A method for determining a drivable area based on millimeter-wave radar and related equipment

Abstract

The present invention discloses a method for determining a drivable area based on millimeter-wave radar and related equipment. The method includes: obtaining obstacle point cloud information based on millimeter-wave radar scanning data, wherein the millimeter-wave scanning data includes periodic scanning data, uniform scanning data, and angle scanning data; performing aggregation processing based on the obstacle point cloud information to obtain aggregated point cloud information; determining obstacle associated point information based on the aggregated point cloud information and the kernel density estimation method; extracting target road information based on the obstacle associated point information; and determining a drivable area based on the target road information. The method provided in the embodiment of the present application is aimed at millimeter-wave radars that can realize partial scanning and tracking functions, and realizes tracking and scanning of environmental scenes in low-speed specific environmental scenarios, and realizes demarcation of drivable areas with the assistance of millimeter-wave radars in specific scenarios, so that subsequent vehicles can be planned to the drivable area for vehicle control.

Description

Technical Field

[0001] This specification relates to the field of vehicle intelligent control, and more specifically, to a method for determining a drivable area based on millimeter-wave radar and related equipment. Background Art

[0002] Current intelligent driving assistance systems for vehicles are typically designed using millimeter-wave radar and environmental perception cameras. However, their application scenarios are primarily roads with clear lane lines and recognizable moving vehicles, enabling autonomous driving under limited conditions. Current technical solutions are unable to meet the requirements for vehicle recognition of stationary obstacles or environmental obstacles, are inadequate for off-road scenarios or scenarios with obstacles on roads, and are unable to assist drivers in environmental assessment.

[0003] However, existing driving technology judgment capabilities cannot effectively address the use of autonomous driving functions on unstructured or non-standardized roads. Even with the assistance of ultrasonic radar, timely detection of surrounding obstacles cannot be guaranteed. Existing millimeter-wave radar technology has already achieved some scanning and tracking functions, primarily for tracking and continuous detection of targets. Summary of the Invention

[0004] The Summary of the Invention introduces a series of simplified concepts that will be further described in the Detailed Description of the Invention. The Summary of the Invention is not intended to limit the key features and essential features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0005] In a first aspect, the present invention proposes a method for determining a drivable area based on millimeter-wave radar, the method comprising:

[0006] Obtaining obstacle point cloud information based on millimeter-wave radar scanning data, wherein the millimeter-wave scanning data includes periodic scanning data, uniform speed scanning data, and angle scanning data;

[0007] Perform aggregation processing based on the above obstacle point cloud information to obtain aggregated point cloud information;

[0008] Determine obstacle-related point information based on the above-mentioned aggregated point cloud information and kernel density estimation method;

[0009] Extracting target road information based on the above obstacle associated point information;

[0010] The drivable area is determined based on the target road information.

[0011] Optionally, performing aggregation processing on the obstacle point cloud information to obtain aggregated point cloud information includes:

[0012] Performing coordinate transformation on the obstacle point cloud information according to the layered method to obtain transformed point cloud information;

[0013] Multiple aggregation processes are performed based on the converted point cloud information to obtain the above-mentioned aggregated point cloud information.

[0014] Optionally, determining obstacle-related point information based on the above-mentioned aggregated point cloud information and kernel density estimation method includes:

[0015] Extract peak area point sets from the above aggregated point cloud information based on kernel density estimation method;

[0016] Obstacle associated point information is determined based on the above peak area point set.

[0017] Optionally, extracting target road information based on the above obstacle-related point information includes:

[0018] Obtaining road boundary information based on the least squares method according to the above obstacle association information;

[0019] Obtaining road surface information based on the above obstacle association information using a kernel density estimation method;

[0020] The target road information is determined based on the road boundary information and the road surface information.

[0021] Optionally, the above method further includes:

[0022] The target road information is optimized based on the obstacle association information, the preset density threshold and the kernel density estimation method.

