Motor vehicle lateral obstacle recognition method and device and computer readable storage medium

Abstract

The embodiment of the invention provides a motor vehicle lateral obstacle identification method and device and a computer readable storage medium. The method comprises the following steps that a detection instruction is sent to an ultrasonic radar to control the ultrasonic radar to send waves once in a fixed-frequency mode and a sweep-frequency mode in each detection period to detect obstacles outside the corresponding side of a vehicle body; analyzing each detection return wave of the ultrasonic radar in each detection period, and calculating to obtain the following detection information: actual fixed-frequency echo height, actual sweep-frequency echo height and real-time point cloud coordinates of an obstacle; determining the shape type of the obstacle based on the actual fixed-frequency echo height and the actual sweep-frequency echo height in the detection period; and based on the shape type of the obstacle, selecting a matched point cloud clustering method to cluster the real-time point cloud coordinates of the obstacle in the plurality of detection periods so as to calculate and obtain the clustering point cloud coordinates and the actual outer contour shape of the obstacle. According to the embodiment, the position and the contour shape of the obstacle can be determined by using one ultrasonic radar.

Description

Technical Field

[0001] Embodiments of the present invention relate to the technical field of motor vehicle obstacle recognition, and more particularly to a motor vehicle lateral obstacle recognition method, device, and computer-readable storage medium. Background Art

[0002] At present, with the increasing development of fully automatic parking technology, the demand for parking in narrow spaces and complex scenes with multiple obstacles is increasing. Therefore, the accuracy of using on-board ultrasonic radar to perceive the position of objects in the environment is becoming increasingly higher. Among them, on the basis of traditional ultrasonic radar that can only provide real-time detection distance, ultrasonic radar is required to construct the outline shape of obstacles around the vehicle body and determine the height attributes of obstacles relative to the vehicle.

[0003] There are two main methods for motor vehicle obstacle recognition based on ultrasonic radar. The first method is to use the cross echo between at least two ultrasonic radars on the motor vehicle, adopt the least squares principle to locate the obstacle, and finally determine the position profile of the obstacle object; however, since the motor vehicle body side usually only has one ultrasonic radar installed at the front and back, and the distance between the two ultrasonic radars is large, it is usually impossible to use cross echoes for positioning, and the cost of using multiple ultrasonic radars is also high, and the calculation is difficult; the second method is to use the two reflected echoes of the obstacle within a detection cycle of the ultrasonic radar, the first reflected echo is the shortest distance position reflection of the obstacle, and the second reflected echo is reflected back to the sensor through the ground, and the height of the obstacle is measured by using the two reflected echoes. However, this method is only applicable to smooth roads in static scenes. The calculation accuracy of this method is poor on rough asphalt roads or when the vehicle speed is fast. Summary of the invention

[0004] The technical problem to be solved by the embodiments of the present invention is to provide a method for identifying lateral obstacles of a motor vehicle, which can determine the position and contour shape of the obstacle using only one ultrasonic radar.

[0005] A further technical problem to be solved by an embodiment of the present invention is to provide a device for identifying lateral obstacles of a motor vehicle, which can determine the position and contour shape of an obstacle using only one ultrasonic radar.

[0006] A further technical problem to be solved by an embodiment of the present invention is to provide a computer-readable storage medium for storing a computer program that can use an ultrasonic radar to determine the position and contour shape of an obstacle.

[0007] In order to solve the above technical problems, the embodiment of the present invention provides the following technical solution: a method for identifying a lateral obstacle of a motor vehicle, comprising the following steps: Sending a detection instruction to an ultrasonic radar installed on a corresponding side of the vehicle body to control the ultrasonic radar to emit waves once in a fixed frequency mode and once in a sweep frequency mode in each detection cycle to detect obstacles outside the corresponding side of the vehicle body; Analyze the detection return waves of the ultrasonic radar corresponding to the fixed frequency mode and the swept frequency mode in each detection cycle to calculate and obtain the following detection information: the actual fixed frequency echo height in the fixed frequency mode, the actual swept frequency echo height in the swept frequency mode, and the real-time point cloud coordinates of the obstacle; Determining a shape type of the obstacle based on the actual fixed-frequency echo height and the actual swept-frequency echo height in each detection cycle, the shape type being a point type or a strip type; and A matching point cloud clustering method is selected based on the shape type of the obstacle to cluster the real-time point cloud coordinates of the obstacle in multiple detection cycles to calculate the clustered point cloud coordinates and the actual outer contour shape of the obstacle.

