier 104 are fitted to obtain a fitting curve or a fitting polyline for the road barrier 104. It can be understood that, compared with the method of directly fitting after obtaining the detection points, in this embodiment, the detection points that are not road barriers can be excluded in the process of road barrier fitting, thereby improving the accuracy of the fitting result, and further improving the accuracy of road barrier detection.

[0038] In the embodiment of the present disclosure, the road fence 104 ahead is first detected by the vehicle-mounted radar 112 to obtain a plurality of detection points. Then, the embodiment of the present disclosure can select two detection points from the plurality of detection points, and calculate the degree of deviation of the direction of the line connecting the two detection points from the predicted driving trajectory of the vehicle 108. Finally, in the embodiment of the present disclosure, the degree of deviation is judged according to the deviation condition, and when the degree of deviation meets the deviation condition, it is determined that one of the two detection points comes from a stationary object other than the road fence 104. In this way, it is possible to avoid misidentifying the detection point of a stationary object as a detection point of the road fence 104 when the stationary object is too close to the road fence 104, thereby improving the accuracy of the recognition of the road fence 104, and further improving the accuracy and safety of the driving assistance system and the automatic driving system.

[0039] Figure 3 A schematic diagram of a scenario 300 for determining the degree of deviation based on the vehicle coordinate system in some embodiments of the present disclosure is shown. Figure 3 , the longitudinal axis Y of the ego vehicle coordinate system can be determined based on the driving direction of the ego vehicle. For example, the longitudinal axis of the ego vehicle coordinate system can be parallel to the current driving direction of the ego vehicle or present a small angle in the direction. The horizontal axis X of the ego vehicle coordinate system is perpendicular to the longitudinal axis Y. The origin of the ego vehicle coordinate system is not limited in the present disclosure, and it can be determined based on the position of the ego vehicle or based on other methods. The stationary object is a stationary vehicle 302 in front of the ego vehicle. The ego vehicle and the stationary vehicle 302 are located in a lane 304. The lane 304 includes a left lane line 304-1 and a right lane line 304-2. The vehicle detects the road fence through radar, and obtains a group of detection points 306 including detection point (X1, Y1) (which can be called the first detection point), detection point (X2, Y2) (which can be called the second detection point) and detection point (X3, Y3) (which can be called the third detection point). The dotted line 308 is the fence curve obtained by directly fitting the group of detection points 306, and the solid line 310 is the fence curve obtained by fitting the detection point (X1, Y1) after object classification judgment according to the following embodiment.

[0040] In some embodiments, after obtaining the detection point (X1, Y1), the detection point (X2, Y2) and the detection point (X3, Y3) in a group of detection points 306, the transverse axis distance (which may be referred to as the first transverse axis distance) between the detection point (X1, Y1) and the detection point (X2, Y2) and the transverse axis distance (which may be referred to as the second transverse axis distance) between the detection point (X1, Y1) and the detection point (X3, Y3) may be determined respectively. The transverse axis distance between the detection point (X1, Y1) and the detection point (X2, Y2) and the transverse axis distance between the detection point (X1, Y1) and the detection point (X3, Y3) are added to obtain the sum of the transverse axis distances. If the sum of the transverse axis distances meets the distance condition and the degree of deviation meets the deviation condition, the stationary vehicle 302 corresponding to the detection point (X1, Y1) is determined as an object outside the road fence. In some embodiments, the distance condition may be established according to a distance threshold, and then it is determined whether the sum of the transverse axis distances reaches the distance threshold. In some embodiments, the distance threshold may be set to 2.5 meters. It can be understood that in this embodiment, the sum of the horizontal axis distances and the degree of deviation are combined to determine whether the detection point (X1, Y1) belongs to the stationary vehicle 302 or the lane fence, which further improves the accuracy of the recognition result.

[0041] In some embodiments, after obtaining the detection point (X1, Y1), the detection point (X2, Y2) and the detection point (X3, Y3) in a group of detection points 306, the line 312 formed by the detection point (X1, Y1) and the detection point (X2, Y2) can be further determined. Then, the intersection point formed by the intersection of the line 312 and the left lane line 304-1 is obtained, and the tangent line 314 (which can be called the first tangent line) of the left lane line 304-1 is determined according to the intersection point, and finally the angle between the line 312 and the tangent line 314 is determined as the degree of deviation. In some embodiments, if the angle exceeds the angle threshold, the stationary vehicle 302 corresponding to the detection point (X1, Y1) can be determined as an object other than the road fence. In some embodiments, the angle threshold can be set to 30 degrees. It can be understood that in this embodiment, the tangent 314 is used as the reference direction of the driving trajectory, and the angle between the connecting line 312 and the tangent 314 is obtained, so that the angle can more accurately measure the degree of deviation between the connecting line 312 and the driving trajectory, thereby improving the accuracy of road barrier detection.