[0023] Optionally, determining the drivable area based on the above road information includes:

[0024] Obtain the maximum height and depth information of the obstacle based on the elevation value;

[0025] The drivable area is determined based on the maximum height and depth information and the road information.

[0026] Optionally, the above method further includes:

[0027] Acquire the above periodic scanning data based on a preset period;

[0028] Acquire the above-mentioned uniform speed scanning data based on a preset speed;

[0029] The angle scanning data is obtained based on a preset angle.

[0030] In a second aspect, the present invention further proposes a device for determining a drivable area based on millimeter-wave radar, comprising:

[0031] A first acquisition unit is configured to acquire obstacle point cloud information based on millimeter wave radar scanning data, wherein the millimeter wave scanning data includes periodic scanning data, uniform speed scanning data, and angle scanning data;

[0032] A second acquiring unit is configured to perform aggregation processing based on the obstacle point cloud information to acquire aggregated point cloud information;

[0033] A first determining unit is configured to determine obstacle-related point information based on the aggregated point cloud information;

[0034] An extraction unit, configured to extract road contour information based on the obstacle-related point information;

[0035] The second determining unit is configured to determine a drivable area based on the road profile information.

[0036] In a third aspect, an electronic device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is configured to implement the steps of the method for determining a drivable area based on millimeter-wave radar as described in any one of the first aspects above when executing the computer program stored in the memory.

[0037] In a fourth aspect, the present invention further proposes a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it implements the method for determining a drivable area based on millimeter-wave radar according to any one of the above items in the first aspect.

[0038] In summary, the method for determining a drivable area based on millimeter-wave radar in an embodiment of the present application includes: obtaining obstacle point cloud information based on millimeter-wave radar scanning data, wherein the millimeter-wave scanning data includes periodic scanning data, uniform scanning data, and angle scanning data; performing aggregation processing based on the obstacle point cloud information to obtain aggregated point cloud information; determining obstacle-related point information based on the aggregated point cloud information and the kernel density estimation method; extracting target road information based on the obstacle-related point information; and determining a drivable area based on the target road information. The method for determining a drivable area based on millimeter-wave radar provided in an embodiment of the present application is for a millimeter-wave radar that can realize partial scanning and tracking functions, and realizes tracking and scanning of environmental scenes in low-speed specific environmental scenarios, and realizes demarcation of the drivable area with the assistance of millimeter-wave radar in specific scenarios, so that subsequent vehicles can be planned to the drivable area for vehicle control.

[0039] The millimeter-wave radar-based drivable area determination method of the present invention, and other advantages, objectives, and features of the present invention will be partially reflected in the following description, and will also be understood by those skilled in the art through research and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present description. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0041] Figure 1 A flowchart of a method for determining a drivable area based on millimeter-wave radar provided in an embodiment of the present application;

[0042] Figure 2 A schematic diagram of the architecture of an intelligent driving system based on millimeter-wave radar provided in an embodiment of the present application;

[0043] Figure 3 A schematic diagram of the software structure for determining a drivable area based on millimeter-wave radar provided in an embodiment of the present application;

[0044] Figure 4 A schematic diagram of the structure of a device for determining a drivable area based on millimeter-wave radar provided in an embodiment of the present application;

[0045] Figure 5 A schematic diagram of the structure of an electronic device for determining a drivable area based on millimeter-wave radar provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] The method for determining the drivable area based on millimeter-wave radar provided in the embodiment of the present application is for the millimeter-wave radar that can realize partial scanning and tracking functions, so as to realize tracking and scanning of environmental scenes in low-speed specific environmental scenes, and realize the demarcation of the drivable area with the assistance of millimeter-wave radar in specific scenes, so that subsequent vehicles can be planned to the drivable area for vehicle control.

[0047] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the embodiments described are only some embodiments of the present application, not all embodiments.

