Roadside sensing method, device, electronic device, and storage medium

By receiving information from road-side perception devices, determining the scene and switching the perception fusion strategy, the problem of insufficient road-side perception accuracy and robustness in the prior art is solved, dynamic perception fusion is achieved, and the effect of road-side perception is improved.

CN115909728BActive Publication Date: 2024-11-26ZHIDAO NETWORK TECH (BEIJING) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202211366855.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-11-02
Publication Date
2024-11-26
Estimated Expiration
2042-11-02

AI Technical Summary

Technical Problem

The existing roadside perception technology mainly relies on vehicle-side solutions, and fails to effectively utilize the roadside characteristics to formulate targeted strategies, resulting in insufficient perception accuracy and robustness.

Method used

By receiving the perception information of the road-side perception device, the scene is determined and the perception fusion strategy is dynamically switched. According to the model and parameter configuration in the road-side configuration file, perceived fusion is adapted to different environmental conditions to improve perception accuracy and robustness.

Benefits of technology

Dynamic switching of roadside perception fusion algorithm based on environmental recognition is realized, improving the accuracy and robustness of roadside perception, and adapting to different scenarios and environmental conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115909728B_ABST
    Figure CN115909728B_ABST
Patent Text Reader

Abstract

The present application discloses a roadside perception method, device, electronic device, and storage medium. The method includes receiving first perception information of a roadside perception device; determining a first roadside perception scene according to the first perception information; and calling a corresponding first perception fusion strategy in a roadside configuration file according to the first roadside perception scene to perform perception fusion on the scene of a target area. The present application selects different perception fusion strategies for the perception environment in a typical scene to improve the accuracy and robustness of roadside perception.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of roadside sensing technology, and in particular to a roadside sensing method, device, electronic device, and storage medium. Background Art

[0002] With the continuous development of smart transportation, roadside monitoring systems are becoming more and more perfect. However, the current roadside perception fusion method basically refers to the vehicle-side solution. For the vehicle-side perception problem, the roadside perception has not formulated targeted strategies based on the characteristics of the roadside.

[0003] Compared with vehicle-side perception equipment, the environment in which roadside perception equipment is located is relatively stable, and it is easier to model, analyze and develop solutions for perception problems caused by some environments. Summary of the invention

[0004] The embodiments of the present application provide a roadside perception method, device, electronic device, and storage medium to implement the calling and switching of perception strategies in different scenarios.

[0005] The present application embodiment adopts the following technical solutions:

[0006] In a first aspect, an embodiment of the present application provides a roadside perception method, wherein the method includes:

[0007] Receiving first perception information from a roadside perception device;

[0008] Determining a first roadside perception scenario according to the first perception information;

[0009] The first perception fusion strategy corresponding to the roadside configuration file is called according to the first roadside perception scenario to perform perception fusion on the scene of the target area.

[0010] In some embodiments, the method further comprises:

[0011] Determining a second roadside perception scenario according to the second perception information, wherein the second perception information and the first perception information are perception results in different environments;

[0012] The first roadside perception scenario calls the corresponding first perception fusion strategy in the roadside configuration file, and dynamically switches to the second perception fusion strategy corresponding to the second roadside perception scenario, wherein the second perception fusion strategy and the first perception fusion strategy are different algorithm strategies in different or same environments.

[0013] In some embodiments, the perception fusion strategy includes multiple roadside perception fusion models stored in a roadside configuration file.

[0014] In some embodiments, the perception fusion strategy includes different parameter configurations for the same algorithm strategy saved in a roadside configuration file.

[0015] In some embodiments, calling a first perception fusion strategy corresponding to a roadside configuration file according to the first roadside perception scene to perform perception fusion on a scene in a target area includes:

[0016] Obtaining a scene detection result through a scene recognition engine according to the local weather conditions and / or the roadside image recognition result in the first roadside perception scene;

[0017] According to the scene detection result, searching the roadside configuration file for a perception fusion strategy that matches the scene detection result;

[0018] The target area is sensed according to the parameters of the perception fusion model or algorithm strategy in the search result.

