Obstacle point cloud identification method and system, electronic equipment and medium

Abstract

The invention provides an obstacle point cloud recognition method and system, electronic equipment and a medium, and the method comprises the steps: obtaining the ultrasonic radar data of an obstacle, carrying out the distance correction of the ultrasonic radar data of the obstacle, and obtaining the point cloud coordinate data of the obstacle detected at this time, the obstacle point cloud coordinate data comprises an obstacle point cloud coordinate and a detection distance; judging whether the obstacle point cloud coordinate data obtained this time meets a storage condition or not, if yes, storing the obstacle point cloud coordinate data, and if not, abandoning the obstacle point cloud coordinate data; reading obstacle point cloud coordinate data stored in a continuous time interval, and matching a corresponding identification type according to a detection distance change trend of multiple groups of obstacle point cloud coordinate data; and carrying out obstacle coordinate identification according to the corresponding identification type so as to obtain the position of the obstacle. According to the obstacle point cloud recognition method, the point cloud can be classified and recognized, and the accurate obstacle position is obtained.

Description

Technical Field

[0001] The present application belongs to the field of point cloud recognition technology, and relates to an obstacle point cloud recognition method, and in particular to an obstacle point cloud recognition method, system, electronic equipment and medium. Background Art

[0002] In recent years, with the advancement of intelligent driving technology, automatic parking systems have gradually become a standard feature in vehicles. Their core technology lies in high-precision perception of the parking environment. While current mainstream solutions generally utilize a fusion of cameras and ultrasonic radar, significant accuracy limitations persist in practical applications. Camera-based visual recognition systems primarily identify obstacles through image segmentation and object detection algorithms. However, in low light, rain, snow, or complex background interference, cameras are prone to increased image noise and motion blur, significantly increasing the algorithm's error rate. While wide-angle lenses can expand the field of view, edge distortion reduces ranging accuracy, especially at close range to the side of the vehicle. Furthermore, obstacles made of unusual materials, such as reflective surfaces and transparent glass, often lack texture features, making them difficult to effectively identify. Ultrasonic radar systems rely on the reflection time of sound waves to calculate distance, and their inherent physical properties limit recognition accuracy.

[0003] In short, these defects may lead to missed or misdetected obstacles in complex scenarios, directly affecting the safety of parking path planning and becoming a key bottleneck restricting the development of automatic parking technology. Summary of the Invention

[0004] The purpose of this application is to provide an obstacle point cloud recognition method, system, electronic device and medium to solve the problems in the prior art of lacking a classification and recognition mechanism and low obstacle recognition accuracy.

[0005] In a first aspect, the present application provides an obstacle point cloud recognition method. The obstacle point cloud recognition method includes: acquiring obstacle ultrasonic radar data, performing distance correction on the obstacle ultrasonic radar data to obtain obstacle point cloud coordinate data detected this time, wherein the obstacle point cloud coordinate data includes obstacle point cloud coordinates and detection distance; determining whether the obstacle point cloud coordinate data acquired this time meets storage conditions, and if so, storing the obstacle point cloud coordinate data; if not, discarding the obstacle point cloud coordinate data; reading obstacle point cloud coordinate data stored within a continuous time interval, matching corresponding recognition types based on detection distance change trends of multiple groups of obstacle point cloud coordinate data; and performing obstacle coordinate recognition based on the corresponding recognition type to obtain the obstacle position.

[0006] In this application, the acquired obstacle point cloud coordinate data is screened and stored if it meets the storage requirements. The corresponding identification type is determined based on the trend of the stored obstacle point cloud coordinate data over several consecutive time intervals. The obstacle position is then identified based on the corresponding identification type. This obstacle point cloud recognition method can classify and identify point clouds, eliminate physical property errors of on-board ultrasonic radar, and improve the precision and accuracy of obstacle recognition.

[0007] In an implementation of the first aspect, performing distance correction on the obstacle ultrasonic radar data to obtain the obstacle point cloud coordinate data detected this time includes: obtaining the refresh time of the vehicle-mounted ultrasonic radar, the current operating time of the system, the speed at the refresh time, and the speed of the system at the current operating time; and performing coordinate transformation on the obstacle ultrasonic radar data based on the refresh time, the current operating time of the system, the speed at the refresh time, the speed at the current operating time of the system, and the speed at the current operating time of the system, as well as the angle between the forward direction of the vehicle and the vehicle-mounted ultrasonic radar, to obtain the obstacle point cloud coordinate data, wherein the obstacle point cloud coordinates are coordinates of a coordinate system with the center of the vehicle's rear axle as the origin.

[0008] In an implementation of the first aspect, performing distance correction on the obstacle ultrasonic radar data to obtain the obstacle point cloud coordinate data detected this time also includes: correcting the obstacle point cloud coordinate data using the difference between the refresh time of the vehicle-mounted ultrasonic radar and the vehicle travel distance at the current operation time of the system.

[0009] In an implementation of the first aspect, determining whether the obstacle point cloud coordinate data obtained this time meets the storage conditions includes: determining whether the vehicle driving distance is greater than or equal to a preset driving distance, where the vehicle driving distance is the vehicle driving distance within a time interval between two consecutive storages of the obstacle point cloud coordinate data; or, determining whether the system operation time interval is greater than or equal to a preset operation time, where the system operation time interval is a time interval between two consecutive storages of the obstacle point cloud coordinate data; or, determining whether the echo intensity of the on-board ultrasonic radar this time is greater than a preset echo intensity; if so, storing the obstacle point cloud coordinate data; if not, discarding the obstacle point cloud coordinate data.