[0008] Furthermore, the real-time point cloud coordinates of the obstacle are obtained by analyzing the detection return waves of the ultrasonic radar corresponding to the fixed frequency mode and the sweeping frequency mode within one detection cycle to calculate the real-time detection distance of the ultrasonic radar relative to the obstacle when it emits wave detection in the fixed frequency mode and the sweeping frequency mode, respectively; obtaining the actual moving distance of the motor vehicle within the time difference between the wave detection in the fixed frequency mode and the wave detection in the sweeping frequency mode within one detection cycle; and combining the real-time detection distance and the actual moving distance to calculate the real-time point cloud coordinates of the obstacle based on the plane geometric relationship.

[0009] Further, the determining the shape type of the obstacle based on the actual fixed-frequency echo height and the actual swept-frequency echo height in each detection cycle specifically includes: Finding out the standard fixed-frequency echo height corresponding to the real-time detection distance in the fixed-frequency mode from a standard fixed-frequency echo height data table reflecting the corresponding relationship between the detection distance and the fixed-frequency echo height, and finding out the standard swept-frequency echo height corresponding to the real-time detection distance in the swept-frequency mode from a standard swept-frequency echo height data table reflecting the corresponding relationship between the detection distance and the swept-frequency echo height; and The standard fixed-frequency echo height is compared with the actual fixed-frequency echo height, and the standard swept-frequency echo height is compared with the actual swept-frequency echo height. If the standard fixed-frequency echo height is smaller than the actual fixed-frequency echo height and the standard swept-frequency echo height is smaller than the actual swept-frequency echo height, it is determined that the obstacle is a point type; otherwise, it is determined that the obstacle is a strip type.

[0010] Further, the selecting a matching point cloud clustering algorithm model based on the shape type of the obstacle to cluster the real-time point cloud coordinates of the obstacle to obtain the clustered point cloud coordinates and the actual outer contour of the obstacle specifically refers to: If the shape type of the obstacle is a point type, a mean shift clustering method is selected to cluster the real-time point cloud coordinates of the obstacle; If the shape type of the obstacle is a strip type, a density clustering method is selected to cluster the real-time point cloud coordinates of the obstacle.

[0011] Furthermore, the method further comprises: Acquire the real-time position of the motor vehicle each time the ultrasonic radar transmits a wave for detection in each detection cycle, and sequentially match and cache the following available information corresponding to each time the ultrasonic radar transmits a wave for detection: the real-time position, the actual fixed-frequency echo height, and the actual swept-frequency echo height; When the available information within at least two detection cycles is cached, the actual fixed-frequency echo height of the cached fixed-frequency mode is sequentially extracted according to the wave transmission order of the ultrasonic radar to calculate the height ratio of the actual fixed-frequency echo height corresponding to the previous sweeping frequency mode to obtain the echo height ratio coefficient; constructing a first variation curve graph of echo height varying with the real-time position based on the actual fixed-frequency echo height and the actual swept-frequency echo height, and constructing a second variation curve graph of echo height ratio coefficient varying with the real-time position based on the echo height ratio coefficient; and The first variation curve graph and the second variation curve graph are analyzed to determine the height attribute of the obstacle relative to the motor vehicle.

[0012] Furthermore, the frequency variation directions of the wave transmission in the frequency sweeping modes of two adjacent detection periods are opposite.

[0013] Furthermore, the number of waving pulses in the fixed frequency mode is less than the number of waving pulses in the sweep frequency mode.

[0014] On the other hand, in order to solve the above-mentioned further technical problems, an embodiment of the present invention provides the following technical solutions: a motor vehicle lateral obstacle recognition device is connected to an ultrasonic radar installed on one side of a motor vehicle body, the device includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the motor vehicle obstacle recognition method as described in any one of the above items.

[0015] On the other hand, in order to solve the above-mentioned further technical problems, an embodiment of the present invention provides the following technical solutions: a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the motor vehicle lateral obstacle recognition method as described in any one of the above-mentioned items.