[0042] In some embodiments, after obtaining the detection point (X1, Y1), the detection point (X2, Y2) and the detection point (X3, Y3) in the set of detection points 306, the angle A (which may be referred to as the first angle) of the line 312 between the detection point (X1, Y1) and the detection point (X2, Y2) may be determined according to the established vehicle coordinate system. The specific calculation formula is shown in formula (1):

[0043]

[0044] Then, the curve expression of the left lane line 304-1 can be determined according to the vehicle coordinate system, as shown in formula (2):

[0045] y=ax 2 +bx+c Formula (2)

[0046] Where a, b, and c are coefficients in the curve expression. Obtain the intersection coordinates (X0, Y0) of the connecting line 312 and the left lane line 304-1, and substitute the intersection coordinates (X0, Y0) into the slope formula to obtain the slope of the tangent line 314, wherein the slope formula is shown in formula (3):

[0047] y′=2ax+b Formula (3)

[0048] Furthermore, the angle B (which may be referred to as the second angle) of the tangent line 314 may be calculated. The specific calculation formula is shown in formula (4):

[0049] B=arctan (2aX0+b) Formula (4)

[0050] Finally, after the angle A of the connecting line 312 and the angle B of the tangent line 314 are calculated, the angle between the connecting line 312 and the tangent line 314 can be calculated according to the difference between the angle A and the angle B and used as the degree of deviation.

[0051] Figure 4 A schematic diagram of another scenario 400 for determining the degree of deviation based on the vehicle coordinate system of certain embodiments of the present disclosure is shown. In some embodiments, after obtaining a detection point (X1, Y1), a detection point (X2, Y2), and a detection point (X3, Y3) in a set of detection points, a projection point 402 of the detection point (X1, Y1) in the left lane line can be determined. After determining the projection point 402, a tangent 406 (which may be referred to as a second tangent) to the left lane line at the projection point 402 is determined. It can be understood that the line connecting the detection point (X1, Y1) and the projection point 402 is perpendicular to the tangent 406. Then, the angle between the line 404 between the detection point (X1, Y1) and the detection point (X2, Y2) and the tangent 406 is calculated and determined as the degree of deviation.

[0052] In some embodiments, after acquiring multiple detection points and determining the first detection point among the multiple detection points, firstly, a candidate detection point can be determined based on the first detection point, for example, a detection point adjacent to or close to the first detection point is selected as a candidate detection point. Then, the distance between the candidate detection point and the first detection point on the longitudinal axis can be calculated. Finally, if the distance between the candidate detection point and the first detection point on the longitudinal axis is within a preset distance range, the candidate detection point can be determined as the second detection point, and then the object corresponding to the first detection point is identified according to the degree of deviation between the connecting line of the first detection point and the second detection point and the predicted driving trajectory. It can be understood that if the distance between the first detection point and the second detection point on the longitudinal axis is too small, the direction of the connecting line between the first detection point and the second detection point will deviate excessively from the predicted driving trajectory, making the degree of deviation too large; if the distance between the first detection point and the second detection point on the longitudinal axis is too large, the direction of the connecting line between the first detection point and the second detection point will be too close to the predicted driving trajectory, making the degree of deviation too small. Therefore, by preliminarily screening the candidate detection points through the longitudinal distance in this embodiment to obtain the second detection point with a suitable distance, the accuracy of the deviation degree calculation can be improved, thereby improving the accuracy of the road fence detection.