[0010] In an implementation of the first aspect, the identification type is one or more combinations of: a detection distance of continuously storing the obstacle point cloud coordinate data is equal; a detection distance of continuously storing the obstacle point cloud coordinate data is increasing and is not greater than a change threshold; a detection distance of continuously storing the obstacle point cloud coordinate data is decreasing and is not greater than a change threshold; and an internal detection distance of continuously storing the obstacle point cloud coordinate data is neither equal, increasing nor decreasing and is not greater than a change threshold; wherein the detection distance is the obstacle distance detected in the obstacle ultrasonic radar data.

[0011] In an implementation of the first aspect, performing obstacle coordinate identification according to the corresponding identification type to obtain the obstacle position includes: when the corresponding identification type is that the detection distances of the continuously stored obstacle point cloud coordinate data are equal, using the average value of the continuously stored obstacle point cloud coordinates to obtain the obstacle position; and / or when the corresponding identification type is that the detection distances of the continuously stored obstacle point cloud coordinate data are not any of equal, increasing, or decreasing detection distances and are not greater than a change threshold, using the perimeter and area of ​​a triangle formed by the ultrasonic radar and two obstacle point clouds to obtain the obstacle position.

[0012] In an implementation of the first aspect, performing obstacle coordinate identification according to the corresponding identification type to obtain the obstacle position includes: when the corresponding identification type is that the obstacle point cloud coordinate data is continuously stored and the detection distance decreases and is not greater than a change threshold, estimating the obstacle position using the N-1 and N-2 obstacle point cloud coordinate data that are continuously stored in chronological order and are parallel to the line connecting the ultrasonic radar; and / or when the corresponding identification type is that the obstacle point cloud coordinate data is continuously stored and the detection distance increases and is not greater than the change threshold, estimating the obstacle position using the N-1 and N-2 obstacle point cloud coordinate data that are continuously stored in chronological order and are parallel to the line connecting the ultrasonic radar, and obtaining the obstacle position using the N and N-1 obstacle point cloud coordinate data.

[0013] In a second aspect, the present application provides an obstacle point cloud recognition system. The obstacle point cloud recognition system includes: a data acquisition module for acquiring obstacle ultrasonic radar data, performing distance correction on the obstacle ultrasonic radar data to obtain obstacle point cloud coordinate data detected this time, wherein the obstacle point cloud coordinate data includes obstacle point cloud coordinates and detection distance; a data screening module for determining whether the obstacle point cloud coordinate data acquired this time meets the storage conditions, and if so, storing the obstacle point cloud coordinate data; if not, discarding the obstacle point cloud coordinate data; a type matching module for reading obstacle point cloud coordinate data stored in continuous time intervals, and matching corresponding recognition types based on the detection distance change trends of multiple groups of obstacle point cloud coordinate data; and a position acquisition module for performing obstacle coordinate recognition based on the corresponding recognition type to obtain the obstacle position.

[0014] In a third aspect, the present application provides an electronic device comprising: a memory for storing a computer program; and a processor for executing the computer program stored in the memory, so that the electronic device performs the obstacle point cloud recognition method as described in any one of the first aspects.

[0015] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the obstacle point cloud recognition method according to any one of the first aspects is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1A Shown is a schematic diagram of an application scenario of the obstacle point cloud recognition method described in this application.

[0017] Figure 1B Shown are structural diagrams of the client-cloud interaction scenarios in these implementations.

[0018] Figure 2 Shown is a flow chart of the obstacle point cloud recognition method described in an embodiment of the present application.

[0019] Figure 3 Shown is a schematic diagram of a radar according to an embodiment of the present application.

[0020] Figure 4 Shown is a schematic diagram of the identification type described in an embodiment of the present application.

[0021] Figure 5 Shown is a schematic diagram of an identification type described in an embodiment of the present application.

[0022] Figure 6 Shown is a structural schematic diagram of the obstacle point cloud recognition system described in an embodiment of the present application.

[0023] Figure 7 Shown is a structural schematic diagram of an electronic device described in an embodiment of the present application.

[0024] Component number description

[0025] 1 Obstacle recognition device

[0026] 11 Radar equipment

[0027] 12 Local Processor

[0028] 13 Display Terminal

[0029] 2-end-cloud interactive system

[0030] 20 Terminal

[0031] 21 Cloud Servers

[0032] 100 Obstacle Point Cloud Recognition System

[0033] 110 Data Acquisition Module

[0034] 120 Data Filtering Module

[0035] 130 Type matching module

[0036] 140 Position Acquisition Module

[0037] 700 Electronic Equipment

[0038] 710 Memory

[0039] 720 processor

[0040] 730 Display

[0041] Steps S11 to S14 DETAILED DESCRIPTION

[0042] The following describes the embodiments of the present application through specific examples. Those skilled in the art can easily understand the other advantages and effects of the present application from the content disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0043] It should be noted that in the embodiments of this application, words such as "optionally" or "for example" represent examples, illustrations, or descriptions. Any embodiment or design described in this application as "optionally" or "for example" should not be interpreted as being more preferred or advantageous than other embodiments or designs. Rather, the use of words such as "optionally" or "for example" is intended to present the relevant concepts in a concrete manner.

[0044] In the embodiments of the present application, "at least one" refers to one or more, and "more" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: the existence of A alone, the existence of A and B at the same time, and the existence of B alone, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items 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, c can be single or multiple.

[0045] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application. Therefore, the illustrations only show components related to the present application and are not drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component can be changed at will, and the component layout type may also be more complicated.