[0016] After adopting the above technical scheme, the embodiment of the present invention has at least the following beneficial effects: the embodiment of the present invention controls an ultrasonic radar installed on the corresponding side of the motor vehicle body to emit waves in a fixed frequency mode and a sweep frequency mode in each detection cycle to detect obstacles outside the corresponding side of the vehicle body, effectively utilizing the characteristic of the ultrasonic radar having a larger field of view when emitting waves in the sweep frequency mode to detect obstacles, with a wide detection range and improved obstacle detection accuracy, and further by analyzing each detection return wave of the ultrasonic radar corresponding to the fixed frequency mode and the sweep frequency mode in the detection cycle, respectively, thereby calculating and obtaining the detection accuracy of the fixed frequency mode. The actual fixed-frequency echo height, the actual swept-frequency echo height in the swept-frequency mode, and the real-time point cloud coordinates of the obstacle are obtained by utilizing the characteristic that the echo heights of point-shaped and strip-shaped obstacles are significantly different when reflecting the detection return wave. The shape type of the obstacle can be preliminarily determined according to the actual fixed-frequency echo height and the actual swept-frequency echo height. Then, the real-time point cloud coordinates of the obstacle in multiple detection cycles are clustered according to the shape type of the obstacle to accurately obtain the clustered point cloud coordinates and the actual outer contour shape of the obstacle. The detection accuracy is high, and the obstacle detection is achieved with only one ultrasonic radar, and the cost is low. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The present invention is a flowchart of the steps of an optional embodiment of the method for identifying a lateral obstacle of a motor vehicle.

[0018] Figure 2 This is a schematic diagram of a method for identifying lateral obstacles of a motor vehicle according to an optional embodiment of the present invention for calculating the real-time point cloud coordinates of an obstacle.

[0019] Figure 3 This is a schematic diagram of calculating the real-time point cloud coordinates of an obstacle according to another optional embodiment of the method for identifying a lateral obstacle of a motor vehicle of the present invention.

[0020] Figure 4 This is a schematic diagram of real-time point cloud coordinate clustering of obstacles in an optional embodiment of the method for identifying lateral obstacles of a motor vehicle of the present invention.

[0021] Figure 5 This is a schematic diagram of real-time point cloud coordinate clustering of obstacles in another optional embodiment of the method for identifying lateral obstacles of a motor vehicle of the present invention.

[0022] Figure 6 This is a first variation curve diagram of a type of obstacle in an optional embodiment of the method for identifying lateral obstacles of a motor vehicle of the present invention.

[0023] Figure 7 This is a first variation curve diagram of another type of obstacle in an optional embodiment of the method for identifying lateral obstacles of a motor vehicle of the present invention.

[0024] Figure 8 This is a first variation curve diagram of another type of obstacle in an optional embodiment of the method for identifying lateral obstacles of a motor vehicle of the present invention.

[0025] Fig. 9 This is a second variation curve diagram of an optional embodiment of the method for identifying lateral obstacles of a motor vehicle of the present invention.

[0026] Fig.10 The figure is a principle block diagram of an optional embodiment of the motor vehicle lateral obstacle recognition device of the present invention.

[0027] Fig.11 This is a functional module diagram of an optional embodiment of the motor vehicle lateral obstacle recognition device of the present invention. DETAILED DESCRIPTION

[0028] The present application is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the following illustrative embodiments and descriptions are only used to explain the present invention and are not intended to limit the present invention. In addition, the embodiments and features in the embodiments of the present application may be combined with each other without conflict.

[0029] like Figure 1 As shown, an optional embodiment of the present invention provides a method for identifying a lateral obstacle of a motor vehicle, comprising the following steps: S1: Sending a detection instruction to an ultrasonic radar 1 installed on a corresponding side of a motor vehicle body to control the ultrasonic radar to emit waves once in a fixed frequency mode and once in a sweep frequency mode in each detection cycle to detect obstacles outside the corresponding side of the vehicle body; S2: Analyze the detection return waves of the ultrasonic radar 1 corresponding to the fixed frequency mode and the swept frequency mode in each detection cycle to obtain the following detection information: the actual fixed frequency echo height in the fixed frequency mode, the actual swept frequency echo height in the swept frequency mode, and the real-time point cloud coordinates of the obstacle; S3: determining the shape type of the obstacle based on the actual fixed-frequency echo height and the actual swept-frequency echo height in each detection cycle, wherein the shape type is a point type or a strip type; and S4: Based on the shape type of the obstacle, a matching point cloud clustering method is selected to cluster the real-time point cloud coordinates of the obstacle within multiple detection cycles to calculate the clustered point cloud coordinates and the actual outer contour shape of the obstacle.

