Road region detection method and apparatus, and electronic device

Abstract

The present invention relates to the technical field of autonomous driving. Disclosed are a road region detection method and apparatus, and an electronic device. The method comprises: using LiDAR provided on a vehicle to acquire original point cloud data of a region to be subjected to identification, and performing grid cell division on the original point cloud data, so as to obtain first-type grid cells and second-type grid cells, wherein each first-type grid cell comprises at least one piece of original point cloud data, and none of the second-type grid cells comprises any original point cloud data; using a preset road detection model to determine a road type of each first-type grid cell on the basis of the original point cloud data in the first-type grid cell, wherein the road type comprises a road region and a road boundary; and selecting reference grid cells from among the first-type grid cells, and using the road type of each reference grid cell and a sensing range parameter of the LiDAR to determine a road type of each second-type grid cell. Thus, the accuracy of road region detection is improved.

Description

Road area detection method, device and electronic equipment Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a road area detection method, device, and electronic equipment. Background Art

[0002] With the development of computer and artificial intelligence technologies, autonomous driving technology has become a research hotspot and is undergoing rapid development and iteration. Road area detection can provide information for driving decisions, path planning, and control, and can also provide auxiliary information for obstacle detection. Therefore, road area detection is a hot topic of research and attention in the field.

[0003] Related technologies typically utilize visual information and deep learning models to detect road areas. However, visual information is easily affected by external lighting conditions and lacks depth information. Therefore, it is not suitable for sensing certain special or complex scenes, resulting in low accuracy in detecting road areas and difficulty meeting the high robustness requirements of autonomous driving.

[0004] Summary of the Invention

[0005] The present invention provides a road area detection method, device and electronic equipment to solve the problem of low road area detection accuracy in related technologies.

[0006] According to one aspect of the present invention, a road area detection method is provided, comprising:

[0007] Acquiring raw point cloud data of the area to be identified using a laser radar provided on the vehicle, and performing grid division on the raw point cloud data to obtain a first type of grid and a second type of grid; wherein the first type of grid includes at least one raw point cloud data, and the second type of grid does not include any raw point cloud data;

[0008] Using a preset road detection model, determining the road type of the first type of grid based on the original point cloud data of the first type of grid; wherein the road type includes a road area and a road boundary;

[0009] A reference grid is selected from the first type of grids, and the road type of the reference grid and the perception range parameter of the laser radar are used to determine the road type of the second type of grid.

[0010] According to another aspect of the present invention, there is provided a road area detection device, comprising:

[0011] An acquisition unit, configured to acquire original point cloud data of the area to be identified using a laser radar provided on the vehicle;

[0012] A gridding unit, configured to grid the original point cloud data to obtain a first type of grid and a second type of grid; wherein the first type of grid includes at least one original point cloud data; and the second type of grid does not include any original point cloud data;

[0013] a detection unit, configured to determine the road type of the first type of grid based on the original point cloud data of the first type of grid using a preset road detection model; wherein the road type includes a road area and a road boundary;

[0014] The detection unit is further configured to select a reference grid from the first type of grids, and determine the road type of the second type of grid using the road type of the reference grid and the perception range parameter of the laser radar.

[0015] According to another aspect of the present invention, an electronic device is provided, comprising:

[0016] at least one processor; and

[0017] a memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program executable by the at least one processor. The computer program is executed by the at least one processor so that the at least one processor can perform the road area detection method described in any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the road area detection method described in any embodiment of the present invention when executed.

[0020] The technical solutions of the embodiments of the present invention can detect road areas using point cloud data. Because point cloud data is less susceptible to external lighting conditions and contains scene depth information, it can improve the accuracy of road area detection, particularly in special or complex scenarios, such as extremely severe weather, thereby meeting the high robustness requirements of autonomous driving. Furthermore, gridding the raw point cloud data and performing detection on a grid-by-grid basis can reduce computational complexity and increase detection speed. For the second type of grids without raw point cloud data, the road type can be predicted based on the road type of the reference grid in the first type and the lidar's sensing range parameters, improving the detectability of the area to be identified.