[0019] In some embodiments, determining a first roadside perception scenario according to the first perception information includes:

[0020] Using roadside weather information as redundant information of the first perception information;

[0021] Based on the redundant information of the first perception information, characteristic information of the first roadside perception scene is determined, wherein the characteristic information includes at least one of the following: rain, night, fog, and snow.

[0022] In some embodiments, the calling of the first perception fusion strategy corresponding to the roadside configuration file according to the first roadside perception scene, after perceiving and fusing the scene of the target area, further includes:

[0023] Communicate with the vehicle end and synchronize the perception fusion result of the target area with the vehicle end.

[0024] In a second aspect, an embodiment of the present application further provides a roadside sensing device, wherein the device comprises:

[0025] A receiving module, used to receive first perception information from a roadside perception device;

[0026] A determination module, configured to determine a first roadside perception scenario according to the first perception information;

[0027] A calling module is used to call the corresponding first perception fusion strategy in the roadside configuration file according to the first roadside perception scene, and perform perception fusion on the scene of the target area.

[0028] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor; and a memory arranged to store computer executable instructions, wherein the executable instructions, when executed, cause the processor to perform the above method.

[0029] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple application programs, the electronic device executes the above method.

[0030] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects: by receiving the first perception information of the roadside perception device, a first roadside perception scene can be determined; then, according to the first roadside perception scene, a corresponding first perception fusion strategy in the roadside configuration file is called to perceive the target area. Different perception fusion strategies are selected for the perception environment in typical scenes to improve the accuracy and robustness of roadside perception. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0032] Figure 1 This is a flow chart of a roadside sensing method in an embodiment of the present application;

[0033] Figure 2 This is a schematic diagram of the structure of a roadside sensing device in an embodiment of the present application;

[0034] Figure 3 This is a schematic diagram of the implementation process of the roadside sensing method in the embodiment of the present application;

[0035] Figure 4 It is a schematic diagram of the structure of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0037] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0038] The present application embodiment provides a roadside sensing method, such as Figure 1 As shown, a schematic flow chart of a roadside sensing method in an embodiment of the present application is provided, and the method at least includes the following steps S110 to S130:

[0039] Step S110: receiving first perception information from a roadside perception device.

[0040] The roadside perception method in the embodiment of the present application is based on the constructed roadside system.

[0041] The roadside system includes roadside sensing equipment, roadside servers, etc., and is usually deployed on multiple roadside units, with the roadside units as carriers. The roadside units are installed at the target intersection or target road section, and synchronize the perception fusion results of the roadside system to the vehicles in its coverage area to provide perception information for the vehicles. Roadside sensing equipment includes, but is not limited to, laser radar, cameras, and may also include temperature and humidity sensing modules, etc.

[0042] Here, a roadside unit is used as the smallest implementation unit for description. The roadside server on the roadside unit receives the first perception information of the roadside perception device. Since the position of the roadside unit is relatively fixed, the perception information of the roadside perception device also has relatively fixed / static background (environmental) information, and the moving objects mainly include vehicles, pedestrians, etc.

[0043] The roadside sensing device can use cameras and laser radars to sense the environment and obtain image information and corresponding laser point cloud information. At the same time, since the longitude and latitude positioning information of the roadside unit is fixed, the environmental information in the area where the roadside unit is located, such as weather information, can also be determined.

[0044] It should be noted that the roadside perception device itself can be written with image vision algorithms, laser radar and other algorithm programs to enhance the perception capabilities of the roadside perception device itself. Among them, the image vision algorithm includes but is not limited to YOLOv4. The laser radar algorithm includes but is not limited to the 2D / 3D object detection algorithm. The embodiment of the present application shows a specific implementation method of the aforementioned step S110. Of course, it should be understood that other methods can also be used for implementation, and the embodiment of the present application does not limit this.

[0045] Step S120: determining a first roadside perception scenario based on the first perception information.

[0046] Since the first perception information includes the environmental perception results in the current scene, and the position of the roadside equipment is fixed, the change in the environmental perception results may affect the accuracy of perception fusion in the roadside unit. It can be understood that the change in the environmental perception results includes but is not limited to perception information in typical scenes such as daytime, nighttime, rainy days, and snowy days.