[0030] In the embodiment of the present invention, an ultrasonic radar 1 installed on the corresponding side of the vehicle body is controlled to emit waves in a fixed frequency mode and a sweep frequency mode in each detection cycle to detect obstacles outside the corresponding side of the vehicle body. The ultrasonic radar 1 is effectively used to detect obstacles with a larger field of view when emitting waves in the sweep frequency mode, and the detection range is wide, thereby improving the detection accuracy of obstacles. Further, by analyzing the detection return waves of the ultrasonic radar 1 corresponding to the fixed frequency mode and the sweep frequency mode in the detection cycle, the actual fixed frequency echo height in the fixed frequency mode, the actual sweep frequency echo height in the sweep frequency mode, and the real-time point cloud coordinates of the obstacle are calculated and obtained respectively. By using the characteristic that the echo heights of point-shaped and strip-shaped obstacles are significantly different when reflecting the detection return wave, the shape type of the obstacle can be preliminarily determined according to the actual fixed frequency echo height and the actual sweep frequency echo height. Then, the real-time point cloud coordinates of the obstacle in multiple detection cycles are clustered according to the shape type of the obstacle, and the clustered point cloud coordinates and the actual outer contour shape of the obstacle can be accurately obtained. The detection accuracy is high, and the obstacle detection is achieved with only one ultrasonic radar 1, and the cost is low.

[0031] In a specific implementation, there is no strict restriction on the order of the fixed frequency mode and the sweep frequency mode within a detection cycle, but the order of the fixed frequency mode and the sweep frequency mode within each detection cycle should remain consistent.

[0032] In an optional embodiment of the present invention, the real-time point cloud coordinates of the obstacle are obtained by: analyzing the detection return waves of the ultrasonic radar 1 corresponding to the fixed frequency mode and the sweep frequency mode in one detection cycle to calculate the real-time detection distance of the ultrasonic radar 1 relative to the obstacle when it is transmitting wave detection in the fixed frequency mode and the sweep frequency mode respectively, obtaining the actual moving distance of the motor vehicle within the time difference between the wave detection in the fixed frequency mode and the wave detection in the sweep frequency mode in one detection cycle, and combining the real-time detection distance and the actual moving distance to calculate the real-time point cloud coordinates of the obstacle based on the plane geometric relationship. In this embodiment, after calculating the real-time detection distance of the ultrasonic radar 1 relative to the obstacle when it is transmitting wave detection in each wave transmission mode according to the detection return waves, combined with the actual moving distance of the motor vehicle in the wave transmission interval on both sides, the real-time point cloud coordinates of the obstacle can be determined by using a triangulation algorithm, and the calculation is very convenient.

[0033] exist Figure 2 The obstacle 2 is a dot. Figure 3 The obstacle 2 is in the shape of a strip. Similarly, within a detection cycle, the ultrasonic radar 1 on the motor vehicle emits waves at time T1 and time T2 respectively. D1 and D2 are the real-time detection distances at time T1 and time T2 respectively. D0 is the actual moving distance of the motor vehicle between time T1 and time T2. The real-time point cloud coordinates can be calculated by combining the triangulation algorithm.

[0034] In an optional embodiment of the present invention, determining the shape type of the obstacle based on the actual fixed-frequency echo height and the actual swept-frequency echo height in each detection cycle specifically includes: Finding out the standard fixed-frequency echo height corresponding to the real-time detection distance in the fixed-frequency mode from a standard fixed-frequency echo height data table reflecting the corresponding relationship between the detection distance and the fixed-frequency echo height, and finding out the standard swept-frequency echo height corresponding to the real-time detection distance in the swept-frequency mode from a standard swept-frequency echo height data table reflecting the corresponding relationship between the detection distance and the swept-frequency echo height; and The standard fixed-frequency echo height is compared with the actual fixed-frequency echo height, and the standard swept-frequency echo height is compared with the actual swept-frequency echo height. If the standard fixed-frequency echo height is smaller than the actual fixed-frequency echo height and the standard swept-frequency echo height is smaller than the actual swept-frequency echo height, it is determined that the obstacle is a point type; otherwise, it is determined that the obstacle is a strip type.