[0053] In some embodiments, reference Figure 3 or Figure 4 , when the candidate detection point is set to the detection point between the detection point (X2, Y2) and the detection point (X3, Y3) (i.e. Figure 3 The detection point 316 in Figure 4 When the candidate detection point is set as the detection point 408 in the left lane, since the longitudinal axis distance between the candidate detection point and the detection point (X1, Y1) is too small, the line connecting the detection point (X1, Y1) and the candidate detection point will be nearly perpendicular to the left lane line, resulting in a large deviation. Therefore, the candidate detection point may not be used as the second detection point. When the candidate detection point is set as the detection point before the detection point (X2, Y2) or the detection point after the detection point (X3, Y3), for example Figure 3 The detection point 318 or Figure 4 In the detection point 410, since the longitudinal axis distance between the candidate detection point and the detection point (X1, Y1) is too large, the line connecting the detection point (X1, Y1) and the candidate detection point will be nearly parallel to the left lane line, resulting in a small deviation, so the candidate detection point cannot be used as the second detection point. However, the longitudinal axis distance between the detection point (X2, Y2) and the detection point (X1, Y1) is within the appropriate distance range, so the detection point (X2, Y2) can be used as the second detection point.

[0054] In some embodiments, multiple reflection points for road fences can be collected by the radar of the vehicle, and then the multiple reflection points can be clustered to remove reflection points that are not highly related to the road fences. For example, a group of reflection points for road fences can be obtained by clustering the positions or attributes of the reflection points. Further, a group of detection points and the sparsity of the group of detection points are determined from the multiple reflection points. If the sparsity meets the sparsity condition, such as the sparsity is within a certain sparsity range, the deviation degree is continued to be obtained based on the group of detection points to perform road fence detection. It can be understood that by screening a group of detection points according to the sparsity of the detection points, it can be ensured that the longitudinal axis distance between the detection points is within a suitable range, thereby ensuring the accuracy of the deviation degree.

[0055] Figure 5 A schematic diagram of a road barrier fitting result 500 under a straight road condition in some embodiments of the present disclosure is shown. In some embodiments, a stationary vehicle 502 and an ego vehicle 504 are located in a lane 508, and a road barrier 506 is provided on the right side of the lane 508. During the process of the ego vehicle 504 detecting the road barrier 506, if the stationary vehicle 502 corresponding to the first detection point is not identified according to the degree of deviation of the line between the two detection points relative to the predicted driving trajectory, the stationary vehicle 502 will be mistakenly identified as a part of the road barrier 506, and then an incorrect fitting result 510 shown in the dotted line portion is obtained. If the stationary vehicle 502 corresponding to the first detection point is identified according to the degree of deviation of the line between the two detection points relative to the predicted driving trajectory, the first detection point can be removed from the multiple detection points before the barrier fitting, so as to fit and obtain the correct fitting result 512 shown in the solid line portion.

[0056] Figure 6 A schematic diagram of a road fence fitting result 600 under curved road conditions in certain embodiments of the present disclosure is shown. In some embodiments, a stationary vehicle 602 and an ego vehicle 604 are located in a lane 608, and a road fence 606 is provided inside the lane 608. During the process of the ego vehicle 604 detecting the road fence 606, if the stationary vehicle 602 corresponding to the first detection point is not identified based on the degree of deviation of the line between the two detection points relative to the predicted driving trajectory, the stationary vehicle 602 will be mistakenly identified as a part of the road fence 606, thereby obtaining an erroneous fitting result 610 shown in the dotted line portion. If the stationary vehicle 602 corresponding to the first detection point is identified based on the degree of deviation of the line between the two detection points relative to the predicted driving trajectory, the first detection point can be removed from the multiple detection points before the fence fitting, thereby obtaining the correct fitting result 612 shown in the solid line portion. Combined with Figure 5 as well as Figure 6It can be seen that since the degree of deviation is determined by the angle between the line between the detection points and the tangent of the predicted driving trajectory (i.e., the tangent of the lane line), its calculation process and results are not affected by straight road conditions and curved road conditions. It can be seen that the solution in this embodiment has high robustness and adaptability to different road conditions.

[0057] Figure 7 A flowchart of another detection method 700 for road barriers according to certain embodiments of the present disclosure is shown. In some embodiments, in box 702, a detection task of a stationary object next to a road barrier is started. In box 704, the driving trajectory of the vehicle is fitted into a quadratic curve equation in the vehicle coordinate system, and the tangent angle at the intersection of the current detection point (which may be called the first detection point) and the previous detection point (which may be called the second detection point) or the next detection point (which may be called the third detection point) with the driving trajectory is obtained. In some embodiments, the previous detection point or the next detection point is not limited to the detection point adjacent to the current detection point, and it is only necessary to ensure that the distance between the previous detection point or the next detection point and the current detection point on the longitudinal axis is within a reasonable distance range. In some embodiments, the fitted quadratic curve equation can be differentiated according to the coordinates of the intersection point to obtain the slope of the tangent, and then the tangent angle of the tangent is obtained by the inverse trigonometric function.