[0046] Automatic parking systems generally use cameras and ultrasonic radar sensors to fuse information to perceive the external environment and calculate information about the parking space to be parked and its surrounding environment. For parking spaces without stop lines and formed by objects in front, behind, or on the left and right (i.e., parking spaces), it is often difficult to detect them using camera sensors. Ultrasonic radar is often required to detect obstacles around the parking space to obtain parking space information. The working principle of ultrasonic radar is to emit ultrasonic waves in a fan-shaped manner through an ultrasonic transmitter and measure the distance by the time difference between the time the ultrasonic waves are received by the receiver. Due to this characteristic, automatic parking systems using ultrasonic radar cannot accurately describe the location of obstacles and cannot obtain the precise location and size of parking spaces.

[0047] To address at least the above-mentioned issues, embodiments of the present application provide an obstacle point cloud recognition method. The obstacle point cloud recognition method comprises: acquiring obstacle ultrasonic radar data, performing distance correction on the obstacle ultrasonic radar data to obtain obstacle point cloud coordinate data detected this time, wherein the obstacle point cloud coordinate data includes obstacle point cloud coordinates and detection distance; determining whether the obstacle point cloud coordinate data acquired this time meets storage conditions, and if so, storing the obstacle point cloud coordinate data; if not, discarding the obstacle point cloud coordinate data; reading obstacle point cloud coordinate data stored within consecutive time intervals, matching corresponding recognition types based on detection distance change trends of multiple groups of obstacle point cloud coordinate data; and performing obstacle coordinate recognition based on the corresponding recognition type to obtain the obstacle position.

[0048] In this embodiment, the acquired obstacle point cloud coordinate data is screened and stored if it meets the storage criteria. The corresponding identification type is determined based on the changing trend of the obstacle point cloud coordinate data over several consecutive time intervals. The obstacle position is then determined based on the corresponding identification type. This obstacle point cloud recognition method can classify and identify point clouds, eliminate errors caused by the physical characteristics of ultrasonic radar, and improve the precision and accuracy of obstacle recognition.

[0049] Figure 1A The obstacle recognition device 1 can be used to implement the obstacle point cloud recognition method provided in the embodiment of the present application, but the application scenario of the obstacle point cloud recognition method provided in the embodiment of the present application is not limited to Figure 1A The obstacle recognition device 1 shown in FIG. Figure 1A As shown, the obstacle recognition device 1 includes a radar device 11, a local processor 12, and a display terminal 13. The obstacle point cloud recognition method provided in the embodiment of the present application can be applied to the local processor 12.

[0050] in, Figure 1A The local processor 12 in the embodiment can be a local processor or a local processor cluster composed of multiple local processors or a cloud computing center, etc., which is not limited here. Figure 1A Only one radar device 11, one local processor 12 and one display terminal 13 are shown in FIG, but it should be understood that Figure 1A The examples are only used to understand this solution, and the specific numbers of local processors 12 and display terminals 13 should be flexibly determined based on actual conditions.

[0051] In other implementations, the obstacle identification device 1 may not include the display terminal 13, but may only include a local processor 12 with a display function and the radar device 11. The obstacle point cloud identification method provided in the embodiments of the present application can be applied to the local processor 12. The local processor 12 with a display function may include a tablet computer, a laptop computer, a PDA, a mobile phone, a personal computer, and a voice interaction device, or may be a monitoring device, an in-vehicle driving recorder, etc., without limitation herein.

[0052] In some other implementations, the obstacle point cloud recognition method described in this application can be applied to end-cloud interaction scenarios. Figure 1B Shown is a schematic diagram of the structure of the end-cloud interaction scenario in these implementation methods. Figure 1B 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.

[0053] Among them, the terminal 20 can be mobile or fixed. For example, the terminal 20 can be a wireless terminal or a wired terminal. The wireless terminal can refer to a device with wireless transceiver functions, which can be deployed indoors, outdoors, and in industrial workshops. The terminal 20 can be a vehicle-mounted processor, a mobile phone, a tablet computer, a laptop computer, etc., which is not limited here. 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 called a server cluster, a management platform, a data processing center, etc., which is not limited in the embodiments of this application.

[0054] The technical solutions in the embodiments of the present application will be described in detail below with reference to the accompanying drawings in the embodiments of the present application.

[0055] The following embodiments of the present application provide a method for identifying obstacle point clouds. For example, the method can be Figure 1A The local processor 12 shown or Figure 1B It is implemented by the cloud server 21 shown. Figure 2 The figure shows a flow chart of the obstacle point cloud recognition method described in the embodiment of the present application. Figure 2 As shown, the obstacle point cloud recognition method includes steps S11 to S14.

[0056] Step S11 : Obtain obstacle ultrasonic radar data, and perform distance correction on the obstacle ultrasonic radar data to obtain obstacle point cloud coordinate data detected this time, wherein the obstacle point cloud coordinate data includes obstacle point cloud coordinates and detection distance.

[0057] Step S12: determining whether the obstacle point cloud coordinate data obtained this time meets the storage condition; if so, storing the obstacle point cloud coordinate data; if not, discarding the obstacle point cloud coordinate data.

[0058] Step S13 , reading the obstacle point cloud coordinate data stored in continuous time intervals, and matching corresponding recognition types according to the detection distance change trends of the plurality of groups of obstacle point cloud coordinate data.

[0059] Step S14: performing obstacle coordinate recognition according to the corresponding recognition type to obtain the obstacle position.