[0035] In this embodiment, by referring to the standard fixed-frequency echo height data table and the standard swept-frequency echo height data table reflecting the corresponding relationship between the detection distance and the echo height, the standard fixed-frequency echo height corresponding to the real-time detection distance in the fixed-frequency mode and the standard swept-frequency echo height corresponding to the real-time detection distance in the swept-frequency mode are found respectively. Since the actual fixed-frequency echo height and the actual swept-frequency echo height of a point obstacle are usually smaller, based on this, after comparing the standard fixed-frequency echo height with the actual fixed-frequency echo height and comparing the standard swept-frequency echo height with the actual swept-frequency echo height, if the standard fixed-frequency echo height is less than the actual fixed-frequency echo height and the standard swept-frequency echo height is less than the actual swept-frequency echo height, it is determined that the obstacle is a point type; otherwise, the corresponding obstacle is a strip type, thereby achieving a preliminary determination of the shape of the obstacle.

[0036] In an optional embodiment of the present invention, the selecting a matching point cloud clustering algorithm model based on the shape type of the obstacle to cluster the real-time point cloud coordinates of the obstacle to obtain the clustered point cloud coordinates and the actual outer contour of the obstacle specifically refers to: If the shape type of the obstacle is a point type, a mean shift clustering method is selected to cluster the real-time point cloud coordinates of the obstacle; If the shape type of the obstacle is a strip type, a density clustering method is selected to cluster the real-time point cloud coordinates of the obstacle.

[0037] In this embodiment, point-type obstacles use the mean shift clustering method to realize real-time point cloud coordinate clustering, while strip-type obstacles use the density clustering method to realize real-time point cloud coordinate clustering. The corresponding clustering method realizes effective clustering of obstacles of corresponding shapes, and realizes accurate calculation of the clustered point cloud coordinates and actual outer contour of the obstacle.

[0038] like Figure 4 In the figure, Pn on the left side is the real-time point cloud coordinate. The mean shift clustering method is used for clustering this type of point cloud as shown on the right side of the figure, where P1 is the noise point, P2 is the clustered point cloud coordinate, and the dotted outline in the figure is the actual outer contour of the obstacle.

[0039] like Figure 5 In the figure, Pn on the left side is the real-time point cloud coordinate. The density clustering method is used for clustering this type of point cloud as shown on the right side of the figure, where P1 is a noise point. The dotted outline L1 in the figure is the actual outer contour of the obstacle. The clustered point cloud coordinates of this method are the real-time point cloud coordinates along the dotted outline L1.

[0040] In an optional embodiment of the present invention, the method further includes: S4: obtaining the real-time position of the motor vehicle each time the ultrasonic radar performs wave detection in each detection cycle, and sequentially matching and caching the following available information corresponding to each time the ultrasonic radar performs wave detection: the real-time position, the actual fixed-frequency echo height, and the actual swept-frequency echo height; S5: when the available information within at least two detection cycles is cached, the actual fixed-frequency echo height of the cached fixed-frequency mode is sequentially extracted according to the wave transmission order of the ultrasonic radar, and the height ratio of the actual fixed-frequency echo height corresponding to the previous sweeping frequency mode is calculated to obtain the echo height ratio coefficient; S6: constructing a first variation curve diagram of the echo height varying with the real-time position based on the actual fixed-frequency echo height and the actual swept-frequency echo height, and constructing a second variation curve diagram of the echo height ratio coefficient varying with the real-time position based on the echo height ratio coefficient; and S7: Analyze the first change curve graph and the second change curve graph to determine the height attribute of the obstacle relative to the motor vehicle.