[0021] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] FIG1 is a flow chart of a road area detection method provided in Embodiment 1 of the present invention;

[0024] FIG2 is a schematic diagram of a road area detection result provided according to the first embodiment of the present invention;

[0025] FIG3 is a flow chart of a road area detection method provided according to a second embodiment of the present invention;

[0026] FIG4 is a schematic structural diagram of a road area detection device provided in a third embodiment of the present invention;

[0027] FIG5 is a schematic structural diagram of an electronic device for implementing the road area detection method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "original", "first", "second", etc. in the description and claims of the present invention 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 of the present invention described herein can be implemented in an order other than those 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.

[0030] Example 1

[0031] Figure 1 is a flow chart of a road area detection method according to a first embodiment of the present invention. This embodiment is applicable to the detection of road areas and road boundaries of a moving vehicle. The method can be performed by an electronic device. As shown in Figure 1, the method includes:

[0032] In step 101, the laser radar installed on the vehicle is used to obtain the original point cloud data of the area to be identified, and the original point cloud data is gridded to obtain a first type of grid and a second type of grid; wherein the first type of grid includes at least one original point cloud data; and the second type of grid does not have original point cloud data.

[0033] Specifically, a laser radar can be set up in the four directions of the vehicle to obtain the original point cloud data of the front, back, left and right of the vehicle.

[0034] Specifically, a rectangular area of ​​a preset size centered on the vehicle can be determined as the area to be identified. For example, a range of 200 meters in front and behind and 64 meters in left and right centered on the vehicle can be determined as the area to be identified.

[0035] Specifically, the original point cloud data in the area to be identified may be grid-divided according to a preset unit grid size to obtain a first type of grid and a second type of grid.

[0036] Step 102 : Using a preset road detection model, the road type of the first type of grid is determined based on the original point cloud data of the first type of grid; wherein the road type includes a road area and a road boundary.

[0037] Among them, the road detection model is a neural network model pre-trained according to the point cloud data to be trained.

[0038] Specifically, all original point cloud data contained in the first type of grid and the coordinates of the first type of grid relative to the area to be identified can be input into the road detection model, and the road detection model outputs the road type of the area to be identified corresponding to the first type of grid.

[0039] Step 103 : Select a reference grid from the first type of grids, and use the road type of the reference grid and the perception range parameter of the laser radar to determine the road type of the second type of grid.

[0040] Specifically, considering that the wide-angle spread of the LiDAR point cloud may not fully cover the area to be identified, when detecting the road area in the area to be identified, a second type of grid that does not contain the original point cloud data may exist. The second type of grid generally exists at the far end of the LiDAR's perception range. Therefore, a reference grid that is closer to the second type of grid can be selected from the first type of grid. To ensure the accuracy of the prediction, the LiDAR's perception range parameters can be used to determine whether the reference grid and the second type of grid are within the LiDAR's perception range. If so, the road type of the second type of grid can be determined based on the road type of the reference grid.

[0041] Among them, the perception range parameters of the laser radar include at least the horizontal dynamic perception range, effective detection distance, maximum detection distance, etc.

[0042] Figure 2 shows a schematic diagram of the road area detection results. 0 represents a grid area within the area to be identified that is neither a road area nor a road boundary; 2 represents a grid area within the area to be identified that is a road boundary; and 1 represents a grid area within the area to be identified that is a road area. Connecting the grid areas marked as 2 forms the road boundary during vehicle travel. This road area detection result can be displayed to the user for easy viewing.

[0043] The technical solution provided by the embodiment of the present invention can use point cloud data to detect road areas. Since point cloud data is not easily affected by the external light environment and has depth information of the scene, it can improve the accuracy of road area detection, especially for some special or complex scenes, such as extremely bad weather, and thus meet the high robustness requirements of autonomous driving. In addition, the original point cloud data is gridded and detection is performed in units of grids, which can reduce computational complexity and increase detection speed. For the second type of grid without original point cloud data, the road type can be predicted based on the road type of the reference grid in the first type of grid and the perception range parameters of the lidar, thereby improving the detectability of the area to be identified.