[0058] In block 706, the angle of the line between the current detection point and the previous detection point or the next detection point in the vehicle coordinate system, and the sum of the lateral distances between the current detection point and the previous detection point and the next detection point are obtained. In some embodiments, the slope of the line can be determined according to the coordinates of the current detection point and the previous detection point or the next detection point, and then the line angle of the line can be obtained by an inverse trigonometric function. In some embodiments, the first horizontal axis distance between the current detection point and the previous detection point can be determined first, and then the second horizontal axis distance between the current detection point and the next detection point can be determined, and then the first horizontal axis distance and the second horizontal axis distance are added to obtain the sum of the horizontal axis distances.

[0059] In box 708, determine whether the difference between the tangent angle and the connecting line angle is greater than 30 degrees and whether the sum of the horizontal axis distances is greater than 2.5 meters. In some embodiments, the angle threshold can be set based on the radar sampling period of the detection point. When the radar sampling period is short, the distance between the detection points is small, and the deviation of the connecting line between the detection points from the driving trajectory is large. At this time, the angle threshold can be increased, for example, set to 35 degrees. When the radar sampling period is long, the distance between the detection points is large, and the deviation of the connecting line between the detection points from the driving trajectory is small. At this time, the angle threshold can be lowered, for example, set to 25 degrees. In some embodiments, the distance threshold can be set according to the actual road scene. When the road lane is wide, the distance threshold can be increased, for example, set to 3 meters. When the road lane is narrow, the distance threshold can be lowered, for example, set to 2 meters.

[0060] In box 710, if the above two conditions are satisfied at the same time, it indicates that the stationary object corresponding to the current detection point is far away from the road fence, and the stationary object corresponding to the current detection point is determined as an object other than the road fence. In box 712, if at least one of the above two conditions is not satisfied, it indicates that the object corresponding to the current detection point is very close to the road fence or belongs to the road fence, and the object corresponding to the current detection point is determined as the road fence. In box 714, the detection task of the stationary object next to the road fence is terminated. In this way, it can be avoided that the detection point of the stationary object is mistakenly identified as the detection point of the road fence when the stationary object is too close to the road fence, thereby improving the accuracy of road fence recognition, and further improving the accuracy and safety of the driving assistance system.

[0061] Figure 8 FIG. 8 is a block diagram of a detection device 800 for road barriers according to some embodiments of the present disclosure. Figure 8 As shown, the device 800 includes a detection point acquisition module 802, which is configured to acquire multiple detection points for the road fence through the radar of the vehicle. The device 800 also includes a deviation degree determination module 804, which is configured to determine the deviation degree of the connection line between the first detection point and the second detection point in the multiple detection points relative to the predicted driving trajectory of the vehicle. In addition, the device 800 also includes an object determination module 806, which is configured to determine that the stationary object corresponding to the first detection point is an object outside the road fence in response to the deviation degree satisfying the deviation condition.

[0062] In some embodiments, the position of each of the multiple detection points is determined based on the vehicle's ego-vehicle coordinate system, the longitudinal axis of the ego-vehicle coordinate system corresponds to the vehicle's driving direction, and the transverse axis of the ego-vehicle coordinate system is perpendicular to the longitudinal axis, and the object determination module 806 is further configured to: determine a first transverse axis distance between a first detection point and a second detection point; determine a second transverse axis distance between the first detection point and a third detection point among the multiple detection points; determine the sum of the transverse axis distances based on the first transverse axis distance and the second transverse axis distance; and in response to the sum of the transverse axis distances satisfying a distance condition and the degree of deviation satisfying a deviation condition, determine that the stationary object corresponding to the first detection point is an object outside the road fence.

[0063] In some embodiments, the distance condition is that the sum of the horizontal axis distances is greater than 2.5 meters.

[0064] In some embodiments, the deviation degree determination module 804 is further configured to: determine the intersection of the connecting line and the predicted driving trajectory; based on the intersection, determine the first tangent of the predicted driving trajectory; and determine the angle between the connecting line and the first tangent as the deviation degree.