[0047] After the first roadside perception scenario is determined, the environmental characteristics of the area where the roadside unit is currently located are known, and these environmental characteristics usually affect the stability and accuracy of the results of the roadside perception fusion.

[0048] Step S130: Calling a first perception fusion strategy corresponding to a roadside configuration file according to the first roadside perception scenario to perform perception fusion on the scene of the target area.

[0049] For the first roadside perception scenario, the previous perception fusion strategy can be adjusted (or not adjusted) by calling the corresponding first perception fusion strategy in the roadside configuration file, so as to better perceive the target area. The roadside configuration file is usually pre-written in the roadside server.

[0050] The roadside configuration file contains a mapping relationship between the perception fusion strategy and the roadside perception scenario.

[0051] It should be noted that the background environment of the target area here is relatively fixed for the roadside unit, while the changing factors of the foreground environment (such as temperature and light intensity) will affect the perception fusion results, so it is necessary to switch the corresponding perception fusion strategy to perceive the environmental information in the target area.

[0052] Preferably, in order to ensure real-time performance, the called roadside configuration file is saved locally, for example, at an intersection with heavy traffic during peak hours.

[0053] However, if the real-time requirement is not high, asynchronous processing can be used to obtain the roadside configuration file from the cloud, for example, during periods of low traffic.

[0054] Furthermore, for the vehicle side, when entering a complex and unfamiliar target area, the perception results of the vehicle side are often affected. At this time, the environmental information perceived in the target area can be synchronized to the vehicle side through the roadside unit for laser or visual SLAM mapping, etc., thereby helping the vehicle side to better plan paths and avoid obstacles.

[0055] The roadside perception method in the embodiment of the present application realizes dynamic switching of the roadside perception fusion algorithm based on environment recognition, thereby improving the accuracy and robustness of roadside perception. The perception environment is identified based on relatively fixed characteristics in the roadside perception information.

[0056] Based on the above-mentioned roadside perception method, a roadside perception system is provided in an embodiment of the present application, and the system includes: a roadside information collection module, a roadside perception device module, a scene detection engine, and an algorithm selection module.

[0057] Among them, the roadside perception device module mainly provides image data for scene detection through the camera. The roadside information collection module can provide information such as weather information of the day and mainly provide redundant information for scene recognition. The algorithm selection module mainly selects different perception algorithms based on the scene detection results in the scene detection engine.

[0058] Specifically, the roadside perception system obtains local weather information and then uses the camera to identify the environment. It then configures and / or adjusts the perception fusion algorithm based on the specific scenario and dynamically selects the corresponding algorithm based on the scene recognition output results of the scene detection engine.

[0059] In one embodiment of the present application, the method also includes: determining a second roadside perception scenario based on second perception information, wherein the second perception information and the first perception information are perception results in different environments; calling the first perception fusion strategy corresponding to the roadside configuration file for the first roadside perception scenario, and dynamically switching to the second perception fusion strategy corresponding to the second roadside perception scenario, wherein the second perception fusion strategy and the first perception fusion strategy are different algorithm strategies in different or the same environments.

[0060] In specific implementation, if the environmental information changes, such as light intensity, the second perception information will be generated by the roadside perception device. The second perception information and the first perception information are perception results in different environments. It should be noted that the perception information here can be the current weather information and the image information collected by the camera in the current target area. The current weather information can be obtained through the roadside unit's own sensors, or through the roadside server of the roadside unit after accessing the network.

[0061] It should be noted that the frequency of acquiring the second perception information may be periodic according to natural time or customized.

[0062] Furthermore, since the second perception fusion strategy and the first perception fusion strategy are different algorithm strategies in different or the same environment, when different algorithm strategies are selected according to the environment, the first roadside perception scene calls the corresponding first perception fusion strategy in the roadside configuration file, and dynamically switches to the second perception fusion strategy corresponding to the second roadside perception scene.

[0063] It should be noted that different algorithm strategies in different environments may include, but are not limited to, rainy days, foggy days, snowy days, and other different environments. Different algorithm strategies in the same environment may include, but are not limited to, daytime, night, and the same environment but different light intensities.