[0044] Example 2

[0045] FIG3 is a flow chart of a road area detection method according to a second embodiment of the present invention. This embodiment refines steps 101, 102, and 103 in the first embodiment. As shown in FIG3 , the method includes:

[0046] In step 301, the laser radar installed on the vehicle is used to obtain the original point cloud data of the area to be identified, and the original point cloud data is projected to obtain a two-dimensional top view; the two-dimensional top view is meshed to obtain a first type of grid and a second type of grid; wherein the first type of grid includes at least one original point cloud data; and the second type of grid does not contain any original point cloud data.

[0047] The Bird's Eye View (BEV) technology can be used to project the original point cloud data to obtain a two-dimensional top-down view. Specifically, the x-axis and y-axis data of the three-dimensional coordinate information in the original point cloud data are retained, and the z-axis data is removed.

[0048] Specifically, the two-dimensional top view may be divided into grids according to a preset unit grid size to obtain a first type of grid and a second type of grid. The preset unit grid size may be, for example, a square with a side length of 0.2 meters.

[0049] Step 302 : Process the original point cloud data of the first type of grid using the feature extraction layer in the road detection model to obtain a process feature map.

[0050] The feature extraction layer may include a normalization layer, a linear rectification function layer, a convolution layer, and a downsampling layer. Specifically, the original point cloud data included in the first type of grid may be processed using the normalization layer and the linear rectification function layer, and then processed by the convolution layer to obtain an original feature map. The original feature map is then processed using the downsampling layer to obtain a process feature map.

[0051] The downsampling layer may be a maximum pooling layer.

[0052] Specifically, the downsampling layer can compress the size of the original feature map and compress the information in the original feature map to obtain a process feature map to extract more subtle feature information in the original feature map.

[0053] Step 303: Process the process feature map using the detection layer in the road detection model to obtain the road type of the first type of grid; wherein the input size of the feature extraction layer is the same as the output size of the detection layer; and the road type includes a road area and a road boundary.

[0054] Specifically, the detection layer can be the reverse process of the feature extraction layer. The detection layer can have a similar layer group structure as the feature extraction layer, and the input size of the feature extraction layer can be set to be the same as the output size of the detection layer to ensure that the position information of the first type of grid is retained to the greatest extent possible, reduce misclassification, and improve the detection accuracy of the model. The input size of the feature extraction layer refers to the input data size of the feature extraction layer. The output size of the detection layer refers to the output data size of the detection layer.

[0055] In addition, the road detection model does not use a fully connected layer to ensure that higher-resolution feature maps are retained in the output of the detection layer. At the same time, the size of the feature maps is reduced, which greatly reduces the number of parameters and model size during model training. This can effectively reduce memory usage during model training and improve computational efficiency.

[0056] In one implementable manner, the detection layer in the road detection model includes an upsampling layer, a convolution layer, a feature fusion layer, and a Softmax activation function layer; the upsampling layer is used to process the process feature map to obtain a sparse feature map; the convolution layer is used to process the sparse feature map to obtain a dense feature map; the feature fusion layer is used to perform feature fusion processing on the sparse feature map and the dense feature map to obtain a fused feature map; the Softmax activation function layer is used to process the fused feature map to obtain the road type of the first type of grid.

[0057] The Softmax activation function layer can be used as the output layer of the detection layer, and the normalization processing layer of the feature extraction layer can be used as the input layer of the feature extraction layer. The input data size of the normalization processing layer is the first size, and the output data size of the Softmax activation function layer is the second size, and the first size is equal to the second size.

[0058] Specifically, the upsampling layer can be used to expand the size of the process feature map to the size of the original feature map to obtain a sparse feature map.

[0059] The convolution layer can supplement the sparse feature map with information to obtain a dense feature map. The sparse feature map and the dense feature map have the same size.

[0060] Specifically, a weighted summation method can be used to fuse the sparse feature map and the dense feature map. The fused feature map has more comprehensive feature information, so the road type detection using the fused feature map can improve the detection accuracy.

[0061] Optionally, the upsampling layer is an indexed max pooling layer.