[0048] To resolve the above issues, please refer to Figure 1 , which is a flow chart of a method for determining a drivable area based on millimeter-wave radar provided in an embodiment of the present application, which may specifically include:

[0049] S110. Obtaining obstacle point cloud information based on millimeter-wave radar scanning data, wherein the millimeter-wave scanning data includes periodic scanning data, uniform-speed scanning data, and angle scanning data;

[0050] For example, if the target vehicle meets the driving conditions for this function, such as a speed less than a preset speed value, such as 15 km / h, the vehicle's millimeter-wave radar is activated. The millimeter-wave radar collects data from periodic scans, uniform speed scans, and angle scans simultaneously. Preliminary fusion and extraction of these three types of data yields obstacle point cloud information.

[0051] S120, performing aggregation processing based on the obstacle point cloud information to obtain aggregated point cloud information;

[0052] For example, the obstacle point cloud is aggregated according to different density parameters to obtain aggregated point cloud information. Aggregation using multiple aggregation density parameters can fully represent the outline characteristics of the obstacle, and the obtained aggregated point cloud information contains more comprehensive obstacle information.

[0053] S130, determining obstacle-related point information based on the aggregated point cloud information and kernel density estimation method;

[0054] Exemplarily, the point set in the peak area of ​​the aggregated point cloud data is extracted based on the kernel density estimation method, the standard deviation of the horizontal and vertical coordinates of the point set is calculated, and the obstacle-associated point information is determined based on the standard deviation. The obtained obstacle-associated point information is the key point of the obstacle.

[0055] S140, extracting target road information based on the obstacle-related point information;

[0056] For example, target road information can be extracted based on obstacle-related point information. For example, if a road boundary is linearly distributed, the least squares method can be used to fit the road boundary. This method can also identify roadside steps as obstacles, as they are roughly distributed in a straight line. Using the least squares method, the area where the straight line is fitted represents the road boundary. After determining the road boundary, information within the road is extracted to obtain target road information.

[0057] S150: Determine a drivable area based on the target road information.

[0058] For example, based on the extracted target road information and combined with the elevation information, the bumps and depressions on the road are determined, that is, which locations exceed the threshold of normal road undulations, thereby determining which locations on the target road have potholes or bumps, and thus the area where the vehicle can be driven can be determined.

[0059] In summary, the method proposed in the embodiment of the present application clusters radar data so that the smart car can obtain information about the road ahead and obstacles. According to different density parameters, the algorithm is called multiple times to complete multi-density clustering. Combined with information extraction methods such as road edges, road surfaces, and obstacles, the drivable area information of the smart car is accurately extracted. For millimeter-wave radars that can achieve partial scanning and tracking functions, it is possible to track and scan the environmental scene in a low-speed specific environmental scenario, and to achieve the demarcation of the drivable area with the assistance of millimeter radar in a specific scenario. Subsequent vehicles can be planned to the drivable area for vehicle control.

[0060] In some examples, performing aggregation processing based on the obstacle point cloud information to obtain aggregated point cloud information includes:

[0061] Performing coordinate transformation on the obstacle point cloud information according to the layered method to obtain transformed point cloud information;

[0062] Multiple aggregation processes are performed based on the converted point cloud information to obtain the above-mentioned aggregated point cloud information.

[0063] For example, the origin of the vehicle body coordinate system can be specified as the center of the rear axle of the vehicle (the direction of travel is Axis, horizontal to the left Axle, above the vehicle body Axis) to establish a temporary coordinate system, stipulating that the projection center of the millimeter wave radar on the ground is the coordinate origin O, and the horizontal forward is Axis, horizontal to the left Axis, perpendicular to the xoy plane upward Axis. The millimeter wave radar scanning coordinate system takes the center of the millimeter wave radar as the coordinate origin, extracts the first layer as the xoy plane, and the horizontal forward is Axis, horizontal to the left Axis, perpendicular to the vertical upward To ensure that two scanning lines can scan the road surface when the ground is level and to accurately extract road information, the radar is installed with a downward tilt of α>2° (calibrable). Continue to convert the sensor coordinates to the vehicle coordinate system. The vertical scanning accuracy of the millimeter wave radar is a fixed value. =0.8° (calibrable), this application can use the layered method shown in the following formula for coordinate transformation:

[0064]

[0065] The polar coordinate representation of the information point is ( ), when the millimeter wave radar scans to the origin, that is, θ=0°, its The value is expressed as Participate in calculation, indicating , θ=0° The distance from the radar coordinate origin to the ground when the scanning angle θ=0° for the first scanning line of the smart car on the horizontal ground; the vertical scanning accuracy is a fixed value °

[0066] in:

[0067]

[0068] i represents the scanning layer number: by adjusting The value can be fixed according to the vertical scanning accuracy The change of several multiples of ° distinguishes the scanning area and orientation, and finally achieves the purpose of scanning layering. is the distance from the radar origin to the scanning point; is the angle value scanned by the millimeter radar; i represents the scanning layer number; , θ=0° The distance from the radar coordinate origin to the ground when the scanning angle θ=0° of the first scanning line of the smart car on the horizontal ground; is the rotation matrix; is the translation matrix, which represents the translation from the origin of the temporary coordinate system to the origin of the vehicle coordinate system.

[0069] The millimeter wave radar receives the returned millimeter waves and defines the targets that are not aggregated or identified in the formed data set as data points p (the set of returned data points), that is, the targets that are not aggregated into targets. Then the data set p is verified and the number of data points in the non-noise area is judged. If the number of data points is greater than the target threshold, p is defined as the core object and the target object is identified by the target point p. Indicates that at the same time Include high-density identifiable points within the threshold domain.

[0070] examine All unprocessed data points in , which must contain at least a number of data points that can be identified, then the relevant data points are classified into cluster C (by testing the number of data points in the non-noise area of ​​the data set p, and if the number is greater than the target threshold data Repeat the steps until no new objects are added to cluster C, and then proceed to the next step.

[0071] Repeat the above steps to connect the data points C to form the maximum density area, set the data point density compliance value (the test value is 10 / 10,000 pixels), and then merge the density compliance area set. Finally, output the data set D with the environment connected area (that is, the set of points that meet the conditions and density compliance). The points that meet the conditions are , which can be said to be dense point cloud information.

[0072] In summary, the method proposed in the embodiment of the present application can obtain multi-layer point cloud data by coordinate transformation of the point cloud information of the obstacle, and aggregate the multi-layer point cloud data to obtain more comprehensive obstacle information and improve the accuracy of obstacle and lane recognition.

[0073] In some examples, determining obstacle-associated point information based on the aggregated point cloud information and kernel density estimation method includes:

[0074] Extract peak area point sets from the above aggregated point cloud information based on kernel density estimation method;

[0075] Obstacle associated point information is determined based on the above peak area point set.

[0076] For example, we aggregate the raw millimeter-wave radar echo data set (the data set that hasn't been aggregated by the algorithm) and record it using core parameters (such as those in the following formula). The digitized parameters can then display their distribution characteristics. The radar data points are then converted into independent random variables, and the function f(x) of the random variables can be expressed as:

[0077]

[0078] in, Where x is the kernel density estimate of the density function, is the kernel function, h is the window width, and n is the number of estimated data. , expressed as:

[0079]

[0080] Then, a suitable window is selected for kernel density estimation. The window width selected by the present invention is:

[0081]

[0082] in is the standard deviation of the data to be estimated.

[0083] Combined with the current frames of data to be estimated The calculation is performed using the intermediate value of and n. n is adjusted to prevent sudden changes. Specifically, if extremely irregular fluctuations occur, that is, the window width is too small and n needs to be increased. If f(x) is too stable, the recognition is low, that is, the window width is too large and n needs to be reduced.