[0064] In one embodiment of the present application, the perception fusion strategy includes multiple roadside perception fusion models stored in a roadside configuration file.

[0065] In a specific implementation, the first perception fusion strategy corresponding to the roadside configuration file is called according to the first roadside perception scenario, or the second perception fusion strategy is a pre-trained fusion model stored in the roadside configuration file.

[0066] It can be understood that the fusion model is obtained by training the network to convergence through machine learning using multiple sets of data, and each set of data in the multiple sets of data includes: image data and / or weather information of the target area and type labels corresponding to the image data and / or weather information. According to the confidence of the perception fusion results of each roadside perception fusion model in different scenarios, a corresponding perception fusion strategy is established.

[0067] Exemplarily, multiple roadside perception fusion models may be saved according to different algorithm tables LIST.

[0068] In one embodiment of the present application, the perception fusion strategy includes different parameter configurations for the same algorithm strategy stored in a roadside configuration file.

[0069] In a specific implementation, the first perception fusion strategy or the second perception fusion strategy corresponding to the roadside configuration file is called according to the first roadside perception scenario. The strategy is a different parameter configuration for the same algorithm strategy stored in the roadside configuration file.

[0070] It should be noted that since different parameter configurations will affect the perception fusion results, if the same parameter configuration is used for the same algorithm strategy, it will not be able to adapt well to changes in environmental perception results, resulting in a decrease in the accuracy and timeliness of roadside perception fusion.

[0071] Exemplarily, different parameter configurations for the same algorithm strategy may be saved via the algorithm table LIST.

[0072] In one embodiment of the present application, the first perception fusion strategy corresponding to the roadside configuration file is called according to the first roadside perception scene to perform perception fusion on the scene of the target area, including: obtaining a scene detection result through a scene recognition engine according to the local weather conditions and / or roadside image recognition results in the first roadside perception scene; searching the roadside configuration file for a perception fusion strategy that matches the scene detection result according to the scene detection result; and sensing the target area according to the parameters of the perception fusion model or algorithm strategy in the search result.

[0073] When implementing it, Figure 3 As shown, in the embodiment of the present application, a scene recognition engine and an algorithm engine (i.e., an algorithm LIST, which includes Algorithm 1, Algorithm 2, and Algorithm 3, where Algorithms 1, 2, and 3 are only examples and are configured according to actual usage) are included. The scene recognition engine and the perception fusion strategy (algorithm) selection module are in an upstream and downstream relationship. As the upstream scene recognition engine, it will process some information collected in the current scene, including local weather conditions, image content recognition results, etc. The algorithm engine will output the scene detection results, and the algorithm LIST will select a suitable algorithm model or algorithm parameters according to the results of the scene detection.

[0074] It should be noted that multiple algorithm models or different parameter configurations of the same algorithm model will be saved in the roadside configuration file according to the needs. For example, when the algorithm engine recognizes that it is snowing on that day, the algorithm selection module will select the configuration parameters or inference model suitable for snowy days to ensure the vehicle detection performance in special weather conditions.

[0075] Preferably, if the local meteorological conditions and the image content recognition results have relatively obvious features that can be identified, the scene recognition result can be output based on one of the features. In addition, if the feature is not obvious, more information needs to be collected for judgment. For example, the image content recognition results identify the weather features after removing the background, whether it is snowing or raining. For another example, the local meteorological conditions can collect more temperature and humidity information.

[0076] In one embodiment of the present application, determining a first roadside perception scene based on the first perception information includes: using roadside weather information as redundant information of the first perception information; determining characteristic information of the first roadside perception scene based on the redundant information of the first perception information, wherein the characteristic information includes at least one of the following: rain, night, fog, and snow.

[0077] In specific implementation, for the above-mentioned first perception information (mainly image information or laser point cloud information), the roadside weather information can be used as its redundant information, so as to determine the characteristic information of the first roadside perception scene according to the redundant information, and these characteristic information include but are not limited to environmental characteristics such as rain, night, fog, snow, etc. In addition, it should be noted that the different environmental characteristics in the characteristic information cover the influence of factors such as different lighting conditions and visibility.