[0060] In some possible implementations, obstacle point cloud recognition is initiated based on a preset soft switch or user-selected triggering condition for obstacle point cloud recognition. Ultrasonic radar data is acquired using a vehicle-mounted ultrasonic radar. Distance correction is performed on the obstacle ultrasonic radar data to obtain the obstacle point cloud coordinates of the currently detected obstacle point relative to the center of the vehicle's rear axle, as well as the detection distance. The detection distance is the distance of the obstacle directly detected by the ultrasonic radar. Specifically, the detection distance is the straight-line distance between the obstacle detected by the ultrasonic radar and the center of the vehicle's rear axle. A determination is made based on preset storage conditions as to whether the currently acquired obstacle point cloud coordinate data should be stored. If so, the obstacle point cloud coordinate data is stored; otherwise, the obstacle point cloud coordinate data is discarded. Three obstacle point cloud coordinate data stored within two consecutive time intervals are read. The corresponding recognition type is matched based on the detection distance change trend of the obstacle point cloud coordinate data. Obstacle coordinate recognition is then performed based on the corresponding recognition type to obtain the obstacle's location. It should be noted that the number of obstacle point cloud coordinate data read within the consecutive time intervals is not limited to this.

[0061] In this embodiment, the acquired obstacle point cloud coordinate data is screened and stored. Obstacle point cloud coordinate data that meets the storage requirements is then stored. The corresponding identification type is determined based on the changing trends of the stored obstacle point cloud coordinate data over several consecutive time intervals. The obstacle position is then determined based on the corresponding identification type. This obstacle point cloud recognition method can classify and identify point clouds, eliminate physical property errors of on-board ultrasonic radar, and improve the precision and accuracy of obstacle recognition.

[0062] Figure 3 Shown is a schematic diagram of the radar described in the embodiment of the present application. Figure 3As shown, using the obstacle ultrasonic radar data to perform distance correction to obtain the obstacle point cloud coordinate data detected this time includes: obtaining the refresh time of the vehicle-mounted ultrasonic radar, the current operating time of the system, the speed at the refresh time, and the speed of the system at the current operating time; obtaining the obstacle point cloud coordinate data based on the refresh time, the current operating time of the system, the speed at the refresh time, the speed at the current operating time of the system, and the angle between the vehicle forward and the vehicle-mounted ultrasonic radar, wherein the obstacle point cloud coordinates are coordinates of a coordinate system with the center of the vehicle's rear axle as the origin.

[0063] In some possible implementations, ultrasonic radar data is used to obtain the refresh time t1 of the vehicle-mounted ultrasonic radar, the current system operating time t2, the refresh speed v1, and the current system operating speed v2. The refresh time t1, the current system operating time t2, the refresh speed v1, the current system operating speed v2, and the angle between the vehicle's forward direction and the vehicle-mounted ultrasonic radar are used to obtain obstacle point cloud coordinate data, where the obstacle point cloud coordinates are coordinates of a coordinate system with the center of the vehicle's rear axle as the origin. The distance of the obstacle detected by the radar relative to the center of the vehicle's rear axle in the forward x-axis direction is obtained based on the speed-distance formula:

[0064]

[0065] Among them, X k is the forward coordinate of the obstacle point cloud coordinate, x1 is the forward distance between the radar and the center of the rear axle of the vehicle, t2 k is the current running time of the system, t1 k Refresh time, v2 k is the current running speed of the system, v1 k is the refresh speed, d k is the obstacle distance detected by the vehicle-mounted ultrasonic radar, β front It is the angle between the vehicle-mounted ultrasonic radar and the vehicle's forward direction.

[0066] According to the speed-distance formula, the distance of the obstacle detected by the radar relative to the center of the vehicle's rear axle in the longitudinal y-axis direction is:

[0067]

[0068] Among them, Y k The longitudinal coordinate of the obstacle point cloud coordinate, d k is the obstacle distance detected by the vehicle-mounted ultrasonic radar, β front It is the angle between the vehicle-mounted ultrasonic radar and the vehicle's forward direction.

[0069] In one embodiment of the present application, performing distance correction on the obstacle ultrasonic radar data to obtain the obstacle point cloud coordinate data detected this time also includes: correcting the obstacle point cloud coordinate data using the difference between the refresh time of the vehicle-mounted ultrasonic radar and the vehicle travel distance at the current operation time of the system.

[0070] In one embodiment of the present application, determining whether the obstacle point cloud coordinate data obtained this time meets the storage conditions includes: determining whether the vehicle driving distance is greater than or equal to a preset driving distance, and the vehicle driving distance is the vehicle driving distance within the time interval of two consecutive storages of the obstacle point cloud coordinate data.

[0071] It is determined whether a system operation time interval is greater than or equal to a preset operation time, where the system operation time interval is a time interval between two consecutive storages of the obstacle point cloud coordinate data.

[0072] Determine whether the echo intensity of the vehicle-mounted ultrasonic radar is greater than the preset echo intensity. If so, store the obstacle point cloud coordinate data; if not, discard the obstacle point cloud coordinate data.

[0073] In some possible implementations, when the vehicle's forward motion determines that the vehicle's driving conditions meet preset storage conditions, the system stores the point cloud data (X, Y, d) in a buffer, up to a maximum of K times. The buffer stores the forward coordinate X and longitudinal coordinate Y of each obstacle point cloud coordinate, as well as the corresponding obstacle distance d detected by the ultrasonic radar.