[0062] Step 304 : Determine a first longitudinal distance and a first lateral distance between the first type of grid and the vehicle; and determine a second longitudinal distance and a second lateral distance between the second type of grid and the vehicle; if the difference between the first longitudinal distance and the second longitudinal distance is less than a preset threshold, and the first lateral distance is greater than the second lateral distance, determine the first type of grid as a reference grid.

[0063] Specifically, in order to ensure detection accuracy, the preset threshold may be set to a smaller value.

[0064] In step 305 , if it is determined that the reference grid and the second type grid are within the sensing range of the laser radar according to the sensing range parameter of the laser radar, and the road type of the reference grid is a road boundary, the road type of the second type grid is determined to be a road area.

[0065] Specifically, to ensure detection accuracy, if the road type of the reference grid is determined to be a road boundary, the road type of the second type grid can be determined to be a road area. If the road type of all reference grids is a road area, the road type of the second type grid is not predicted.

[0066] Example 3

[0067] FIG4 is a schematic diagram of the structure of a road area detection device provided according to Embodiment 3 of the present invention. As shown in FIG4 , the device 400 includes:

[0068] An acquisition unit 410 is configured to acquire original point cloud data of the area to be identified using a laser radar provided on the vehicle;

[0069] The gridding unit 420 is configured to grid the original point cloud data to obtain a first type of grid and a second type of grid; wherein the first type of grid includes at least one original point cloud data; and the second type of grid does not include any original point cloud data;

[0070] The detection unit 430 is configured to use a preset road detection model to determine the road type of the first type of grid based on the original point cloud data of the first type of grid; wherein the road type includes a road area and a road boundary;

[0071] The detection unit 430 is further configured to select a reference grid from the first type of grids, and determine the road type of the second type of grid using the road type of the reference grid and the perception range parameter of the laser radar.

[0072] The detection unit 430 is specifically configured to: process the original point cloud data of the first type of grid using the feature extraction layer in the road detection model to obtain a process feature map;

[0073] The detection layer in the road detection model is used to process the process feature map to obtain the road type of the first type of grid; wherein the input size of the feature extraction layer is the same as the output size of the detection layer.

[0074] The detection layer in the road detection model includes an upsampling layer, a convolution layer, a feature fusion layer, and a Softmax activation function layer; the detection unit 430 is specifically used to: use the upsampling layer to process the process feature map to obtain a sparse feature map;

[0075] Use the convolution layer to process the sparse feature map to obtain a dense feature map;

[0076] The feature fusion layer is used to perform feature fusion processing on the sparse feature map and the dense feature map to obtain the fused feature map;

[0077] The fused feature map is processed using the Softmax activation function layer to obtain the road type of the first type of grid.

[0078] In one implementation, the upsampling layer is an indexed maximum pooling layer.

[0079] The detection unit 430 is specifically configured to: determine a first longitudinal distance and a first lateral distance between the first type of grid and the vehicle; and determine a second longitudinal distance and a second lateral distance between the second type of grid and the vehicle; if the difference between the first longitudinal distance and the second longitudinal distance is less than a preset threshold, and the first lateral distance is greater than the second lateral distance, determine the first type of grid as a reference grid;

[0080] If the reference grid and the second type of grid are determined to be within the perception range of the lidar according to the perception range parameters of the lidar, and the road type of the reference grid is a road boundary, the road type of the second type of grid is determined to be a road area.

[0081] The gridding unit 420 is specifically used to perform projection processing on the original point cloud data to obtain a two-dimensional top view;

[0082] The two-dimensional top view is meshed to obtain the first type of mesh and the second type of mesh.

[0083] The road area detection device provided in the embodiment of the present invention can execute the road area detection method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the road area detection method.

[0084] Example 4

[0085] FIG5 shows a block diagram of an electronic device 10 that can be used to implement an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0086] As shown in FIG5 , the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 and a random access memory (RAM) 13, that is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0087] Multiple components in electronic device 10 are connected to I / O interface 15, including an input unit 16, such as a keyboard and mouse; an output unit 17, such as various types of displays and speakers; a storage unit 18, such as a magnetic disk and optical disk; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0088] The processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any other suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the road area detection method.