[0078] In one embodiment of the present application, after calling the corresponding first perception fusion strategy in the roadside configuration file according to the first roadside perception scene and perceptually fusing the scene of the target area, it also includes: communicating with the vehicle side to synchronize the perception fusion result of the target area with the vehicle side.

[0079] By communicating with the vehicle side through the V2X protocol, the perception fusion result of the target area can be synchronized with the vehicle side that enters the target area. The vehicle side may or may not have autonomous driving capabilities. While improving the accuracy of roadside perception, the perception data and environmental information in the perception area can be sent to the vehicle side, thereby compensating for the disadvantages of the vehicle side in perception in complex and unfamiliar environments, and helping the vehicle side with autonomous driving capabilities to better plan paths and avoid obstacles.

[0080] The present application also provides a roadside sensing device 200, such as Figure 2 As shown, a schematic diagram of the structure of a roadside sensing device in an embodiment of the present application is provided, wherein the roadside sensing device 200 at least includes: a receiving module 210, a determining module 220 and a calling module 230, wherein:

[0081] In one embodiment of the present application, the receiving module 210 is specifically used to: receive first perception information from a roadside perception device.

[0082] The roadside system includes roadside sensing equipment, roadside servers, etc., and is usually deployed on multiple roadside units, with the roadside units as carriers. The roadside units are installed at the target intersection or target road section, and synchronize the perception fusion results of the roadside system to the vehicles in its coverage area to provide perception information for the vehicles.

[0083] Here, a roadside unit is used as the smallest implementation unit for description. The roadside server on the roadside unit receives the first perception information of the roadside perception device. Since the position of the roadside unit is relatively fixed, the perception information of the roadside perception device also has relatively fixed background (environmental) information.

[0084] The roadside sensing device can use cameras and laser radars to sense the environment and obtain image information and corresponding laser point cloud information. At the same time, since the longitude and latitude positioning information of the roadside unit is fixed, the environment in the area where the roadside unit is located, such as weather information, can also be determined.

[0085] It should be noted that the roadside perception device itself can be written with image vision algorithms, laser radar and other algorithm programs to enhance the perception capabilities of the roadside perception device itself. Among them, image vision algorithms include but are not limited to YOLOv4. Laser radar algorithms include but are not limited to 2D / 3D object detection algorithms. The embodiment of the present application shows a specific implementation of the aforementioned receiving module 210. Of course, it should be understood that other methods can also be used for implementation, and the embodiment of the present application does not limit this.

[0086] In one embodiment of the present application, the determination module 220 is specifically used to: determine a first roadside perception scene according to the first perception information.

[0087] Since the first perception information includes the environmental perception results in the current scene, and the position of the roadside equipment is fixed, the change in the environmental perception results may affect the accuracy of perception fusion in the roadside unit. It can be understood that the change in the environmental perception results includes but is not limited to perception information in typical scenes such as daytime, nighttime, rainy days, and snowy days.

[0088] After the first roadside perception scenario is determined, the environmental characteristics of the area where the roadside unit is currently located are known, and these environmental characteristics usually affect the stability and accuracy of the results of the roadside perception fusion.

[0089] In one embodiment of the present application, the calling module 230 is specifically used to: call the corresponding first perception fusion strategy in the roadside configuration file according to the first roadside perception scene, and perform perception fusion on the scene of the target area.

[0090] For the first roadside perception scenario, the previous perception fusion strategy can be adjusted (or not adjusted) by calling the corresponding first perception fusion strategy in the roadside configuration file, so as to better perceive the target area. The roadside configuration file is usually pre-written in the roadside server. The roadside configuration file has a mutual mapping relationship between the perception fusion strategy and the roadside perception scenario.

[0091] It should be noted that the background environment of the target area here is relatively fixed for the roadside unit, while the changing factors of the foreground environment (such as temperature and light intensity) affect the perception fusion results, so it is necessary to switch the corresponding perception fusion strategy to perceive the environmental information in the target area.

[0092] Preferably, in order to ensure real-time performance, the called roadside configuration file is saved locally, for example, at an intersection with heavy traffic during peak hours.