[0041] In this embodiment, after calculating and obtaining the clustered point cloud coordinates and the actual outer contour of the obstacle, the real-time position (i.e., world coordinates) of the motor vehicle during each wave detection can also be calculated through the data provided by the odometer or IMU module of the motor vehicle, and the real-time position of each detection cycle, the actual fixed-frequency echo height, and the actual swept-frequency echo height are cached, and then after the above-mentioned available information reaches a predetermined amount, the height ratio is calculated by calculating the actual fixed-frequency echo height of the fixed-frequency mode and the actual fixed-frequency echo height corresponding to the previous swept-frequency mode, and then the echo height ratio coefficients β1, β2...βn are obtained in sequence, and then, the real-time position of the motor vehicle (CarPos, unit: cm) is used as the independent variable, and the actual fixed-frequency echo height and the actual swept-frequency echo height are used as dependent variables respectively, and the echo height (Echo) is constructed in the same coordinate system. A first variation curve graph of the echo height ratio coefficients β1, β2, ..., βn as the real-time position changes, and a second variation curve graph of the echo height ratio coefficients β1, β2, ..., βn as the real-time position changes is constructed in another coordinate system with the real-time position (CarPos, in cm) as the independent variable and the echo height ratio coefficients β1, β2, ..., βn as the dependent variable; like Figure 6 The figure shows a first variation curve diagram constructed when the ultrasonic radar 1 detects obstacles such as pillars or vehicle fronts, wherein C1 is a variation curve of the actual swept frequency echo height, and C2 is a variation curve of the actual fixed frequency echo height. As can be seen from the figure, both curves are in a U-shaped shape with the opening facing downwards; like Figure 7 The figure shows a first variation curve diagram constructed when the ultrasonic radar 1 detects obstacles such as curbs that are lower than the height of a motor vehicle, wherein C3 is a variation curve of the actual swept frequency echo height, and C4 is a variation curve of the actual fixed frequency echo height. As can be seen from the figure, both curves are in the shape of wavy lines; like Figure 8 The figure shows the first variation curve diagram constructed when the ultrasonic radar 1 detects obstacles such as large walls and long objects, among which C5 is the variation curve of the actual swept frequency echo height, and C6 is the variation curve of the actual fixed frequency echo height. It can be seen from the figure that after the fixed frequency echo height value reaches the maximum value, it no longer increases, and as the vehicle runs, the curve basically presents a straight line shape; while the swept frequency echo height curve presents a wavy line shape as the vehicle runs, and the swept frequency echo height is relatively high, exceeding 60,000.

[0042] like Fig. 9 The figure shows a second variation curve diagram constructed by ultrasonic radar 1 after detecting different types of obstacles, wherein C21 corresponds to obstacles such as pillars or vehicle fronts, which present a U-shaped shape with the opening facing downward; C22 corresponds to obstacles such as low curbs, which presents mainly a straight line shape; C23 corresponds to obstacles such as high walls, which presents mainly a wavy line shape.

[0043] Based on Figure 6-Figure 9 The step of determining the height attribute of the obstacle relative to the motor vehicle based on the first change curve diagram and the second change curve diagram is as follows: Analyze the first change curve graph. If both curves in the first change curve graph are U-shaped with the opening facing downward, it is accurately determined that the obstacle is higher than the motor vehicle. If the curve of the actual sweep frequency echo height changing with the real-time position of the motor vehicle in the first change curve graph is a wavy line, it is temporarily impossible to determine the height of the object. If the curve of the actual fixed frequency echo height changing with the real-time position of the motor vehicle in the first change curve graph is a straight line, it is preliminarily determined that the obstacle is higher than the motor vehicle. If it is a wavy line, it is preliminarily determined that the obstacle is lower than the motor vehicle. In order to avoid misjudgment in complex scenarios, the second change curve is further analyzed. If the obstacle that was initially determined to be higher than the motor vehicle in the previous step is a wavy line in the second change curve, it is accurately determined that the obstacle is higher than the motor vehicle. If the obstacle that was initially determined to be higher than the motor vehicle in the previous step is a straight line in the second change curve, it is accurately determined that the obstacle is lower than the motor vehicle.

[0044] In addition, it is understandable that the height of an obstacle relative to a motor vehicle is usually determined by the lowest point of the motor vehicle bumper from the ground. For example, if the lowest point of the motor vehicle bumper from the ground is 25 cm, then obstacles below 25 cm are defined as low; obstacles above 25 cm are defined as high. Usually, the critical value is set between 15 and 25 cm. In actual environments, wheel stoppers and 10 cm curbs are identified as low, and others above the critical value are identified as high.