[0074] The preset storage condition is the vehicle travel distance S k Whether it is greater than or equal to the preset driving distance L, the vehicle driving distance is the vehicle driving distance within the time interval of two consecutive storage of obstacle point cloud coordinate data.

[0075] Or determine the system operation time interval T k Whether it is greater than or equal to the preset running time T, the system running time interval is the time interval between two consecutive storages of the obstacle point cloud coordinate data.

[0076] Alternatively, it is determined whether the echo intensity of the vehicle-mounted ultrasonic radar is greater than a preset echo intensity. If so, the obstacle point cloud coordinate data is stored in a buffer; if not, the obstacle point cloud coordinate data is discarded.

[0077] Specifically, the preset driving distance L is related to the ultrasonic radar refresh time t1, the maximum speed of the vehicle, the program operation cycle and other factors, and is, for example, 3cm-10cm. k The constraint formula is: S k =[v2k ×dt], [] is a rounding symbol, and dt is the difference between the current system runtime and the last time the obstacle point cloud coordinates were stored. The preset runtime T value depends on factors such as the ultrasonic radar refresh time t1, the vehicle's maximum operating speed, and the program execution cycle. For example, it ranges from 20ms to 100ms.

[0078] In some other possible implementations, the echo intensity F is introduced k As a storage condition for whether to store the obstacle position coordinates, the echo intensity F k This is the value automatically output by the ultrasonic radar.

[0079] Determine the echo intensity F of the vehicle-mounted ultrasonic radar k Is it greater than or equal to a preset value F? If so, store the obstacle point cloud coordinate data; if not, discard the obstacle point cloud coordinate data.

[0080] In this embodiment, the preset storage conditions for filtering obstacle point cloud coordinates can balance computing resources and accuracy, avoiding excessive computational complexity and resource consumption caused by overly densely stored data. Echo intensity is used to further filter useful data, improving computational accuracy.

[0081] Figure 4 Shown is a schematic diagram of the identification type described in the embodiment of this application. Figure 4 As shown, the identification type is one or more combinations of the following: the detection distances of the obstacle point cloud coordinate data stored continuously are equal; the detection distances of the obstacle point cloud coordinate data stored continuously are increasing and not greater than a change threshold; the detection distances of the obstacle point cloud coordinate data stored continuously are decreasing and not greater than a change threshold; and the internal detection distances of the obstacle point cloud coordinate data stored continuously are neither equal, increasing nor decreasing and are not greater than a change threshold; wherein the detection distance is the obstacle distance detected in the obstacle ultrasonic radar data.

[0082] In some possible implementations, during the vehicle driving process, for a time-based ultrasonic radar data stream {X k |k=1,2,3...}, each time we get X kThen, the detection distance change trend output at three update times is used for judgment. The obstacle point cloud coordinates stored in continuous time intervals are read to obtain the detection distances D1, D2, and D3 between the ultrasonic radar and the obstacle. The recognition types include the following four: 1. The detection distances of the obstacle point cloud coordinate data stored continuously are equal to D1=D2=D3. 2. The detection distances of the obstacle point cloud coordinate data stored continuously are increasing and are not greater than the change threshold D3-D2≤b and D2-D1≤b, 3. The detection distances of the obstacle point cloud coordinate data stored continuously are decreasing and are not greater than the change threshold D3-D2≤b and D2-D1≤b, 4. The detection distances of the obstacle point cloud coordinate data stored continuously are neither equal, increasing, nor decreasing and are not greater than the change threshold D3-D2≤b and D2-D1≤b. Among them, the detection distance is the obstacle distance detected in the obstacle ultrasonic radar data.

[0083] It should be noted that the above are four possible forms of recognition types mentioned in the embodiments of this application. The recognition type corresponding to the obstacle point cloud in a specific application scenario includes any one or more random combinations, and this application is not limited to this.

[0084] In one embodiment of the present application, the corresponding recognition type is to obtain the obstacle position by using an average value of the continuously stored obstacle point cloud coordinates when the detection distances of the continuously stored obstacle point cloud coordinate data are equal.

[0085] In some possible implementations, the corresponding recognition type is that the obstacle detected by the ultrasonic radar is considered parallel to the vehicle's direction of travel when the detection distance is equal within consecutive time intervals. The coordinates of the obstacle position are obtained by taking the average value of the coordinates of the three points. The calculation formula for the obstacle position coordinates is:

[0086]

[0087] Where x(N) and y(N) are the forward and longitudinal coordinates of the obstacle relative to the center of the vehicle's rear axle, N is the Nth point cloud, X(i) is the longitudinal coordinate of the ith point cloud, Y(i) is the forward coordinate of the ith point cloud, and Q is the number of point clouds detected at consecutive time intervals. For example, Q is preset to 3.

[0088] In other possible implementations, assume that three obstacle point cloud coordinate data are stored within consecutive time intervals. The obstacle point cloud coordinate data includes the obstacle point cloud coordinates and the detection distance. Specifically, the three obstacle point cloud coordinate data (X, Y, d) are recorded as (3000mm, 1500mm, 100mm), (3050mm, 1500mm, 100mm), and (3100mm, 1500mm, 100mm). If the corresponding recognition type is determined to be equal to the detection distance within the consecutive time interval, the calculation result of the corresponding obstacle coordinates is:

[0089]

[0090] The coordinates of the obstacle position are (3050, 1500).