[0084] Density object merging process: connect the objects into cluster C with the maximum density, and form a set of such objects, and then merge the set of objects that meet the density standard. Extract the distance between adjacent density points, and set the typical target distance as the n-dimensional vector , the distance expression formula is as follows:

[0085]

[0086] In the formula, when =1, it is the Euclidean distance; when ≠1, it is weighted Euclidean distance; when n=2, is the distance between two points in the plane; when n=3, is the distance between two points in space. In order to simplify and engineer the calculation, Restricted to a fixed range, i.e. =1 is weighted Euclidean distance; is the distance between two points in a plane (two-dimensional distance); and is the distance between two points in space (3D distance). After performing kernel density estimation on the radar data, extract the point set in each peak area, calculate the standard deviation of its horizontal and vertical coordinates, and use The distance between two points in the plane (two-dimensional distance) is used as the λ value, which is the standard deviation of the horizontal and vertical coordinates of the peak area point set estimated based on the kernel density of the current frame. If σ≠0, then = , = 1; otherwise, σ = ε (very small). Then, the x and y values ​​of the points scanned on the roadside are assigned different weights, and the points on the obstacle are scanned again.

[0087] In summary, the method provided in the embodiment of the present application extracts the point set within the peak area based on the kernel density estimation method, and determines the obstacle association information based on the motors in the peak area to identify the obstacles with higher recognition accuracy.

[0088] In some examples, extracting target road information based on the obstacle-associated point information includes:

[0089] Obtaining road boundary information based on the least squares method according to the above obstacle association information;

[0090] Obtaining road surface information based on the above obstacle association information using a kernel density estimation method;

[0091] The target road information is determined based on the road boundary information and the road surface information.

[0092] Exemplary, least squares method for fitting road boundaries: Based on experience, the points scanned on the road edge are generally linearly distributed. The present invention fits the data points generated by clustering and uses the least squares method to optimize the surface contour of the scanned road edge. The straight line Y = kx + b is fitted, and the k and b are extracted as follows:

[0093]

[0094] The present invention performs one-dimensional kernel density analysis on one frame of millimeter-wave radar raw data, estimates the characteristic distribution of millimeter-wave radar data, obtains the data density value and 4A value (the distance of the four nearest neighbor points) around the peak point, and sorts the aggregated data 4A values. (from low to high); when clustering, cluster in order from large to small density, and use it as an input parameter for multi-density clustering. Different obstacle data will have different density inputs and specific extraction methods. Different extraction schemes are preset according to different obstacles in the early stage. After extraction according to different density sizes using specific extraction methods, verify whether the extraction result is a certain type of obstacle.

[0095] The peak neighborhood scanning and step-by-step processing method is used to reduce the search complexity of the algorithm. The present invention will estimate the area where the kernel density estimate value is higher than a certain threshold (which can be calibrated). As a preparation area, Represents the neighborhood range of kernel density estimation. The results of four consecutive millimeter-to-meter radar scans are then further subdivided into two groups. The lowest and second-lowest layers, as they scan directly onto the road surface, are primarily used to extract lateral road information (such as boundaries). The highest and second-highest layers, due to their longer scanning range, are primarily used to extract longitudinal road information (such as obstacles) ahead of the road surface (partial four-layer data can be used when necessary). This combined processing of the two layers significantly reduces false alarms and false positives associated with single-layer scanning, while also enabling better detection of a sufficiently large driving area ahead of the vehicle and reducing algorithm complexity.

[0096] In summary, the method provided in the embodiment of the present application can quickly identify the boundaries of the road and determine the corresponding range of the road based on the kernel density method and the least squares method.

[0097] In some examples, the method further includes:

[0098] The target road information is optimized based on the obstacle association information, the preset density threshold and the kernel density estimation method.

[0099] For example, we first perform a one-dimensional kernel density estimation analysis on the horizontal coordinate y of the first layer of radar data (data when i=1), find the peak point of the density estimation, which serves as the core object of the algorithm, and then extract the density data point area near the peak point that meets the following conditions:

[0100]

[0101] Where, is the density estimate of the data points near the peak point; is the density threshold.

