Obstacle positioning method, system, device, and medium

Abstract

The application provides an obstacle positioning method, system, device and medium, comprising: for each APA ultrasonic radar sensor, acquiring ultrasonic radar data of an obstacle identified in a continuous driving process of a vehicle based on the APA ultrasonic radar sensor, and acquiring a sequence of original radar sensor coordinates based on the ultrasonic radar data; taking each original radar sensor coordinate point in the sequence of radar sensor coordinates as a center, and taking a distance of an obstacle actually detected by the APA ultrasonic radar sensor at the moment as a radius, a virtual circle is constructed respectively; calculating a geometric intersection point of two virtual circles at adjacent moments to obtain an intersection coordinate; using a radar beam angle to constrain the intersection coordinate to screen out effective obstacle coordinate points identified by the APA ultrasonic radar sensor, and performing checking and thinning operations on the effective obstacle coordinate points respectively to obtain an obstacle coordinate point. The obstacle positioning method of the application can obtain more accurate obstacle coordinate points for obstacle positioning.

Description

Technical Field

[0001] This application belongs to the field of automatic parking technology, and relates to an obstacle positioning method, and in particular to an obstacle positioning method, system, device and medium. Background Technology

[0002] Ultrasonic radar, as a mature environmental perception sensor, has been widely used in fields such as automatic parking, assisted driving, and robot navigation due to its advantages of low cost, stable technology, and insensitivity to lighting conditions. Its basic working principle is as follows: a transmitter emits ultrasonic waves of a specific frequency; these waves are reflected when they encounter obstacles during propagation; and a receiver captures the echo signal. By measuring the transit time between transmission and reception and multiplying it by the speed of sound, the distance from the radar to the obstacle can be calculated. With the continuous improvement in the accuracy requirements of environmental perception in intelligent driving systems, relying solely on single-point distance measurement is no longer sufficient to meet the obstacle recognition needs in complex scenarios. In recent years, the industry has begun to explore using the memory of obstacle coordinates obtained by ultrasonic radar at various scanning positions, and integrating the radar's installation pose on the vehicle with the vehicle's positional relationship before and after movement. This transforms multiple ranging points into a unified coordinate system, thereby piecing together point cloud data reflecting the outer contour of obstacles to support subsequent tasks such as target recognition, parking space detection, and path planning. However, the obstacle point cloud generated by this method has obvious shortcomings in positioning accuracy: the ultrasonic radar has poor ranging accuracy in non-direct directions, and the single-point measurement itself has low angular resolution, making it impossible to accurately obtain the orientation information of the obstacle edge.

[0003] The above factors together make it difficult for existing ultrasonic point cloud generation methods based on memory coordinate fusion to accurately restore the true outline and spatial position of obstacles, thus failing to meet the actual requirements of high-precision automatic parking and complex scene perception. Summary of the Invention

[0004] This application provides an obstacle localization method, system, device, and medium to solve the problems of existing technologies that rely on specific obstacle edge shapes and require a large amount of computing resources.

[0005] Firstly, this application provides an obstacle localization method. The obstacle localization method includes: for each APA ultrasonic radar sensor, acquiring ultrasonic radar data of obstacles identified during continuous vehicle operation based on the APA ultrasonic radar sensor, and acquiring an original radar sensor coordinate sequence based on the ultrasonic radar data; constructing virtual circles with each original radar sensor coordinate point in the radar sensor coordinate sequence as the center and the actual obstacle distance detected by the APA ultrasonic radar sensor at that moment as the radius; calculating the geometric intersection of two virtual circles at adjacent moments to obtain the intersection coordinates; constraining the intersection coordinates using radar beam angles to filter out valid obstacle coordinate points identified by the APA ultrasonic radar sensor, wherein when the intersection coordinates are simultaneously within the coverage area of ​​the radar beam angles at two adjacent moments, they are determined to be valid obstacle coordinate points; converting the valid obstacle coordinate points acquired by different APA ultrasonic radar sensors to the same coordinate system, and performing verification and separation operations on the valid obstacle coordinate points to obtain the actual obstacle coordinate points.

[0006] In one implementation of the first aspect, obtaining the intersection coordinates includes: calculating an intermediate quantity based on the intersection coordinates, the center distance, the radius of the virtual circle at the previous moment, and the radius of the virtual circle at the current moment, wherein the intermediate quantity is used to characterize intermediate parameters for calculating the intersection coordinates; and calculating the intersection coordinates using the original radar sensor coordinates at the previous moment, the original radar sensor coordinates at the current moment, and the intermediate quantity.

[0007] In one implementation of the first aspect, constraining the intersection coordinates using the radar beam angle to filter out valid obstacle coordinates identified by the APA ultrasonic radar sensor includes: applying a first constraint to the intersection coordinates based on the obstacle distance actually detected by the APA ultrasonic radar sensor at the previous moment and the forward mounting angle between the APA ultrasonic radar sensor and the vehicle; applying a second constraint to the intersection coordinates based on the obstacle distance actually detected by the APA ultrasonic radar sensor at the current moment and the forward mounting angle between the APA ultrasonic radar sensor and the vehicle; wherein the first constraint and the second constraint are used to characterize the coverage range of the radar beam angle at the corresponding moment, and when the intersection point is simultaneously within the coverage range of the radar beam angle at two adjacent moments, the intersection point is determined to be a valid obstacle coordinate point.

[0008] In one implementation of the first aspect, the verification operation of the valid obstacle coordinate points to obtain obstacle coordinate points includes: determining whether the slope between any valid obstacle coordinate point and the remaining valid obstacle coordinate points within a defined window is within a preset slope interval; if so, it is marked as a neighbor point; and / or determining whether the Euclidean distance between any valid obstacle coordinate point and the remaining valid obstacle coordinate points within the defined window is less than a preset maximum direct neighbor distance; if so, it is marked as a direct neighbor point; counting the number of neighbor points and direct neighbor points of any valid obstacle coordinate point; if the number of neighbor points is greater than 2 or the number of direct neighbor points is greater than 1, the valid obstacle coordinate point is confirmed as the obstacle coordinate point; otherwise, the valid obstacle coordinate point is discarded.