[0089] In some embodiments, any of the above-described road area detection methods may be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of any of the above-described road area detection methods may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform any of the above-described road area detection methods in any other appropriate manner (e.g., via firmware).

[0090] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0091] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0092] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0093] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0094] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0095] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0096] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0097] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A road area detection method, characterized in that: include: Acquiring raw point cloud data of the area to be identified using a laser radar provided on the vehicle, and performing grid division on the raw point cloud data to obtain a first type of grid and a second type of grid; wherein the first type of grid includes at least one raw point cloud data, and the second type of grid does not include any raw point cloud data; Using a preset road detection model, determining the road type of the first type of grid based on the original point cloud data of the first type of grid; wherein the road type includes a road area and a road boundary; A reference grid is selected from the first type of grids, and the road type of the reference grid and the perception range parameter of the laser radar are used to determine the road type of the second type of grid.

2. The method according to claim 1, characterized in that The method of using a preset road detection model to determine the road type of the first type of grid according to the original point cloud data of the first type of grid includes: Processing the original point cloud data of the first type of grid using a feature extraction layer in the road detection model to obtain a process feature map; The process feature map is processed using a detection layer in the road detection model to obtain the road type of the first type of grid; wherein the input size of the feature extraction layer is the same as the output size of the detection layer.

3. The method according to claim 2, characterized in that The detection layer in the road detection model includes an upsampling layer, a convolution layer, a feature fusion layer, and a Softmax activation function layer; the process feature map is processed by the detection layer in the road detection model to obtain the road type of the first type of grid, including: Processing the process feature map using the upsampling layer to obtain a sparse feature map; Processing the sparse feature map using the convolutional layer to obtain a dense feature map; Using the feature fusion layer to perform feature fusion processing on the sparse feature map and the dense feature map to obtain a fused feature map; The fused feature map is processed using the Softmax activation function layer to obtain the road type of the first type of grid.

4. The method according to claim 3, characterized in that The upsampling layer is an indexed maximum pooling layer.

5. The method according to claim 1, wherein The selecting a reference grid from the first type of grids and determining the road type of the second type of grid using the road type of the reference grid and the perception range parameter of the laser radar includes: determining a first longitudinal distance and a first lateral distance from the first type of grid to the vehicle; and determining a second longitudinal distance and a second lateral distance from the second type of grid to the vehicle; if a difference between the first longitudinal distance and the second longitudinal distance is less than a preset threshold, and the first lateral distance is greater than the second lateral distance, determining the first type of grid as a reference grid; If, according to the perception range parameters of the laser radar, it is determined that the reference grid and the second type of grid are within the perception range of the laser radar, and the road type of the reference grid is a road boundary, then the road type of the second type of grid is determined to be a road area.

6. The method according to claim 1, characterized in that The grid division of the original point cloud data to obtain the first type of grid and the second type of grid includes: Projecting the original point cloud data to obtain a two-dimensional top view; The two-dimensional top view is meshed to obtain a first type of mesh and a second type of mesh.

7. A road area detection device, characterized in that: include: An acquisition unit, configured to acquire original point cloud data of the area to be identified using a laser radar provided on the vehicle; A gridding unit, configured to grid the original point cloud data to obtain a first type of grid and a second type of grid; wherein the first type of grid includes at least one original point cloud data; and the second type of grid does not include any original point cloud data; A detection unit is configured to use a preset road detection model to determine the road type of the first type of grid according to the original point cloud data of the first type of grid; wherein the road type includes road district and road boundaries; The detection unit is further configured to select a reference grid from the first type of grids, and determine the road type of the second type of grid using the road type of the reference grid and the perception range parameter of the laser radar.

8. The device according to claim 7, characterized in that The detection unit is specifically used for: Processing the original point cloud data of the first type of grid using a feature extraction layer in the road detection model to obtain a process feature map; The process feature map is processed using a detection layer in the road detection model to obtain the road type of the first type of grid; wherein the input size of the feature extraction layer is the same as the output size of the detection layer.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor. The computer program is executed by the at least one processor so as to enable the at least one processor to perform the road area detection method according to any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the road area detection method according to any one of claims 1 to 6 when executed.

Images