[0091] Figure 5 The diagram shows a recognition type described in an embodiment of the present application. The corresponding recognition type is to continuously store the obstacle point cloud coordinate data and when the detection distance decreases and is not greater than the change threshold, estimate the obstacle point cloud data stored in chronological order in parallel with the line connecting the ultrasonic radar, and use the N-1 and N-2 obstacle point cloud coordinate data to obtain the obstacle position. Specifically, Figure 5 As shown, the corresponding recognition type is to continuously store the obstacle point cloud coordinate data with a detection distance that decreases and is not greater than the change threshold. Assume that the line connecting the first point cloud detected by the ultrasonic radar and the ultrasonic radar in chronological order is parallel to the line connecting the second point cloud detected by the ultrasonic radar, that is, the angle ΔABC≈ΔGEF. It is assumed that the vehicle is traveling in a straight line along the x-axis. The calculation formula for the obstacle position coordinates is:

[0092]

[0093] Among them, x(N) and y(N) are the forward coordinates and longitudinal coordinates of the obstacle relative to the center of the rear axle of the vehicle, N is the Nth point cloud, X(N-1) and X(N-2) are the forward coordinates of the second to last and third point clouds, Y(N-1) and Y(N-2) are the forward coordinates of the second to last and third point clouds, D(N-1) and D(N-2) are the distances between the second to last and third point clouds and the ultrasonic radar, and AG is the distance moved by the vehicle.

[0094] In some possible implementations, assume that three obstacle point cloud coordinate data are stored within consecutive time intervals. The obstacle point cloud coordinate data include obstacle point cloud coordinates and detection distances. Specifically, the three obstacle point cloud coordinate data (X, Y, d) are recorded as (3000mm, 1500mm, 1000mm), (3050mm, 1600mm, 1300mm), and (3100mm, 1700mm, 1500mm), respectively. If the corresponding recognition type is determined to be that the detection distance of the obstacle point cloud coordinate data is decreasing and is not greater than the change threshold, then the calculation result of the corresponding obstacle coordinates is:

[0095]

[0096] In one embodiment of the present application, the corresponding recognition type is to continuously store obstacle point cloud coordinate data with the detection distance increasing and not exceeding a change threshold. The obstacle position is estimated based on the parallelism between the line connecting the N-1st and N-2nd obstacle point clouds and the ultrasonic radar. The obstacle position is obtained using the Nth and N-1st obstacle point cloud coordinate data. Specifically, assuming that the line connecting the second-to-last point cloud detected by the ultrasonic radar and the line connecting the third-to-last point cloud detected by the ultrasonic radar is parallel, the obstacle position coordinate calculation formula is:

[0097]

[0098] Among them, x(N) and y(N) are the forward coordinates and longitudinal coordinates of the obstacle relative to the center of the vehicle's rear axle, N is the Nth point cloud, X(N) and X(N-1) are the forward coordinates of the first and second to last point clouds, Y(N) and Y(N-1) are the forward coordinates of the first and second to last point clouds, D(N) and D(N-1) are the distances between the first and second to last point clouds and the ultrasonic radar, and AG is the distance moved by the vehicle.

[0099] In some possible implementations, assume that three obstacle point cloud coordinate data are stored within a continuous time interval. The obstacle point cloud coordinate data includes the obstacle point cloud coordinates and the detection distance. Specifically, the three obstacle point cloud coordinate data (X, Y, d) are recorded as (3100mm, 1700mm, 1500mm), (3050mm, 1600mm, 1300mm), and (3000mm, 1500mm, 1000mm). If the corresponding recognition type is determined to be that the detection distance increases within the continuous time interval and is not greater than the change threshold, the calculation result of the corresponding obstacle coordinates is:

[0100]

[0101] In one embodiment of the present application, the corresponding identification type is a detection distance for continuously storing the obstacle point cloud coordinate data that is not equal, increasing, or decreasing. Obstacle coordinate identification based on the corresponding identification type to obtain the obstacle position includes: obtaining the obstacle position using the perimeter and area of ​​a triangle formed by the ultrasonic radar and the two obstacle point clouds. In this case, the point cloud accuracy of the non-obstacle edge is appropriately relaxed compared to the accuracy of the obstacle edge. The obstacle position coordinate calculation formula is:

[0102]

[0103] Among them, x(N) and y(N) are the forward coordinates and longitudinal coordinates of the obstacle relative to the center of the rear axle of the vehicle, N is the Nth point cloud, X(N) is the forward coordinate of the penultimate point cloud, Y(N) is the forward coordinate of the penultimate point cloud, D(N) and D(N-1) are the distances between the penultimate and second point clouds and the ultrasonic radar. Assuming that there are 3 obstacle point cloud coordinates, X1 and X2 are the forward coordinates of the third to last and second point clouds, and Y1 and Y2 are the longitudinal coordinates of the third to last and second point clouds. is the distance between the third to last point cloud and the second point cloud, s is the area of ​​the triangle formed by the coordinates of the ultrasonic radar and the two obstacle point clouds, and p is half of the perimeter of the triangle formed by the coordinates of the ultrasonic radar and the two obstacle point clouds.

[0104] Figure 6 The diagram shows the structure of the obstacle point cloud recognition system described in the embodiment of the present application. Figure 6 As shown, the obstacle point cloud recognition system 100 includes a data acquisition module 110 , a data screening module 120 , a type matching module 130 and a position acquisition module 140 .

[0105] The data acquisition module 110 is used to acquire obstacle ultrasonic radar data, and perform distance correction on the obstacle ultrasonic radar data to obtain obstacle point cloud coordinate data detected this time. The obstacle point cloud coordinate data includes obstacle point cloud coordinates and detection distance.