[0009] In one implementation of the first aspect, the separation operation of the effective obstacle coordinate points to obtain the obstacle coordinate points includes: during vehicle travel, taking the coordinates of the first detected effective obstacle coordinate point as the starting coordinates of the obstacle, and taking the coordinates of the last detected effective obstacle coordinate point as the ending coordinates of the obstacle; using a fixed grid length to evenly divide the grid based on a coordinate axis direction of the starting and ending coordinates of the obstacle, and selecting a fixed number of points in each grid as the obstacle coordinate points.

[0010] In one implementation of the first aspect, the obstacle localization method further includes: simultaneously acquiring data from each UPA ultrasonic radar during vehicle operation; performing curve fitting processing on the UPA ultrasonic radar data and the corresponding APA ultrasonic radar data respectively, and obtaining the intersection points of the corresponding curves based on the installation position of the radar, which are respectively used as obstacle corner points.

[0011] In one implementation of the first aspect, the process of performing curve fitting processing on the UPA ultrasonic radar data and the corresponding APA ultrasonic radar data, and obtaining the corresponding curve intersection points based on the radar installation position, respectively, as obstacle corner points, includes: performing curve fitting processing on the forward radar data in the UPA ultrasonic radar data and the forward radar data in the APA ultrasonic radar data, and obtaining the curve intersection points as obstacle start points; performing curve fitting processing on the backward radar data in the UPA ultrasonic radar data and the backward radar data in the APA ultrasonic radar data, and obtaining the curve intersection points as obstacle end points; and supplementing the obstacle start points and obstacle end points as obstacle corner points as obstacle coordinate points.

[0012] In one implementation of the first aspect, the obstacle localization method further includes: clustering the obstacle coordinate points using a clustering method to obtain at least one obstacle cluster; calculating the overlap between the obstacle cluster and the obstacle corner points to obtain a degree of coincidence; determining whether the degree of coincidence is less than a preset degree of coincidence threshold; if not, confirming them as the same obstacle; and using the obstacle corner points to filter the obstacle coordinate points of the obstacle cluster to obtain the obstacle coordinate points; if so, confirming them as different obstacles.

[0013] Secondly, this application provides an obstacle positioning system. The obstacle localization system includes: a raw calculation module configured to, for each APA ultrasonic radar sensor, acquire ultrasonic radar data of obstacles identified during continuous vehicle driving based on the APA ultrasonic radar sensor, and acquire a raw radar sensor coordinate sequence based on the ultrasonic radar data; a virtual circle construction module configured to construct virtual circles with each raw radar sensor coordinate point in the radar sensor coordinate sequence as the center and the actual obstacle distance detected by the APA ultrasonic radar sensor at that moment as the radius; an intersection point acquisition module configured to calculate the geometric intersection point of two virtual circles at adjacent moments to acquire the intersection point coordinates; a constraint processing module configured to constrain the intersection point coordinates using radar beam angles to filter out valid obstacle coordinate points identified by the APA ultrasonic radar sensors, wherein when the intersection point coordinates are simultaneously within the coverage area of ​​the radar beam angles at two adjacent moments, they are determined to be valid obstacle coordinate points; and a verification and isolation module configured to convert valid obstacle coordinate points acquired by different APA ultrasonic radar sensors to the same coordinate system, and perform verification and isolation operations on the valid obstacle coordinate points to acquire obstacle coordinate points.

[0014] Thirdly, this application provides an electronic device. The electronic device includes: a memory for storing a computer program; and a processor for executing the computer program stored in the memory to cause the electronic device to perform the obstacle localization method as described in any one of the first aspects.

[0015] Fourthly, this application provides a computer-readable storage medium. The computer-readable storage medium stores a computer program that, when executed by a processor, implements the obstacle localization method described in any one of the first aspects.

[0016] As described above, the obstacle location method, system, device, and medium of this application have the following beneficial effects:

[0017] The coordinates of the virtual circle intersection points constructed from the radar sensor coordinates at adjacent time points are constrained and used as valid obstacle coordinate points. This avoids the complicated calibration process, provides better interpretability, filters out invalid pseudo-points generated by geometric calculations, and retains obstacle boundary points that conform to the laws of physical detection, thereby improving the authenticity and reliability of the data.

[0018] By using neighbor verification and separation operations, the density of obstacle coordinate points in overly dense areas is reduced, making the overall distribution of obstacle coordinate points more uniform, reducing redundant data, and improving the processing efficiency and accuracy of subsequent algorithms.

[0019] Furthermore, by integrating radar data from both the ultrasonic parking assist system and the automatic parking assist system, random errors are effectively reduced, providing precise obstacle coordinates for key functions such as parking path planning and vehicle control. Attached Figure Description

[0020] Figure 1 The diagram shown illustrates an application scenario of the obstacle localization method described in this application.

[0021] Figure 2 The diagram shown is a structural schematic of the end-to-cloud interaction scenario described in the embodiments of this application.

[0022] Figure 3 The diagram shown is a flowchart illustrating the obstacle localization method described in an embodiment of this application.

[0023] Figure 4 The diagram shown is a schematic representation of the original radar sensor coordinates as described in the embodiments of this application.

[0024] Figure 5 The diagram shown illustrates the point cloud verification and alienation operations described in the embodiments of this application.

[0025] Figure 6 The diagram shown is a schematic representation of curve fitting as described in an embodiment of this application.

[0026] Figure 7 The diagram shown is a structural schematic of the obstacle location system described in an embodiment of this application.

[0027] Figure 8 The diagram shown is a structural schematic of the electronic device described in an embodiment of this application.