[0045] In an optional embodiment of the present invention, the frequency of the wave transmission in the sweep mode of two adjacent detection cycles is changed in opposite directions. In this embodiment, the frequency of the wave transmission in the sweep mode of two adjacent detection cycles is changed in opposite directions, that is, the sweep mode of two adjacent detection cycles is an upper sweep mode and a lower sweep mode, and the two sweep modes are alternated with each other to improve the accuracy of obstacle detection.

[0046] In an optional embodiment of the present invention, the number of waving pulses in the fixed frequency mode is less than the number of waving pulses in the sweep frequency mode. Specifically, the number of waving pulses in the fixed frequency mode is set to 16, and its waving frequency is 48kHZ; while the number of waving pulses in the sweep frequency mode is 96, and the waving frequency of the upper sweep frequency mode increases from 44kHZ to 52kHZ in sequence, and the waving frequency of the lower sweep frequency mode decreases from 52kHZ to 44kHZ in sequence.

[0047] On the other hand, Fig.10As shown, an embodiment of the present invention provides a motor vehicle lateral obstacle recognition device 3, which is connected to an ultrasonic radar 1 installed on one side of a motor vehicle body. The device 3 includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, a motor vehicle lateral obstacle recognition method as described in any one of the above items is implemented.

[0048] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory 32 and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the motor vehicle side obstacle recognition device 3. For example, the computer program may be divided into Fig.11 The functional modules in the motor vehicle lateral obstacle recognition device 3, wherein the radar control module 41, the return wave analysis module 42, the obstacle shape type determination module 43 and the point cloud clustering module 44 respectively execute the above steps S1 to S4.

[0049] The motor vehicle lateral obstacle recognition device 3 may be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The motor vehicle lateral obstacle recognition device 3 may include, but is not limited to, a processor 30 and a memory 32. Those skilled in the art may understand that the schematic diagram is only an example of the motor vehicle lateral obstacle recognition device 3 and does not constitute a limitation on the motor vehicle lateral obstacle recognition device 3. It may include more or less components than shown in the figure, or combine certain components, or different components. For example, the motor vehicle lateral obstacle recognition device 3 may also include input and output devices, network access devices, buses, etc.

[0050] The processor 30 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor 30 is the control center of the motor vehicle lateral obstacle recognition device 3, and uses various interfaces and lines to connect various parts of the entire motor vehicle lateral obstacle recognition device 3.

[0051] The memory 32 can be used to store the computer program and / or module. The processor 30 realizes various functions of the motor vehicle lateral obstacle recognition device 3 by running or executing the computer program and / or module stored in the memory 32 and calling the data stored in the memory 32. The memory 32 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a graphic recognition function, a graphic overlay function, etc.), etc.; the data storage area can store data (such as graphic data, etc.) created according to the use of the control device, etc. In addition, the memory 32 can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0052] If the functions described in the embodiments of the present invention are implemented in the form of software function modules or units and sold or used as independent products, they can be stored in a storage medium readable by a computing device. Based on this understanding, the embodiments of the present invention implement all or part of the processes in the above-mentioned embodiments, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor 30, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0053] On the other hand, an embodiment of the present invention provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the motor vehicle lateral obstacle recognition method as described in the above embodiment.

[0054] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0055] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, which all belong to the protection scope of the present invention.

Claims

1. A method for identifying lateral obstacles of a motor vehicle, characterized in that: The method comprises the following steps: Sending a detection instruction to an ultrasonic radar installed on a corresponding side of the vehicle body to control the ultrasonic radar to emit waves once in a fixed frequency mode and once in a sweep frequency mode in each detection cycle to detect obstacles outside the corresponding side of the vehicle body; Analyze the detection return waves of the ultrasonic radar corresponding to the fixed frequency mode and the swept frequency mode in each detection cycle to calculate and obtain the following detection information: the actual fixed frequency echo height in the fixed frequency mode, the actual swept frequency echo height in the swept frequency mode, and the real-time point cloud coordinates of the obstacle; Determining a shape type of the obstacle based on the actual fixed-frequency echo height and the actual swept-frequency echo height in each detection cycle, wherein the shape type is a point type or a strip type; A matching point cloud clustering method is selected based on the shape type of the obstacle to cluster the real-time point cloud coordinates of the obstacle in multiple detection cycles to calculate the clustered point cloud coordinates and the actual outer contour shape of the obstacle.