[0106] The data screening module 120 is used to determine whether the obstacle point cloud coordinate data obtained this time meets the storage condition. If so, the obstacle point cloud coordinate data is stored; if not, the obstacle point cloud coordinate data is discarded.

[0107] The type matching module 130 is used to read the obstacle point cloud coordinate data stored in continuous time intervals, and match corresponding recognition types according to the detection distance change trends of multiple groups of the obstacle point cloud coordinate data.

[0108] The position acquisition module 140 is configured to perform obstacle coordinate recognition according to the corresponding recognition type to obtain the obstacle position.

[0109] In some possible implementations, obstacle point cloud recognition is initiated based on a preset soft switch or user-selected triggering condition. The data acquisition module 110 is configured to acquire ultrasonic radar data using a vehicle-mounted ultrasonic radar and perform distance correction on the obstacle ultrasonic radar data to obtain the obstacle point cloud coordinates of the currently detected obstacle point relative to the center of the vehicle's rear axle, as well as the detection distance. The detection distance is the distance of the obstacle directly detected by the ultrasonic radar. Specifically, the detection distance is the straight-line distance between the obstacle detected by the ultrasonic radar and the center of the vehicle's rear axle. The data screening module 120 is configured to determine whether to store the currently acquired obstacle point cloud coordinate data based on preset storage conditions. If so, the obstacle point cloud coordinate data is stored; otherwise, the obstacle point cloud coordinate data is discarded. The type matching module 130 is configured to read three obstacle point cloud coordinate data stored within two consecutive time intervals and match the corresponding recognition type based on the detection distance change trend of the obstacle point cloud coordinate data. The position acquisition module 140 is configured to perform obstacle coordinate recognition based on the corresponding recognition type and obtain the obstacle position. It should be noted that the number of obstacle point cloud coordinate data read at consecutive time intervals is not limited thereto.

[0110] In this embodiment of the present application, the data screening module 120 filters the acquired obstacle point cloud coordinate data and stores the obstacle point cloud coordinate data that meets the storage requirements. The type matching module 130 determines the corresponding identification type based on the changing trend of the obstacle point cloud coordinate data stored over several consecutive time intervals. The position acquisition module 140 identifies the obstacle based on the corresponding identification type and obtains the obstacle position. This obstacle point cloud recognition system 100 can classify and identify point clouds, eliminate physical property errors of vehicle-mounted ultrasonic radar, and improve the precision and accuracy of obstacle recognition.

[0111] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices or methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of modules / units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules or units, which can be electrical, mechanical or other forms.

[0112] The modules / units described as separate components may or may not be physically separate, and the components displayed as modules / units may or may not be physical modules, that is, they may be located in one place or distributed across multiple network elements. Some or all of the modules / units may be selected according to actual needs to achieve the purpose of the embodiments of the present application. For example, the functional modules / units in the various embodiments of the present application may be integrated into a processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into a single module / unit.

[0113] Those skilled in the art should further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the composition and steps of each example according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0114] An embodiment of the present application also provides an electronic device. Figure 7 Shown is a schematic diagram of the structure of the electronic device 700 according to an embodiment of the present application. Figure 7 As shown, in this embodiment, the electronic device 700 includes a memory 710 and a processor 720.

[0115] The memory 710 is used to store computer programs; preferably, the memory 710 includes: ROM, RAM, magnetic disk, USB flash drive, memory card or optical disk, etc., various media that can store program codes.

[0116] Specifically, the memory 710 may include a computer system readable medium in the form of a volatile memory, such as a random access memory (RAM) and / or a cache memory. The electronic device 700 may further include other removable / non-removable, volatile / non-volatile computer system storage media. The memory 710 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present application. It will be understood that the memory 710 may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM) or a programmable read-only memory (PROM), which is used 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). Memory described in the embodiments of the present invention is intended to include, but not be limited to, these and any other suitable types of memory.

[0117] The processor 720 is connected to the memory 710 and is used to execute the computer program stored in the memory 710 so that the electronic device 700 executes the obstacle point cloud recognition method described in any embodiment of the present application.

[0118] Optionally, the processor 720 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can 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, and discrete hardware components.

[0119] Optionally, the electronic device 700 in this embodiment may further include a display 730. The display 730 is communicatively connected to the memory 710 and the processor 720, and is used to display a graphical user interface (GUI) interactive interface related to the obstacle point cloud recognition method described in the embodiment of the present application.

[0120] The present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the obstacle point cloud recognition method described in any embodiment of the present application is implemented.

[0121] As used in this specification, the terms "component," "module," "system," and the like are used to represent computer-related entities, hardware, firmware, a combination 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. By way of illustration, both an application running on a computing device and a computing device can be a component. One or more components can reside in a process and / or an execution thread, and a component can be located on a computer and / or distributed between two or more computers. In addition, these components can be executed from various computer-readable media having various data structures stored thereon. 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 on a local system, a distributed system, and / or a network, such as the Internet interacting with other systems via signals).

[0122] The descriptions of the processes or structures corresponding to the above figures have different emphases. For parts that are not described in detail in a certain process or structure, please refer to the relevant descriptions of other processes or structures.

[0123] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical concepts disclosed in this application shall be covered by the claims of this application.

Claims

1. A method for identifying obstacle point clouds, characterized in that: include: Obtaining obstacle ultrasonic radar data, and performing distance correction on the obstacle ultrasonic radar data to obtain obstacle point cloud coordinate data detected this time, wherein the obstacle point cloud coordinate data includes obstacle point cloud coordinates and detection distance; Determine whether the obstacle point cloud coordinate data obtained this time meets the storage conditions, if so, store the obstacle point cloud coordinate data, if not, discard the obstacle point cloud coordinate data; Reading obstacle point cloud coordinate data stored in continuous time intervals, and matching corresponding recognition types according to detection distance change trends of multiple sets of obstacle point cloud coordinate data; Obstacle coordinates are identified according to the corresponding identification type to obtain the obstacle position.