[0028] Component designation explanation

[0029] 1 obstacle positioning device 11 radar equipment 12 Local processor 13 Display terminal 2 End-to-Cloud Interaction System 20 terminal 21 cloud server 100 Obstacle positioning system 110 Original calculation module 120 Virtual circle building block 130 Intersection acquisition module 140 Constraint processing module 150 Verification of the alienation module 800 electronic devices 810 memory 820 processor 830 monitor S11~S15 step Detailed Implementation

[0030] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0031] It should be noted that in the embodiments of this application, the words "optionally" or "for example" indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "optionally" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "optionally" or "for example" is intended to present the relevant concepts in a specific manner.

[0032] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0033] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. Therefore, the drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0034] The beam angle of an ultrasonic radar can be approximated as a fan-shaped area. As a vehicle equipped with ultrasonic radar gradually approaches an obstacle and eventually enters the radar's beam angle range, the radar's detection range changes, and the obstacle's coordinates can be obtained through mathematical formulas. This point cloud generation method, which approximates the normal direction, inflates the actual size of the obstacle. To address this issue, common industry practices include: 1. Accuracy calibration based on different obstacle types and distances; 2. Estimating the obstacle's outline based on the changing trends of memorized points. Existing methods suffer from dependence on specific obstacle edge shapes, require significant computational resources, and cannot be adequately expressed mathematically or are not precise enough.

[0035] To address at least the aforementioned problems, this application provides an obstacle localization method. The obstacle localization method includes: for each APA ultrasonic radar sensor, acquiring ultrasonic radar data of obstacles identified during continuous vehicle operation based on the APA ultrasonic radar sensor, and acquiring an original radar sensor coordinate sequence based on the ultrasonic radar data; constructing virtual circles with each original radar sensor coordinate point in the radar sensor coordinate sequence as the center and the actual obstacle distance detected by the APA ultrasonic radar sensor at that moment as the radius; calculating the geometric intersection of two virtual circles at adjacent moments to obtain the intersection coordinates; constraining the intersection coordinates using the radar beam angle to filter out valid obstacle coordinate points identified by the APA ultrasonic radar sensor, wherein when the intersection coordinates are simultaneously within the coverage area of ​​the radar beam angle at two adjacent moments, they are determined to be valid obstacle coordinate points; converting the valid obstacle coordinate points acquired by different APA ultrasonic radar sensors to the same coordinate system, and performing verification and separation operations on the valid obstacle coordinate points to obtain the actual obstacle coordinate points.

[0036] In this embodiment, the coordinates of the virtual circular intersection points constructed from the radar sensor coordinates at adjacent time points are constrained and used as valid obstacle coordinate points. This avoids the complicated calibration process, provides better interpretability, filters out invalid pseudo-points generated by geometric calculations, retains obstacle boundary points that conform to the laws of physical detection, and improves the authenticity and reliability of the data.

[0037] Figure 1 This diagram illustrates an application scenario of the obstacle location method described in this application. The obstacle location device 1 can be used to implement the obstacle location method provided in the embodiments of this application, but the application scenarios of the obstacle location method provided in the embodiments of this application are not limited to this. Figure 1 The obstacle positioning device 1 shown is an example. Figure 1 As shown, the obstacle location device 1 includes a radar device 11, a local processor 12, and a display terminal 13. The obstacle location method provided in this embodiment can be applied to the local processor 12.

[0038] in, Figure 1 The local processor 12 can be a single local processor, a cluster of multiple local processors, or a cloud computing center, etc., and is not specifically limited here. Although Figure 1 Only one radar device 11, one local processor 12, and one display terminal 13 are shown, but it should be understood that... Figure 1 The examples in this paper are only for understanding this solution. The specific number of radar devices 11, local processors 12 and display terminals 13 should be flexibly determined based on the actual situation.

[0039] In some other implementations, the obstacle location device 1 may not include the display terminal 13, but only a local processor 12 with display function and a radar device 11. The obstacle location method provided in this application embodiment can be applied to the local processor 12. The local processor 12 with display function may include tablet computers, PDAs, mobile phones, personal computers and voice interaction devices, or it may be a monitoring device, a face recognition device, etc., which are not limited here.

[0040] In some other implementations, the obstacle localization method described in this application can be applied to edge-cloud interaction scenarios. Figure 2 The diagram shown is a structural schematic of the end-to-cloud interaction scenario described in an embodiment of this application. For example... Figure 2 As shown, the terminal-cloud interaction system 2 includes a terminal 20 and a cloud server 21. The terminal 20 and the cloud server 21 can communicate with each other, and the communication method is not limited to wired or wireless.

[0041] The terminal 20 can be mobile or fixed. For example, it can be a wireless terminal or a wired terminal. A wireless terminal can refer to a device with wireless transceiver capabilities and can be deployed indoors, outdoors, and in industrial workshops. The terminal 20 can be a vehicle-mounted terminal, a mobile phone, a tablet computer, a laptop computer, etc., and is not limited thereto. The cloud server 21 can include one or more servers, or one or more processing nodes, or one or more virtual machines running on the server. The cloud server 21 can also be referred to as a server cluster, a management platform, a data processing center, etc., and is not limited thereto in this embodiment.

[0042] The technical solutions in the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0043] The following embodiments of this application provide an obstacle location method, which, for example, can be achieved through... Figure 1 The local processor 12 shown or Figure 2 The cloud server 21 shown is used to implement this. Figure 3The diagram shown is a flowchart illustrating the obstacle localization method described in an embodiment of this application. Figure 3 As shown, the obstacle localization method includes steps S11 to S15.

[0044] Step S11: For each APA ultrasonic radar sensor, acquire ultrasonic radar data of obstacles identified during continuous vehicle driving based on the APA ultrasonic radar sensor, and acquire the original radar sensor coordinate sequence based on the ultrasonic radar data.