[0093] However, if the real-time requirement is not high, asynchronous processing can be used to obtain the roadside configuration file from the cloud, for example, during periods of low traffic.

[0094] Furthermore, for the vehicle side, when entering a complex and unfamiliar target area, the perception results of the vehicle side are often affected. At this time, the environmental information perceived in the target area can be synchronized to the vehicle side through the roadside unit for laser or visual SLAM mapping, etc., thereby helping the vehicle side to better plan paths and avoid obstacles.

[0095] It can be understood that the above-mentioned roadside sensing device can implement each step of the roadside sensing method provided in the aforementioned embodiment, and the relevant explanations about the roadside sensing method are applicable to the roadside sensing device and will not be repeated here.

[0096] Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 4 At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0097] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0098] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0099] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a roadside sensing device at the logical level. The processor executes the program stored in the memory and is specifically used to perform the following operations:

[0100] Receiving first perception information from a roadside perception device;

[0101] Determining a first roadside perception scenario according to the first perception information;

[0102] The first perception fusion strategy corresponding to the roadside configuration file is called according to the first roadside perception scenario to perform perception fusion on the scene of the target area.

[0103] The above application Figure 1 The method performed by the roadside sensing device disclosed in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or instructions in software form. The above processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the field such as random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0104] The electronic device may also perform Figure 1 The method executed by the roadside sensing device in the Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.

[0105] The present application also provides a computer-readable storage medium, which stores one or more programs, wherein the one or more programs include instructions, which, when executed by an electronic device including multiple application programs, enable the electronic device to execute Figure 1 The method performed by the roadside sensing device in the illustrated embodiment is specifically used to perform:

[0106] Receiving first perception information from a roadside perception device;

[0107] Determining a first roadside perception scenario according to the first perception information;

[0108] The first perception fusion strategy corresponding to the roadside configuration file is called according to the first roadside perception scenario to perform perception fusion on the scene of the target area.

[0109] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0110] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks 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 processor, 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.

[0111] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate 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 A function specified in one or more boxes.

[0112] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0113] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0114] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0115] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0116] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0117] 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 adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0118] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A roadside perception method, wherein: The roadside perception system is implemented based on the roadside perception system, which includes: a roadside information collection module, a roadside perception device module, a scene detection engine, and an algorithm selection module. The roadside sensing device module at least includes a laser radar, a camera, and a temperature and humidity sensing module; the sensing information of the roadside sensing device module includes relatively static environmental information and moving object information; The scene detection engine and the algorithm selection module are in an upstream and downstream relationship. The scene detection engine as the upstream will process the information collected in the current scene, and the algorithm selection module outputs the scene detection result and selects a suitable algorithm model or algorithm parameters; The method comprises: Receive first perception information from a roadside perception device, where the first perception information includes an environmental perception result in a current scene, and a change in the environmental perception result includes perception information in daytime, nighttime, rainy day, and snowy day scenes; Determining a first roadside perception scenario according to the first perception information; Determining a second roadside perception scenario according to the second perception information, wherein the second perception information and the first perception information are perception results in different environments; Dynamically switch the first perception fusion strategy corresponding to the roadside configuration file called according to the first roadside perception scenario to the second perception fusion strategy corresponding to the second roadside perception scenario, wherein the second perception fusion strategy and the first perception fusion strategy are different algorithm strategies in different or same environments; the second perception information and the first perception information include current weather information and image information collected by a camera in the current target area, the current weather information is obtained through a sensor in a roadside unit, or is obtained through a roadside server after accessing a network, and the frequency of obtaining the second perception information is periodic acquisition according to natural time or custom acquisition; According to the first roadside perception scenario, a first perception fusion strategy corresponding to the roadside configuration file is called to perform perception fusion on the scene of the target area, including: Obtaining a scene detection result through a scene recognition engine according to the roadside weather information and / or the roadside image recognition result in the first roadside perception scene; According to the scene detection result, searching the roadside configuration file for a perception fusion strategy that matches the scene detection result; Perceiving the target area according to the parameters of the perception fusion model or algorithm strategy in the search result; The perception fusion strategy includes multiple roadside perception fusion models stored in the roadside configuration file, the fusion model is obtained by training the network to convergence through machine learning using multiple sets of data, each set of data in the multiple sets of data includes: image data and / or weather information of the target area and type labels corresponding to the image data and / or weather information; according to the confidence of the perception fusion results of each roadside perception fusion model in different scenarios, a corresponding perception fusion strategy is established; Multiple roadside perception fusion models can be saved according to different algorithm tables LIST; The perception fusion strategy includes different parameter configurations for the same algorithm strategy saved in the roadside configuration file. The different parameter configurations for the same algorithm strategy can be saved through an algorithm table LIST.