[0102] This method maximizes the retention of scanned curb points while filtering out scattered data points, such as scanned flowers and plants. The extracted target area data points are sorted from high to low density to cluster high-density clusters. The remaining data are then clustered sequentially to group data points of varying density into distinct categories. Before clustering the target area, the 4A nearest neighbor distances of the peak point are calculated using weighted Euclidean distance as input parameters. The adaptively generated parameters are then fed into the improved D algorithm to cluster the linearly distributed curb data points into clusters. The least squares method is then used to fit several roadside edges. Similarly, the next lower layer data is analyzed to obtain several roadside edges, retaining one each on the left and right sides closest to the vehicle. The resulting z-values ​​are used to verify the height characteristics of the extracted curb points (typically around 20 cm). Comparisons with actual scene images and experiments confirm that this method is highly effective in extracting roadside edges.

[0103] In some examples, determining a drivable area based on the road information includes:

[0104] Obtain the maximum height and depth information of the obstacle based on the elevation value;

[0105] The drivable area is determined based on the maximum height and depth information and the road information.

[0106] For example, based on the extracted roadside information, the drivable area can be defined horizontally to provide a basis for lateral control of the vehicle. At the same time, the longitudinal driving area of ​​the vehicle needs to be further detected within this range. The process can be divided into two parts: detection and extraction of road surface and extraction of concave and convex obstacle information on the road surface. In order to reduce the computational complexity, the road surface is scanned and the vehicle is detected.

[0107] Based on the data characteristics of closely arranged points, the pre-processed radar 3D data is projected on the 2D planes xoy and yoz. In the yoz plane, combined with the elevation information, the road surface points scanned by each radar scanning layer can be extracted. Within a certain range, the road surface has a certain continuity, so its height should also vary within a certain range. Assume that the current vehicle position height is The local height difference caused by the uneven ground is , then the height of the part of the road scanned by the laser radar should meet [ Since the distances between points on the road surface are not much different and they are continuously and densely distributed, in the xoy plane, the remaining road surface points that meet the following conditions can be further marked as road surface points:

[0108]

[0109] The above formula represents the set of points that can meet the pass condition, where i and j are the serial numbers of adjacent points in the same scanning layer; is the proportionality coefficient; is the horizontal resolution of each radar layer.

[0110] After filtering out the scanning point data outside the road and on the road surface, the remaining data points on the road are clustered and mainly divided into two situations: concave obstacles, which have negative height values, such as potholes, etc., which pose certain safety hazards to the driving of smart cars; convex obstacles, which have positive height values, such as pedestrians, vehicles, shoulders or walls at intersections, etc. When these two types of obstacles appear within the road range, their information needs to be extracted. Because the radar scanning method is from left to right, only the surfaces of objects with the same angle and close to the radar are scanned during the scanning process. Therefore, the Y values ​​of the data of the highest two layers (locally 4 layers when necessary) of millimeter-wave radars on the road are used with an improved algorithm to cluster the obstacles on the road with different densities. The maximum height (depth) information of convex and concave obstacles is extracted using the elevation value on the yoz plane. The category to which the obstacle belongs can be determined by the obstacle height (depth) information, so that the obstacle attribute information representation within the road area can be obtained.

[0111] object={distance, center_x, center_y, object_width, angle}.

[0112]

[0113]

[0114] Where i represents the left and right curbs; It represents the width safety factor for vehicle passage, and finally outputs the set of passable areas.

[0115] In summary, the method provided in the embodiment of the present application provides a more reasonable route for the vehicle by obtaining the high-depth information of the obstacle and comparing it with the high-depth that the vehicle can pass through.

[0116] In some examples, the method further includes:

[0117] Acquire the above periodic scanning data based on a preset period;

[0118] Acquire the above-mentioned uniform speed scanning data based on a preset speed;

[0119] The angle scanning data is obtained based on a preset angle.