[0045] Step S12: Using each original radar sensor coordinate point in the radar sensor coordinate sequence as the center and the actual obstacle distance detected by the APA ultrasonic radar sensor at that moment as the radius, construct virtual circles respectively.

[0046] Step S13: Calculate the geometric intersection of the two virtual circles at adjacent time points and obtain the coordinates of the intersection point.

[0047] Step S14: Constrain the intersection coordinates using the radar beam angle to filter out the valid obstacle coordinates identified by the APA ultrasonic radar sensor. Specifically, when the intersection coordinates are simultaneously within the coverage area of ​​the radar beam angle at two adjacent moments, they are determined to be valid obstacle coordinates.

[0048] Step S15: Convert the valid obstacle coordinate points obtained by different APA ultrasonic radar sensors to the same coordinate system, and perform verification and isolation operations on the valid obstacle coordinate points to obtain the obstacle coordinate points.

[0049] In some possible implementations, the obstacle localization method of this application is applicable to vehicles equipped with an Auto Parking Assist (APA) radar system that detects the coordinates of obstacles formed by objects in front, behind, to the left, or to the right. For each APA ultrasonic radar sensor, ultrasonic radar data of obstacles identified during continuous vehicle driving is acquired based on the APA ultrasonic radar sensor, and the original radar sensor coordinate sequence is obtained based on the ultrasonic radar data. Specifically, taking the radar detection of the right front APA of the vehicle as an example, during the forward driving of the vehicle, when the vehicle's operating condition is determined to meet the preset data storage conditions, the APA stores radar data once in the buffer, with a maximum of K cached times. The radar data includes the radar sensor's horizontal and vertical coordinates and the obstacle distance. A coordinate system is established with the vehicle's center point as the origin, the vehicle's forward direction as the positive X-axis direction, and 90 degrees counterclockwise from the positive X-axis direction as the positive Y-axis direction. The radar data of a time-based automatic parking assist system is converted using an approximate normal direction to obtain the original radar sensor coordinates. When new radar data is generated, it is combined with the original radar sensor coordinates obtained from the radar data of the previous moment to form two virtual circles. The virtual circles are centered on the original radar sensor coordinates of the adjacent moment and have a radius equal to the actual obstacle distance detected by the APA ultrasonic radar sensor at that moment. The geometric intersection of the two virtual circles is calculated to obtain the intersection coordinates. Since the radar sensor has a certain beam angle (horizontal opening angle), its actual detection range is a fan-shaped area. The intersection coordinates are constrained using the radar beam angle to filter out the valid obstacle coordinates identified by the APA ultrasonic radar sensor. Specifically, when the intersection coordinates are simultaneously within the coverage area of ​​the radar beam angle of two adjacent moments, it is determined to be a valid obstacle coordinate point. The valid obstacle coordinates obtained by different APA ultrasonic radar sensors are transformed to the same coordinate system, and the valid obstacle coordinates are verified and separated to obtain the actual obstacle coordinates.

[0050] In some other possible implementations, Figure 4 The diagram shows the original radar sensor coordinates as described in this embodiment. During vehicle operation, for each APA ultrasonic radar sensor, APA radar data is acquired based on either the distance difference or the time difference between the vehicle's travel and the distance at which the radar data was last stored. The distance difference is the difference between the vehicle's travel distance and the distance at which the radar data was last stored. ≥Preset driving distance difference The driving time condition is the difference between the vehicle's driving time and the time when the radar data was last stored. ≥Preset travel time difference . The value is related to parameters such as radar refresh time, maximum speed allowed for vehicle operation, and program execution cycle; for example, the typical range is 3cm-10cm. (Distance difference) The calculation formula is:

[0051] ,

[0052] The integer symbol, The vehicle's speed at the current moment. This is the time difference between the vehicle's current moment and the last time the radar data was stored.

[0053] Preset travel time difference The value is related to parameters such as radar refresh time, maximum speed allowed for vehicle operation, and program execution cycle. For example, the general range is 20ms-100ms.

[0054] like Figure 4 As shown, for time-based radar data streams Based on the vehicle's driving direction, the positional relationship between the radar and the vehicle's center point, and the geometric relationship between the radar and obstacles, the original radar sensor coordinates are obtained by performing a conversion using an approximate normal direction. , These are the intersection coordinates calculated based on the original radar sensor coordinates.

[0055] In this embodiment, the coordinates of the virtual circular intersection points constructed from the radar sensor coordinates at adjacent time points are constrained and used as valid obstacle coordinate points. This avoids the complicated calibration process, provides better interpretability, filters out invalid pseudo-points generated by geometric calculations, retains obstacle boundary points that conform to the laws of physical detection, and improves the authenticity and reliability of the data.

[0056] In one embodiment of this application, obtaining the intersection coordinates includes: calculating an intermediate quantity based on the intersection coordinates, the center distance, the radius of the virtual circle at the previous moment, and the radius of the virtual circle at the current moment, wherein the intermediate quantity is used to characterize the intermediate parameters for calculating the intersection coordinates; and calculating the intersection coordinates using the original radar sensor coordinates at the previous moment, the original radar sensor coordinates at the current moment, and the intermediate quantity.

[0057] In some possible implementations, a virtual circle is drawn with the original radar sensor coordinates from the previous moment as the center and the distance to the obstacle detected by the radar at the previous moment as the radius, resulting in a virtual circle for the previous moment. Similarly, a virtual circle is drawn with the original radar sensor coordinates from the current moment as the center and the distance to the obstacle detected by the radar at the current moment as the radius, resulting in a virtual circle for the current moment. The center distance is calculated based on the center coordinates of the virtual circle from the previous moment and the center coordinates of the virtual circle from the current moment. The formula for calculating the center distance is:

[0058] ,

[0059] These are the coordinates of the center of the virtual circle at the current moment. The coordinates of the center of the virtual circle at the previous moment.