[0065] In some embodiments, the deviation degree determination module 804 is further configured to: determine a first angle of a line connecting a first detection point and a second detection point; determine an expression for predicting a driving trajectory based on the vehicle coordinate system; determine a second angle of a first tangent based on the expression and the intersection point; and determine an angle between the line and the first tangent based on the first angle and the second angle.

[0066] In some embodiments, the deviation condition is that the angle between the connecting line and the first tangent line is greater than 30 degrees.

[0067] In some embodiments, the deviation degree determination module 804 is further configured to: determine the projection point of the first detection point in the predicted driving trajectory; based on the projection point, determine the second tangent of the predicted driving trajectory; and determine the angle between the connecting line and the second tangent as the deviation degree.

[0068] In some embodiments, the device 800 also includes: a candidate detection point determination module, configured to determine a candidate detection point among multiple detection points; a longitudinal axis distance determination module, configured to determine the longitudinal axis distance between the first detection point and the candidate detection point; and a second detection point determination module, configured to determine the candidate detection point as the second detection point in response to the longitudinal axis distance being within a preset distance range.

[0069] In some embodiments, the apparatus 800 further includes: a target detection point determination module configured to determine a set of target detection points of the road fence from the plurality of detection points; and a fitting module configured to fit the road fence based on the set of target detection points.

[0070] In some embodiments, multiple detection points are determined based on clustering multiple reflection points of the radar, and the deviation degree determination module 804 is further configured to: determine the sparsity of the multiple detection points; and determine the deviation degree in response to the sparsity degree satisfying the sparsity condition.

[0071] In some embodiments, the device 800 also includes: a first predicted driving trajectory determination module, configured to determine the predicted driving trajectory based on the vehicle's heading angle and historical driving trajectory; or a second predicted driving trajectory determination module, configured to determine the predicted driving trajectory based on the lane line of the vehicle's driving lane.

[0072] It can be understood that, by using the device 800 of the present disclosure, at least one of the advantages that can be achieved by the method or process described above can be achieved. For example, the device 800 can avoid misidentifying the detection point of a stationary object as the detection point of a road fence when the stationary object is too close to the road fence, thereby improving the accuracy of road fence identification, and further improving the accuracy and safety of the driving assistance system and the automatic driving system.

[0073] Fig. 9 9 shows a block diagram of an electronic device 900 in which certain embodiments of the present disclosure may be implemented. Fig. 9 As shown, the electronic device 900 includes a computing unit 901, which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) 902 or computer program instructions loaded from a storage unit 908 into a random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the electronic device 900 can also be stored. The computing unit 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0074] A plurality of components in the electronic device 900 are connected to the I / O interface 905, which may include, for example: an input unit 906, such as a keyboard, a mouse, etc.; an output unit 907, such as various types of displays, speakers, etc.; a storage unit 908, such as a disk, an optical disk, etc.; and a communication unit 909, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 909 allows the electronic device 900 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0075] The computing unit 901 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 901 performs the various methods and processes described above, such as method 200. For example, in some embodiments, the method 200 may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 908. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 900 via the ROM 902 and / or the communication unit 909. When the computer program is loaded into the RAM 903 and executed by the computing unit 901, one or more steps of the method 200 described above may be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to perform the method 200 in any other appropriate manner (e.g., by means of firmware).

[0076] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip systems (SOCs), load programmable logic devices (CPLDs), and the like.

[0077] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0078] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In addition, although each operation is depicted in a specific order, this should be understood as requiring such operations to be performed in the specific order shown or in a sequential order, or requiring that all illustrated operations should be performed to obtain the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Certain features described in the context of separate embodiments can also be implemented in a single implementation in combination. On the contrary, the various features described in the context of a single implementation can also be implemented in multiple implementations individually or in any suitable sub-combination.

[0079] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the equipment, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each frame in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the frame can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous frames can actually be executed substantially in parallel, and they can also be executed in the opposite order sometimes, depending on the functions involved. It should also be noted that each frame in the block diagram and / or flow chart, and the combination of frames in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0080] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

Claims

1. A detection method for road barriers, comprising: Acquire multiple detection points for the road barrier by using a radar of the vehicle; Determining a degree of deviation of a line connecting a first detection point and a second detection point among the plurality of detection points relative to a predicted driving trajectory of the vehicle; as well as In response to the deviation degree satisfying a deviation condition, it is determined that the stationary object corresponding to the first detection point is an object outside the road fence.