[0120] For example, when the millimeter-wave radar performs periodic scanning at a preset period, it returns three-dimensional data. The millimeter-wave radar is set to automatically scan, with scanning points arranged sequentially from left to right. The scanning method is periodic scanning, and data points are returned in each frame. The scanning data returned by the millimeter-wave radar changes with the driving environment and is clustered based on the surrounding environment of the smart car. During actual clustering, the clustering algorithm can adapt the number of clusters based on data characteristics. When the millimeter-wave radar performs a uniform scanning at a preset speed, the points scanned by the central controller millimeter-wave radar on the road surface are scanned longitudinally at a uniform and constant speed, resulting in a basically uniform and dense distribution along the y-axis. If there are points on the roadside (generally in the direction of vehicle travel) in the longitudinal direction, they are adjacent points in the same layer, with a sparse distribution along the x-axis and a particularly dense distribution along the y-axis, forming a partially linear shape. When the millimeter-wave radar scans at a preset angle, the central controller controls the millimeter-wave radar to scan at the same scanning angle. When the distance to the obstacle increases during vehicle driving, the central controller monitors that the distance between adjacent data points becomes larger, showing a positive correlation. In the case of equal distance, as the angle between the scanned point on the obstacle and the y-axis increases, it increases, and it is monitored to show a positive correlation.

[0121] For some examples, see Figure 2 Schematic diagram of the intelligent driving system architecture based on millimeter-wave radar. Schematic diagram of the intelligent driving system architecture based on millimeter-wave radar, including a millimeter-wave radar system, a central controller (with a drivable area detection technology based on millimeter-wave radar), a CAN bus, an ultrasonic radar system, and other parts. Under normal circumstances, the millimeter-wave radar system emits millimeter waves at a fixed angle or set FOV to identify moving pedestrians or vehicles. In the present invention, the millimeter-wave radar system provides the central controller with scanned raw images and data. The ultrasonic radar is used to collect ultrasonic reflection information from the real scene environment, accurately determine the target reflection flash echo, and form a subsequent supplement to the original input of the visual camera image. The vehicle CAN bus is used to provide the central controller with point function switch signals and real-time vehicle speed signals, direction angle signals and other signals required by the system. The central controller is used to implement the storage and calculation of the functional algorithm of this application, process the millimeter-wave feedback information, and store the algorithm programs of other functions. The rear display system can provide quasi-image information for the driver to view.

[0122] For some examples, see Figure 3, which is a schematic diagram of the software structure for determining a drivable area based on millimeter-wave radar provided in an embodiment of the present application. The central controller includes a millimeter-wave reflected wave processing module, whose function is to filter and process the original millimeter-wave signal. The modules and algorithms corresponding to the method of the present application include a scanning control module, a coordinate conversion module, a target aggregation module, an echo analysis and wave density processing module, a road information extraction and optimization module, a drivable area delineation module, and the drivable area is displayed in a display module.

[0123] See also Figure 4 In one embodiment of the present application, a device for determining a drivable area based on a millimeter-wave radar may include:

[0124] A first acquiring unit 21 is configured to acquire obstacle point cloud information based on millimeter wave radar scanning data, wherein the millimeter wave scanning data includes periodic scanning data, uniform speed scanning data, and angle scanning data;

[0125] A second acquiring unit 22 is configured to perform aggregation processing based on the obstacle point cloud information to acquire aggregated point cloud information;

[0126] A first determining unit 23 is configured to determine obstacle-related point information based on the aggregated point cloud information;

[0127] An extraction unit 24 is used to extract road contour information based on the above obstacle associated point information;

[0128] The second determining unit 25 is configured to determine a drivable area based on the road profile information.

[0129] like Figure 5 As shown, an embodiment of the present application also provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 320 and executable on the processor. When the processor 320 executes the computer program 311, the steps of any of the above-mentioned methods for determining a drivable area based on millimeter-wave radar are implemented.