[0060] The radius of the virtual circle at the current moment is The radius of the virtual circle at the previous moment was If the distance between the centers or If this is confirmed, then the two circles do not intersect, and the calculation of the coordinates of other original radar sensors is repeated. and If the two circles coincide, the calculation of coordinates for other original radar sensors is resumed. If the two circles intersect, the intersection coordinates are calculated. Based on the intersection coordinates, the distance between the circle centers, the radius of the virtual circle from the previous time step, and the radius of the virtual circle from the current time step, intermediate values ​​are obtained. These intermediate values ​​characterize the intermediate parameters used to calculate the intersection coordinates. The formula for calculating the intermediate values ​​is:

[0061] ,

[0062] Let be the radius of the virtual circle at the current moment. Let be the radius of the virtual circle at the previous moment. This represents the distance between the center of the virtual circle at the current moment and the center of the virtual circle at the previous moment.

[0063] The formula for calculating the coordinates of the midpoint of the line connecting two intersection points is:

[0064] ,

[0065] The formula for calculating the coordinates of the intersection point is:

[0066] ,

[0067] and Let be the coordinates of the intersection point of the two virtual circles. As an intermediate quantity, This represents the distance between the centers of the virtual circle at the current moment and the virtual circle at the previous moment. These are the coordinates of the center of the virtual circle at the current moment. Let be the coordinates of the center of the virtual circle at the previous moment. The coordinates of the midpoint of the line connecting the two intersection points.

[0068] Radar sensors have a certain beam angle (horizontal opening angle), and their actual detection range is a fan-shaped area. To ensure that newly added intersection points conform to the laws of physical detection, beam angle constraints must be applied: an intersection point is considered a valid point cloud only when it is simultaneously within the coverage area of ​​the radar beam angle at two adjacent moments. In one embodiment of this application, constraining the intersection point coordinates using the radar beam angle to filter out valid obstacle coordinate points identified by the APA ultrasonic radar sensor includes: applying a first constraint to the intersection point coordinates based on the obstacle distance actually detected by the APA ultrasonic radar sensor at the previous moment and the forward mounting angle between the APA ultrasonic radar sensor and the vehicle; applying a second constraint to the intersection point coordinates based on the obstacle distance actually detected by the APA ultrasonic radar sensor at the current moment and the forward mounting angle between the APA ultrasonic radar sensor and the vehicle; wherein the first and second constraints are used to characterize the coverage area of ​​the radar beam angle at the corresponding moment, and the intersection point is determined to be a valid obstacle coordinate point when it is simultaneously within the coverage area of ​​the radar beam angle at two adjacent moments.

[0069] In some possible implementations, at a horizontal angle Taking the radar beam angle as an example, the first constraint is applied to the intersection point coordinates based on the obstacle distance detected by the APA ultrasonic radar sensor at the previous moment and the forward mounting angle between the APA ultrasonic radar sensor and the vehicle. The second constraint is applied to the intersection point coordinates based on the obstacle distance detected by the APA ultrasonic radar sensor at the current moment and the forward mounting angle between the APA ultrasonic radar sensor and the vehicle. The constraint conditions are:

[0070] ,

[0071] The radar beam angle is the horizontal angle. The current distance to the obstacle detected by the APA ultrasonic radar sensor. The angle at which the APA ultrasonic radar is mounted facing forward of the vehicle. The coordinates of the virtual circle's center at the current moment.

[0072] Based on the first and second constraints, the intersection point coordinates are filtered. When the intersection point is simultaneously within the coverage area of ​​the radar beam angle at two adjacent time points, it is determined to be a valid obstacle coordinate point. The intersection point calculated from the nth stored radar data and the (n-1)th stored radar data must simultaneously satisfy the constraints of both times. Assume that the radar data acquired by the right front APA radar at the current time is... The radar data at the previous moment was The coordinates of the intersection point of the virtual circles at adjacent time points are calculated as follows: , The calculated ranges for the first and second constraints are as follows:

[0073] ,

[0074] Therefore, the intersection point can be seen. satisfy and Given the constraints, the effective obstacle coordinates are: .

[0075] Each radar in the APA system generates independent point cloud data as described above. However, due to the different coordinate system definitions of each radar, although coordinate transformation is used to unify the point clouds on both sides to the same reference coordinate system (such as the vehicle coordinate system), the point cloud often exhibits local over-density and isolated noise issues after unification. Therefore, neighbor verification is still required. In one embodiment of this application, the verification operation for valid obstacle coordinate points to obtain obstacle coordinate points includes: determining whether the slope between any valid obstacle coordinate point and the remaining valid obstacle coordinate points within a defined window is within a preset slope interval; if so, it is marked as a neighbor point; and / or determining whether the Euclidean distance between any valid obstacle coordinate point and the remaining valid obstacle coordinate points within the defined window is less than a preset maximum direct neighbor distance; if so, it is marked as a direct neighbor point; and counting the number of neighbor points and direct neighbor points of any valid obstacle coordinate point. If the number of neighbor points is greater than 2 or the number of direct neighbor points is greater than 1, the valid obstacle coordinate point is confirmed as an obstacle coordinate point; otherwise, the valid obstacle coordinate point is discarded.

[0076] Among some possible implementations, Figure 5 The diagram shown illustrates the point cloud verification and alienation operations described in the embodiments of this application. Figure 5 As shown, after unifying the effective obstacle coordinates of the front-side radar and the rear-side radar into the same coordinate system for clustering, problems such as excessively dense local point clouds or isolated points may occur. Figure 5 The red dots represent the ideal distribution of effective obstacle coordinates, while the black dots represent redundant effective obstacle coordinates.

[0077] For any valid obstacle coordinate point ,judge The slope of the remaining valid obstacle coordinates within the limited window area Is it within the preset range? If it is within the range, then mark it as a neighboring point. Among these, the maximum slope... The typical value is 0.75, the minimum slope. The value is typically -0.75. The defined window range W is related to the vehicle speed, and is typically set between 20cm and 70cm.