2. The method of claim 1, wherein: The determining, according to the first perception information, a first roadside perception scenario includes: Using roadside weather information as redundant information of the first perception information; Based on the redundant information of the first perception information, characteristic information of the first roadside perception scene is determined, wherein the characteristic information includes at least one of the following: rain, night, fog, and snow.

3. The method according to any one of claims 1 to 2, wherein: After calling the first perception fusion strategy corresponding to the roadside configuration file according to the first roadside perception scene and performing perception fusion on the scene of the target area, the method further includes: Communicate with the vehicle end and synchronize the perception fusion result of the target area with the vehicle end.

4. A roadside sensing device, wherein: The roadside perception system is implemented based on the roadside perception system, which includes: a roadside information collection module, a roadside perception device module, a scene detection engine, and an algorithm selection module. The roadside sensing device module at least includes a laser radar, a camera, and a temperature and humidity sensing module; the sensing information of the roadside sensing device module includes relatively static environmental information and moving object information; The scene detection engine and the algorithm selection module are in an upstream and downstream relationship. The upstream scene detection engine processes the information collected in the current scene, and the algorithm selection module outputs the scene detection result and selects a suitable algorithm model or algorithm parameters. The device includes: A receiving module, configured to receive first perception information of a roadside perception device, wherein the first perception information includes an environmental perception result in a current scene, and a variation in the environmental perception result includes perception information in scenes of day, night, rain, and snow; A determination module, configured to determine a first roadside perception scenario according to the first perception information; The determining module is also used for: Determining a second roadside perception scenario according to the second perception information, wherein the second perception information and the first perception information are perception results in different environments; Dynamically switch the first perception fusion strategy corresponding to the roadside configuration file called according to the first roadside perception scenario to the second perception fusion strategy corresponding to the second roadside perception scenario, wherein the second perception fusion strategy and the first perception fusion strategy are different algorithm strategies in different or same environments; the second perception information and the first perception information include current weather information and image information collected by a camera in the current target area, the current weather information is obtained through a sensor in a roadside unit, or is obtained through a roadside server after accessing a network, and the frequency of obtaining the second perception information is periodic acquisition according to natural time or custom acquisition; A calling module, configured to call a first perception fusion strategy corresponding to a roadside configuration file according to the first roadside perception scene, and perform perception fusion on the scene of the target area; The calling module is also used for: Obtaining a scene detection result through a scene recognition engine according to the roadside weather information and / or the roadside image recognition result in the first roadside perception scene; According to the scene detection result, searching the roadside configuration file for a perception fusion strategy that matches the scene detection result; Perceiving the target area according to the parameters of the perception fusion model or algorithm strategy in the search result; The perception fusion strategy includes multiple roadside perception fusion models stored in the roadside configuration file, the fusion model is obtained by training the network to convergence through machine learning using multiple sets of data, each set of data in the multiple sets of data includes: image data and / or weather information of the target area and type labels corresponding to the image data and / or weather information; according to the confidence of the perception fusion results of each roadside perception fusion model in different scenarios, a corresponding perception fusion strategy is established; Multiple roadside perception fusion models can be saved according to different algorithm tables LIST; The perception fusion strategy includes different parameter configurations for the same algorithm strategy saved in the roadside configuration file. The different parameter configurations for the same algorithm strategy can be saved through an algorithm table LIST.

5. An electronic device, comprising: processor; as well as A memory arranged to store computer executable instructions, which when executed cause the processor to perform the method of any one of claims 1 to 3.

6. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, causes the electronic device to execute any one of the methods of claims 1 to 3.