2. The obstacle point cloud recognition method according to claim 1, characterized in that: Performing distance correction on the obstacle ultrasonic radar data to obtain the obstacle point cloud coordinate data detected this time includes: Get the refresh time of the vehicle-mounted ultrasonic radar, the current operating time of the system, the refresh speed and the current operating speed of the system; Based on the refresh time, the current operating time of the system, the speed at the refresh time and the speed at the current operating time of the system, and the angle between the forward direction of the vehicle and the on-board ultrasonic radar, the obstacle ultrasonic radar data is coordinate transformed to obtain the obstacle point cloud coordinate data, wherein the obstacle point cloud coordinates are coordinates of a coordinate system with the center of the rear axle of the vehicle as the origin.

3. The obstacle point cloud recognition method according to claim 2, characterized in that: Performing distance correction on the obstacle ultrasonic radar data to obtain the obstacle point cloud coordinate data detected this time also includes: correcting the obstacle point cloud coordinate data using the difference between the refresh time of the vehicle-mounted ultrasonic radar and the vehicle travel distance at the current operation time of the system.

4. The obstacle point cloud recognition method according to claim 1, characterized in that: Determining whether the obstacle point cloud coordinate data obtained this time meets the storage conditions includes: Determine whether the vehicle's travel distance is greater than or equal to a preset travel distance, where the vehicle's travel distance is the vehicle's travel distance within a time interval between two consecutive storages of the obstacle point cloud coordinate data, or, Determine whether the system operation time interval is greater than or equal to the preset operation time, the system operation time interval being the time interval between two consecutive storages of the obstacle point cloud coordinate data, or, Determine whether the echo intensity of the vehicle-mounted ultrasonic radar is greater than the preset echo intensity. If so, store the obstacle point cloud coordinate data; if not, discard the obstacle point cloud coordinate data.

5. The obstacle point cloud recognition method according to claim 1, characterized in that: The identification type is one or more combinations of: the detection distances of the obstacle point cloud coordinate data stored continuously are equal; the detection distances of the obstacle point cloud coordinate data stored continuously are increasing and not greater than a change threshold; the detection distances of the obstacle point cloud coordinate data stored continuously are decreasing and not greater than a change threshold; and the internal detection distances of the obstacle point cloud coordinate data stored continuously are neither equal, increasing nor decreasing and are not greater than a change threshold; wherein the detection distance is the obstacle distance detected in the obstacle ultrasonic radar data.

6. The obstacle point cloud recognition method according to claim 5, characterized in that: Performing obstacle coordinate identification according to the corresponding identification type to obtain the obstacle position includes: The corresponding recognition type is to obtain the obstacle position by using the average value of the continuously stored obstacle point cloud coordinates when the detection distances of the continuously stored obstacle point cloud coordinate data are equal; and / or The corresponding recognition type is to continuously store the obstacle point cloud coordinate data. When the detection distance does not belong to any of the detection distances of equal, increasing or decreasing and is not greater than the change threshold, the obstacle position is obtained by using the perimeter and area of ​​the triangle formed by the ultrasonic radar and the two obstacle point clouds.

7. The obstacle point cloud recognition method according to any one of claims 1 or 6, characterized in that: Performing obstacle coordinate identification according to the corresponding identification type to obtain the obstacle position includes: The corresponding recognition type is that when the detection distance of the obstacle point cloud coordinate data is continuously stored and decreases and is not greater than the change threshold, the obstacle point cloud coordinate data of the Nth and N-1th obstacle point clouds stored continuously in chronological order are parallel to the line connecting the ultrasonic radar to obtain the obstacle position using the N-1th and N-2th obstacle point cloud coordinate data; and / or The corresponding recognition type is that when the obstacle point cloud coordinate data is continuously stored and the detection distance increases and is not greater than the change threshold, the N-1th and N-2th obstacle point clouds stored continuously in chronological order are estimated in parallel with the line connecting the ultrasonic radar, and the obstacle position is obtained using the Nth and N-1th obstacle point cloud coordinate data.

8. An obstacle point cloud recognition system, characterized in that: include: a data acquisition module, configured to acquire obstacle ultrasonic radar data, and perform distance correction on the obstacle ultrasonic radar data to obtain obstacle point cloud coordinate data detected this time, wherein the obstacle point cloud coordinate data includes obstacle point cloud coordinates and detection distance; a data screening module, configured to determine whether the obstacle point cloud coordinate data obtained this time meets the storage conditions, and if so, store the obstacle point cloud coordinate data; if not, discard the obstacle point cloud coordinate data; A type matching module is used to read the obstacle point cloud coordinate data stored in continuous time intervals and match the corresponding recognition type according to the detection distance change trend of multiple sets of obstacle point cloud coordinate data; The position acquisition module is used to identify the coordinates of the obstacle according to the corresponding recognition type to obtain the obstacle position.

9. An electronic device, characterized in that: The electronic device comprises: Memory for storing computer programs; A processor, wherein the processor is configured to execute the computer program stored in the memory so as to enable the electronic device to perform the obstacle point cloud recognition method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the obstacle point cloud recognition method according to any one of claims 1 to 7 is implemented.

Images