[0078] judge Euclidean distance to the coordinates of the remaining valid obstacles within the limited window area Is it less than the preset maximum direct neighbor distance? If so, it is marked as a direct neighbor. Here, Euclidean distance... The calculation formula is:

[0079] ,

[0080] For any valid obstacle coordinate point coordinates The coordinates of the remaining valid obstacle points within the limited window area.

[0081] The number of neighboring points and direct neighboring points of any valid obstacle coordinate point is counted. If the number of neighboring points is greater than 2 or the number of direct neighboring points is greater than 1, the valid obstacle coordinate point is confirmed as an obstacle coordinate point; otherwise, the valid obstacle coordinate point is removed.

[0082] In one embodiment of this application, the process of separating valid obstacle coordinate points to obtain obstacle coordinate points includes: during vehicle travel, taking the coordinates of the first detected valid obstacle coordinate point as the starting coordinates of the obstacle and the coordinates of the last detected valid obstacle coordinate point as the ending coordinates of the obstacle; and using a fixed grid length to evenly divide the grid based on a coordinate axis direction of the starting and ending coordinates of the obstacle, selecting a fixed number of points in each grid as obstacle coordinate points.

[0083] In some possible implementations, a fixed grid length is used. The coordinates of the obstacles are evenly divided along a single coordinate axis based on the starting and ending coordinates of the obstacles. A fixed number of points are selected from each section as obstacle coordinates. Taking forward vehicle travel as an example, the effective obstacle coordinate with the smallest horizontal coordinate is the starting coordinate, and the effective obstacle coordinate with the largest horizontal coordinate is the ending coordinate. A separation operation is then performed on the effective obstacle coordinates within the range of the starting and ending coordinates. A fixed grid length is used. The value is generally taken as 10cm-30cm.

[0084] In this embodiment, neighbor verification and alienation operations are used to reduce the density of obstacle coordinate points in overly dense areas, making the overall distribution of obstacle coordinate points more uniform, reducing redundant data, and improving the processing efficiency and accuracy of subsequent algorithms.

[0085] In one embodiment of this application, the obstacle localization method further includes: simultaneously acquiring data from each UPA ultrasonic radar during vehicle operation; performing curve fitting processing on the UPA ultrasonic radar data and the corresponding APA ultrasonic radar data respectively, and obtaining the intersection points of the corresponding curves based on the installation position of the radar, which are respectively used as obstacle corner points.

[0086] In some possible implementations, ultrasonic radar data from each Ultrasonic Parking Assist (UPA) system is acquired synchronously during vehicle operation. The UPA ultrasonic radar data and the corresponding APA ultrasonic radar data are then subjected to curve fitting processing, and the intersection points of the corresponding curves are obtained based on the radar's installation location, serving as the corner points of obstacles.

[0087] In one embodiment of this application, the obstacle localization method further includes: performing curve fitting processing on the UPA ultrasonic radar data and the corresponding APA ultrasonic radar data respectively, and obtaining the corresponding curve intersection points according to the installation position of the radar, which are respectively used as obstacle corner points. This includes: performing curve fitting processing on the forward radar data in the UPA ultrasonic radar data and the forward radar data in the APA ultrasonic radar data respectively, and obtaining the curve intersection points as obstacle start points; performing curve fitting processing on the backward radar data in the UPA ultrasonic radar data and the backward radar data in the APA ultrasonic radar data respectively, and obtaining the curve intersection points as obstacle end points. The obstacle start point and the obstacle end point are used as obstacle corner points to supplement the obstacle coordinate points.

[0088] In some possible implementations, obstacle corner points are utilized to further improve obstacle positioning accuracy. Figure 6 The diagram shown is a schematic representation of curve fitting as described in an embodiment of this application. Figure 6 As shown, during the vehicle's forward movement, radar data from the forward ultrasonic radar UPA, the rearward UPA, and the APA ultrasonic radar are acquired. The original radar sensor coordinates stored in the forward UPA are represented by circles, and the original radar sensor coordinates stored in the forward APA are represented by solid black dots. Curve L2 corresponding to the UPA radar data and curve L1 corresponding to the APA radar data are obtained using the least squares method. The intersection point of the two curves is calculated, and shown in red in the figure as the starting point of the obstacle. L1 and L2 are quadratic curves, and their equations are:

[0089] ,

[0090] The coefficients of the quadratic term determine the orientation of the conic section and the degree of curvature of the conic curve. The coefficient of the linear term represents the slope of the tangent line to the curve at x=0, reflecting the overall inclination of the curve. This is a constant term.

[0091] Similarly, the radar data of the backward UPA and the backward APA are respectively processed by curve fitting using the least squares method to obtain the intersection point of the curves as the obstacle endpoint. The obstacle start point and obstacle endpoint are supplemented as obstacle corner points as obstacle coordinate points.

[0092] In this embodiment, radar data from the ultrasonic parking assist system and the automatic parking assist system are integrated to effectively reduce random errors and provide accurate obstacle coordinates for key functions such as parking path planning and vehicle control.

[0093] In one embodiment of this application, the obstacle localization method further includes: clustering obstacle coordinate points using a clustering method to obtain at least one obstacle cluster; calculating the overlap between the obstacle cluster and obstacle corner points to obtain the degree of overlap; determining whether the degree of overlap is less than a preset degree of overlap threshold; if not, confirming them as the same obstacle; and using obstacle corner points to filter the obstacle coordinate points of the obstacle cluster to obtain obstacle coordinate points; if so, confirming them as different obstacles.