2. The method according to claim 1, wherein the position of each of the plurality of detection points is determined based on a self-vehicle coordinate system of the vehicle, the longitudinal axis of the self-vehicle coordinate system corresponds to the driving direction of the vehicle, the transverse axis of the self-vehicle coordinate system is perpendicular to the longitudinal axis, and determining that the stationary object corresponding to the first detection point is an object outside the road fence comprises: determining a first transverse axis distance between the first detection point and the second detection point; determining a second transverse axis distance between the first detection point and a third detection point among the plurality of detection points; Determining a sum of the transverse axis distances based on the first transverse axis distance and the second transverse axis distance; as well as In response to the sum of the horizontal axis distances satisfying a distance condition and the degree of deviation satisfying the deviation condition, it is determined that the stationary object corresponding to the first detection point is the object outside the road fence.

3. The method according to claim 2, wherein the distance condition is that the sum of the horizontal axis distances is greater than 2.5 meters.

4. The method according to claim 2, wherein determining the degree of deviation of a line connecting a first detection point and a second detection point among the plurality of detection points relative to the predicted driving trajectory of the vehicle comprises: Determining an intersection point between the connecting line and the predicted driving trajectory; Based on the intersection point, determining a first tangent line of the predicted driving trajectory; as well as The angle between the connecting line and the first tangent line is determined as the degree of deviation.

5. The method according to claim 4, wherein determining the angle between the connecting line and the first tangent line as the degree of deviation comprises: Determine a first angle of a line connecting the first detection point and the second detection point; Based on the vehicle coordinate system, determining an expression for the predicted driving trajectory; Determining a second angle of the first tangent line based on the expression and the intersection point; as well as Based on the first angle and the second angle, an angle between the connecting line and the first tangent line is determined. The method according to claim 4 , wherein the deviation condition is that an angle between the connecting line and the first tangent line is greater than 30 degrees.

7. The method according to claim 2, wherein determining the degree of deviation of a line connecting a first detection point and a second detection point among the plurality of detection points relative to the predicted driving trajectory of the vehicle comprises: Determining a projection point of the first detection point in the predicted driving trajectory; Based on the projection point, determining a second tangent line of the predicted driving trajectory; as well as The angle between the connecting line and the second tangent line is determined as the degree of deviation.

8. The method according to claim 2, further comprising: determining a candidate detection point among the plurality of detection points; determining a longitudinal distance between the first detection point and the candidate detection point; as well as In response to the longitudinal axis distance being within a preset distance range, the candidate detection point is determined as the second detection point.

9. The method according to any one of claims 2 to 8, further comprising: determining a set of target detection points of the road fence among the plurality of detection points; as well as The road barrier is fitted based on the set of target detection points.

10. The method according to claim 1, wherein the plurality of detection points are determined based on clustering of a plurality of reflection points of the radar, and determining the degree of deviation of a line connecting a first detection point and a second detection point among the plurality of detection points relative to the predicted driving trajectory of the vehicle comprises: Determining the sparsity of the plurality of detection points; as well as In response to the sparsity level satisfying a sparse condition, the deviation level is determined.

11. The method according to claim 1, further comprising: Determining the predicted driving trajectory based on the heading angle and historical driving trajectory of the vehicle; or The predicted driving trajectory is determined based on a lane line of a driving lane of the vehicle.

12. A detection device for road barriers, comprising: A detection point acquisition module is configured to acquire a plurality of detection points for the road fence through a radar of the vehicle; a deviation degree determination module, configured to determine a deviation degree of a line connecting a first detection point and a second detection point among the plurality of detection points relative to a predicted driving trajectory of the vehicle; as well as The object determination module is configured to determine, in response to the deviation degree satisfying a deviation condition, that the stationary object corresponding to the first detection point is an object outside the road fence.

13. An electronic device comprising: at least one processor; as well as A memory coupled to the at least one processor and having instructions stored thereon, the instructions, when executed by the at least one processor, causing the apparatus to perform the method according to any one of claims 1-11.

14. A vehicle comprising: radar; as well as The electronic device according to claim 13.

15. A computer-readable storage medium having computer-executable instructions stored thereon, wherein the computer-executable instructions are executed by a processor to implement the method according to any one of claims 1 to 11.

Images