[0130] Since the electronic device introduced in this embodiment is a device used to implement a millimeter-wave radar-based drivable area determination device in an embodiment of the present application, based on the method introduced in the embodiment of the present application, technical personnel in this field can understand the specific implementation of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of the present application is no longer introduced in detail here. As long as the equipment used by technical personnel in this field to implement the method in the embodiment of the present application falls within the scope of protection to be protected by this application.

[0131] In the specific implementation process, the computer program 311 can be implemented when executed by the processor Figure 1Any implementation manner in the corresponding embodiments.

[0132] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0133] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0134] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0135] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0136] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0137] The present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device is caused to execute the following Figure 1 The method shown.

[0138] A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, they fully or partially produce the processes or functions according to the embodiments of the present application. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be stored by a computer, or a data storage device such as a server or data center that integrates one or more available media. Available media can include magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

[0139] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0140] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components 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 an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0141] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0142] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0143] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.

[0144] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for determining a drivable area based on millimeter-wave radar, characterized in that: include: Obtaining obstacle point cloud information based on millimeter-wave radar scanning data, wherein the millimeter-wave radar scanning data includes periodic scanning data, uniform speed scanning data, and angle scanning data; Performing aggregation processing according to the obstacle point cloud information to obtain aggregated point cloud information; Determining obstacle-related point information based on the aggregated point cloud information and kernel density estimation method; Extracting target road information based on the obstacle associated point information; determining a drivable area based on the target road information; Extracting target road information according to the obstacle associated point information includes: Acquiring road boundary information based on the least squares method according to the obstacle association information; Acquiring road surface information based on the obstacle association information using a kernel density estimation method; The target road information is determined according to the road boundary information and the road surface information.

2. The method according to claim 1, wherein The performing aggregation processing according to the obstacle point cloud information to obtain aggregated point cloud information includes: Performing coordinate transformation on the obstacle point cloud information according to a layered method to obtain transformed point cloud information; Multiple aggregation processes are performed based on the converted point cloud information to obtain the aggregated point cloud information.

3. The method according to claim 1, wherein Determining obstacle-related point information based on the aggregated point cloud information and the kernel density estimation method includes: Extracting peak area point sets from the aggregated point cloud information based on a kernel density estimation method; Obstacle-related point information is determined based on the peak area point set.

4. The method according to claim 1, wherein The method further comprises: The target road information is optimized according to the obstacle association information, a preset density threshold and a kernel density estimation method.

5. The method according to claim 1, wherein Determining a drivable area based on the road information includes: Obtain the maximum height and depth information of the obstacle based on the elevation value; The drivable area is determined based on the maximum height and depth information and the road information.

6. The method of claim 1 , further comprising: Acquiring the periodic scanning data based on a preset period; Acquiring the uniform speed scanning data based on a preset speed; The angle scanning data is acquired based on a preset angle.

7. A device for determining a drivable area based on millimeter-wave radar, characterized in that: The method for determining a drivable area based on a millimeter-wave radar according to any one of claims 1 to 6, wherein the device for determining a drivable area based on a millimeter-wave radar comprises: A first acquiring unit is configured to acquire obstacle point cloud information based on millimeter-wave radar scanning data, wherein the millimeter-wave radar scanning data includes periodic scanning data, uniform-speed scanning data, and angle scanning data; a second acquiring unit, configured to perform aggregation processing on the obstacle point cloud information to acquire aggregated point cloud information; a first determining unit, configured to determine obstacle-related point information based on the aggregated point cloud information; an extraction unit, configured to extract road contour information based on the obstacle associated point information; The second determining unit is configured to determine a drivable area based on the road profile information.

8. An electronic device comprising: A memory and a processor, characterized in that the processor is used to implement the steps of the method for determining a drivable area based on millimeter-wave radar as described in any one of claims 1 to 6 when executing a computer program stored in the memory.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for determining a drivable area based on millimeter-wave radar according to any one of claims 1 to 6 is implemented.

Images