[0094] In some possible implementations, clustering methods are used to cluster obstacle coordinates to obtain multiple obstacle clusters. The overlap between these clusters and obstacle corner points is calculated to obtain an overlap degree P. It is then determined whether the overlap degree P is less than a preset overlap threshold. If it is, the obstacles are identified as different obstacles; otherwise, they are identified as the same obstacle. The obstacle coordinates within each cluster are then filtered using obstacle corner points to obtain the final obstacle coordinates. The preset overlap threshold is typically set to 50%.

[0095] Figure 7 The diagram shown is a structural schematic of the obstacle location system described in an embodiment of this application. Figure 7 As shown, the obstacle localization system 100 includes an original calculation module 110, a virtual circle construction module 120, an intersection point acquisition module 130, a constraint processing module 140, and a verification and isolation module 150.

[0096] The raw calculation module 110 is configured to, for each APA ultrasonic radar sensor, acquire ultrasonic radar data of obstacles identified during continuous vehicle driving based on the APA ultrasonic radar sensor, and acquire the raw radar sensor coordinate sequence based on the ultrasonic radar data.

[0097] The virtual circle construction module 120 is configured to construct virtual circles with each original radar sensor coordinate point in the radar sensor coordinate sequence as the center and the distance to the obstacle actually detected by the APA ultrasonic radar sensor at that moment as the radius.

[0098] The intersection point acquisition module 130 is configured to calculate the geometric intersection point of two virtual circles at adjacent time points and obtain the intersection point coordinates.

[0099] The constraint processing module 140 is configured to constrain the intersection coordinates using the radar beam angle to filter out the valid obstacle coordinates identified by the APA ultrasonic radar sensor. Specifically, when the intersection coordinates are simultaneously within the coverage area of ​​the radar beam angle at two adjacent moments, they are determined to be valid obstacle coordinates.

[0100] The verification and isolation module 150 is configured to convert the valid obstacle coordinate points obtained by different APA ultrasonic radar sensors to the same coordinate system, and to perform verification and isolation operations on the valid obstacle coordinate points to obtain the obstacle coordinate points.

[0101] It should be noted that the obstacle positioning system 100 includes modules 110 to 150 as described above. Figure 3 Steps S11 to S15 in the obstacle localization method shown correspond one-to-one, and will not be elaborated here.

[0102] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, or methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules / units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of apparatuses or modules or units may be electrical, mechanical, or other forms.

[0103] The modules / units described as separate components may or may not be physically separate. The components shown as modules / units may or may not be physical modules; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules / units can be selected to achieve the objectives of the embodiments of this application, depending on actual needs. For example, the functional modules / units in the various embodiments of this application may be integrated into one processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into one module / unit.

[0104] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0105] This application also provides an electronic device. Figure 8 The diagram shown is a structural schematic of the electronic device described in an embodiment of this application. Figure 8 As shown, in this embodiment, the electronic device 800 includes a memory 810 and a processor 820.

[0106] The memory 810 is used to store computer programs; preferably, the memory 810 includes various media that can store program code, such as ROM, RAM, magnetic disk, USB flash drive, memory card or optical disk.

[0107] Specifically, memory 810 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory. Electronic device 800 may further include other removable / non-removable, volatile / non-volatile computer system storage media. Memory 810 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this application. It is understood that memory 810 may be volatile memory or non-volatile memory, or both. Non-volatile memory may be read-only memory (ROM) or programmable read-only memory (PROM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM) and synchronous static random access memory (SSRAM). The memories described in the embodiments of this invention are intended to include, but are not limited to, these and any other suitable categories of memory.

[0108] The processor 820 is connected to the memory 810 and is used to execute the computer program stored in the memory 810 so that the electronic device 800 performs the obstacle localization method described in any embodiment of this application.

[0109] Optionally, the processor 820 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0110] Optionally, the electronic device 800 in this embodiment may further include a display 830. The display 830 is communicatively connected to the memory 810 and the processor 820, and is used to display the relevant graphical user interface (GUI) of the obstacle localization method described in this application embodiment.

[0111] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, the program implements the obstacle localization method described in any embodiment of this application. Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing a processor. The program can be stored in a computer-readable storage medium, which is a non-transitory medium, such as random access memory, read-only memory, flash memory, hard disk, solid-state drive, magnetic tape, floppy disk, optical disk, and any combination thereof. The storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital video disc (DVD)), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0112] As used in this specification, the terms "component," "module," "system," etc., are used to refer to computer-related entities, hardware, firmware, combinations of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program, and / or a computer. As illustrated, applications running on computing devices and computing devices can both be components. One or more components may reside in a process and / or an execution thread, and components may be located on a single computer and / or distributed among two or more computers. Furthermore, these components can be executed from various computer-readable media on which various data structures are stored. Components can communicate, for example, via local and / or remote processes based on signals having one or more data packets (e.g., data from two components interacting with another component between a local system, a distributed system, and / or a network, such as the Internet interacting with other systems via signals).

[0113] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0114] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. An obstacle localization method, characterized in that, include: For each APA ultrasonic radar sensor, ultrasonic radar data of obstacles identified during continuous vehicle driving is acquired based on the APA ultrasonic radar sensor, and the original radar sensor coordinate sequence is acquired based on the ultrasonic radar data. Using each original radar sensor coordinate point in the radar sensor coordinate sequence as the center and the actual obstacle distance detected by the APA ultrasonic radar sensor at that moment as the radius, virtual circles are constructed respectively. Calculate the geometric intersection point of the two virtual circles at adjacent time points to obtain the coordinates of the intersection point; The intersection coordinates are constrained by the radar beam angle to filter out the valid obstacle coordinates identified by the APA ultrasonic radar sensor. When the intersection coordinates are simultaneously within the coverage area of ​​the radar beam angle at two adjacent times, they are determined to be valid obstacle coordinates. The effective obstacle coordinate points obtained by different APA ultrasonic radar sensors are transformed to the same coordinate system, and the effective obstacle coordinate points are verified and isolated to obtain the obstacle coordinate points.