2. The method for identifying lateral obstacles of a motor vehicle according to claim 1, characterized in that: The real-time point cloud coordinates of the obstacle are obtained in the following manner: analyzing the detection return waves of the ultrasonic radar corresponding to the fixed-frequency mode and the sweeping-frequency mode in one detection cycle to calculate the real-time detection distance of the ultrasonic radar relative to the obstacle when it emits waves for detection in the fixed-frequency mode and the sweeping-frequency mode, respectively; obtaining the actual moving distance of the motor vehicle within the time difference between the wave detection in the fixed-frequency mode and the wave detection in the sweeping-frequency mode in one detection cycle; and combining the real-time detection distance and the actual moving distance to calculate the real-time point cloud coordinates of the obstacle based on the plane geometric relationship.

3. The method for identifying a lateral obstacle of a motor vehicle according to claim 1, wherein: The determining the shape type of the obstacle based on the actual fixed-frequency echo height and the actual swept-frequency echo height in each detection cycle specifically includes: Finding out the standard fixed-frequency echo height corresponding to the real-time detection distance in the fixed-frequency mode from a standard fixed-frequency echo height data table reflecting the corresponding relationship between the detection distance and the fixed-frequency echo height, and finding out the standard swept-frequency echo height corresponding to the real-time detection distance in the swept-frequency mode from a standard swept-frequency echo height data table reflecting the corresponding relationship between the detection distance and the swept-frequency echo height; and The standard fixed-frequency echo height is compared with the actual fixed-frequency echo height, and the standard swept-frequency echo height is compared with the actual swept-frequency echo height. If the standard fixed-frequency echo height is smaller than the actual fixed-frequency echo height and the standard swept-frequency echo height is smaller than the actual swept-frequency echo height, it is determined that the obstacle is a point type; otherwise, it is determined that the obstacle is a strip type.

4. The method for identifying lateral obstacles of a motor vehicle according to claim 1, characterized in that: The selecting a matching point cloud clustering algorithm model based on the shape type of the obstacle to cluster the real-time point cloud coordinates of the obstacle to obtain the clustered point cloud coordinates and the actual outer contour of the obstacle specifically refers to: If the shape type of the obstacle is a point type, a mean shift clustering method is selected to cluster the real-time point cloud coordinates of the obstacle; If the shape type of the obstacle is a strip type, a density clustering method is selected to cluster the real-time point cloud coordinates of the obstacle.

5. The method for identifying lateral obstacles of a motor vehicle according to claim 1, characterized in that: The method further comprises: Acquire the real-time position of the motor vehicle each time the ultrasonic radar transmits a wave for detection in each detection cycle, and sequentially match and cache the following available information corresponding to each time the ultrasonic radar transmits a wave for detection: the real-time position, the actual fixed-frequency echo height, and the actual swept-frequency echo height; When the available information within at least two detection cycles is cached, the actual fixed-frequency echo height of the cached fixed-frequency mode is sequentially extracted according to the wave transmission order of the ultrasonic radar to calculate the height ratio of the actual fixed-frequency echo height corresponding to the previous sweeping frequency mode to obtain the echo height ratio coefficient; constructing a first variation curve graph of echo height varying with the real-time position based on the actual fixed-frequency echo height and the actual swept-frequency echo height, and constructing a second variation curve graph of echo height ratio coefficient varying with the real-time position based on the echo height ratio coefficient; and The first variation curve graph and the second variation curve graph are analyzed to determine the height attribute of the obstacle relative to the motor vehicle.

6. The method for identifying lateral obstacles of a motor vehicle according to claim 1, characterized in that: The frequency variation directions of the wave transmission in the sweep frequency mode of two adjacent detection cycles are opposite.

7. The method for identifying a lateral obstacle of a motor vehicle according to any one of claims 1 to 6, characterized in that: The number of waving pulses in the fixed frequency mode is less than the number of waving pulses in the sweep frequency mode.

8. A motor vehicle lateral obstacle recognition device connected to an ultrasonic radar installed on one side of the motor vehicle body, characterized in that: The device includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and the processor implements the motor vehicle obstacle recognition method according to any one of claims 1 to 7 when executing the computer program.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the motor vehicle lateral obstacle recognition method according to any one of claims 1 to 7.

Images