2. The obstacle localization method according to claim 1, characterized in that, Obtaining the coordinates of the intersection point includes: An intermediate quantity is obtained by calculating the intersection point coordinates, the center distance, the radius of the virtual circle at the previous moment, and the radius of the virtual circle at the current moment. The intermediate quantity is used to characterize the intermediate parameters for calculating the intersection point coordinates. The intersection coordinates are obtained by using the original radar sensor coordinates of the previous moment, the original radar sensor coordinates of the current moment, and the intermediate quantity.

3. The obstacle localization method according to claim 1, characterized in that, The intersection point coordinates are constrained using the radar beam angle to filter out the valid obstacle coordinate points identified by the APA ultrasonic radar sensor, including: The first constraint is applied to the intersection point coordinates based on the obstacle distance actually detected by the APA ultrasonic radar sensor at the previous moment and the forward mounting angle of the APA ultrasonic radar sensor and the vehicle. The intersection point coordinates are subject to a second constraint based on the obstacle distance actually detected by the APA ultrasonic radar sensor at the current moment and the forward mounting angle of the APA ultrasonic radar sensor and the vehicle. The first constraint and the second constraint are used to characterize the coverage range of the radar beam angle at the corresponding time. When the intersection point is simultaneously within the coverage range of the radar beam angle at two adjacent times, the intersection point is determined to be a valid obstacle coordinate point.

4. The obstacle localization method according to claim 1, characterized in that, The valid obstacle coordinate points are verified to obtain the obstacle coordinate points, including: Determine whether the slope between any valid obstacle coordinate point and the remaining valid obstacle coordinate points within the defined window falls within a preset slope interval; if so, mark it as a neighbor point; and / or Determine whether the Euclidean distance between any valid obstacle coordinate point and the remaining valid obstacle coordinate points within the defined window is less than the preset maximum direct neighbor distance; if so, mark it as a direct neighbor point. The number of neighboring points and direct neighboring points of any valid obstacle coordinate point is counted. If the number of neighboring points is greater than 2 or the number of direct neighboring points is greater than 1, the valid obstacle coordinate point is confirmed as the obstacle coordinate point; otherwise, the valid obstacle coordinate point is removed.

5. The obstacle localization method according to claim 1, characterized in that, To obtain the obstacle coordinates by performing a separation operation on the effective obstacle coordinates, the following steps are included: During vehicle movement, the coordinates of the first detected valid obstacle point are taken as the starting coordinates of the obstacle, and the coordinates of the last detected valid obstacle point are taken as the ending coordinates of the obstacle. The grid is divided equally along a coordinate axis based on the starting and ending coordinates of the obstacle using a fixed grid length, and a fixed number of points are selected from each grid as the coordinate points of the obstacle.

6. The obstacle localization method according to claim 1, characterized in that, Also includes: During vehicle operation, data from each UPA ultrasonic radar is acquired simultaneously. The UPA ultrasonic radar data and the corresponding APA ultrasonic radar data are respectively subjected to curve fitting processing, and the intersection points of the corresponding curves are obtained according to the installation position of the radar, which are respectively used as the corner points of the obstacle.

7. The obstacle localization method according to claim 6, characterized in that, The UPA ultrasonic radar data and the corresponding APA ultrasonic radar data are respectively subjected to curve fitting processing, and the intersection points of the corresponding curves are obtained according to the installation position of the radar, which are respectively used as obstacle corner points, including: The forward radar data in the UPA ultrasonic radar data and the forward radar data in the APA ultrasonic radar data are respectively subjected to curve fitting processing, and the intersection of the curves is used as the starting point of the obstacle. The backward radar data in the UPA ultrasonic radar data and the backward radar data in the APA ultrasonic radar data are respectively subjected to curve fitting processing, and the intersection of the curves is taken as the endpoint of the obstacle. The starting point and ending point of the obstacle are used as corner points to supplement the coordinate points of the obstacle.

8. The obstacle localization method according to claim 6, characterized in that, Also includes: The obstacle coordinates are clustered using a clustering method to obtain at least one obstacle cluster. The overlap between the obstacle cluster and the corner points of the obstacle is calculated to obtain the degree of coincidence; If the overlap is less than a preset overlap threshold, it is confirmed as the same obstacle. The obstacle coordinates of the obstacle cluster are filtered using the corner points of the obstacle to obtain the obstacle coordinates. If they are the same, they are confirmed as different obstacles.

9. An obstacle location system, characterized in that, include: The original calculation module is configured to, for each APA ultrasonic radar sensor, acquire ultrasonic radar data of obstacles identified during continuous vehicle driving based on the APA ultrasonic radar sensor, and acquire the original radar sensor coordinate sequence based on the ultrasonic radar data; The virtual circle construction module is configured to construct virtual circles with each original radar sensor coordinate point in the radar sensor coordinate sequence as the center and the actual obstacle distance detected by the APA ultrasonic radar sensor at that moment as the radius. The intersection point acquisition module is configured to calculate the geometric intersection point of two virtual circles at adjacent times and obtain the intersection point coordinates; The constraint processing module is configured to constrain the intersection coordinates using the radar beam angle to filter out the valid obstacle coordinates identified by the APA ultrasonic radar sensor. Specifically, when the intersection coordinates are simultaneously within the coverage area of ​​the radar beam angle at two adjacent moments, they are determined to be valid obstacle coordinates. The verification and isolation module is configured to convert the valid obstacle coordinate points obtained by different APA ultrasonic radar sensors to the same coordinate system, and to perform verification and isolation operations on the valid obstacle coordinate points to obtain the obstacle coordinate points.

10. An electronic device, characterized in that, The electronic device includes: Memory, used to store computer programs; A processor for executing a computer program stored in the memory to cause the electronic device to perform the obstacle localization method as described in any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the obstacle localization method according to any one of claims 1 to 8.

Images