Internet of vehicles multi-scene target detection system and method based on RISC-V architecture

Abstract

The invention relates to the technical field of Internet of Vehicles target detection, in particular to an Internet of Vehicles multi-scene target detection system and method based on an RISC-V architecture, and the method comprises the steps: S1, obtaining image data, storing the image data in a DMA cache region preset on an RISC-V hardware calculation module, and setting the DMA cache region to be of a fixed size; s2, mapping the image data in the DMA cache region to an address space of a process on an RISC-V hardware calculation module by utilizing an mmap mechanism; s3, preprocessing the image data by using an RVV instruction; s4, after target detection model parameters are loaded for the first time, caching the preprocessed image data into a heterogeneous shared memory by using a cache consistency protocol supported by an RISC-V hardware calculation module, performing model reasoning, and obtaining perception information; s5, performing post-processing on the model reasoning result through a self-defined post-processing function, and obtaining display information; and the real-time frame rate of target detection, the overall performance of the system and the memory access efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle networking target detection, and in particular to a vehicle networking multi-scenario target detection system and method based on RISC-V architecture. Background Art

[0002] With the rapid development of intelligent transportation systems, vehicle-to-everything (V2X) technology has become a key factor in achieving autonomous driving and improving road safety. In a vehicle-to-everything environment, real-time communication between vehicles and their surroundings (including other vehicles, pedestrians, infrastructure, etc.) is essential. In order to effectively handle complex traffic scenarios, high-performance target detection algorithms are required to identify potential safety risks. Traditional embedded processor architectures cannot meet this demand in terms of power consumption, cost, and flexibility.

[0003] RISC-V is an open instruction set architecture (ISA) based on the principles of reduced instruction set computing (RISC), which aims to provide a set of free and open standards for the industry. However, there is currently a lack of research based specifically on the RISC-V architecture in the field of vehicle network object detection.

[0004] In addition, since each frame of image data for target detection images usually occupies a large memory space, in the prior art, large-capacity memory is repeatedly requested and released when processing each frame of data, resulting in performance degradation; the YOLOv5 model is reloaded when processing each frame of image. Given that the number of parameters of the YOLOv5 model can reach millions or even tens of millions, this not only increases processing time but also leads to unnecessary memory usage; after the image data model inference is completed, the inference result is usually saved to a file, and then the result is read from the file, and finally the bounding box result is rendered to the original image. This method not only adds additional I / O operations but also causes processing delays. Summary of the invention

[0005] In view of the shortcomings of the prior art mentioned above, the purpose of the present invention is to provide a multi-scenario target detection system and method for the Internet of Vehicles based on the RISC-V architecture, which is used to solve the technical problems of low efficiency, high memory usage, high energy consumption, slow response speed, etc. of the vehicle target detection system.

[0006] To achieve the above object, the present invention provides a multi-scenario target detection method for Internet of Vehicles based on RISC-V architecture, comprising:

[0007] Step S1: storing the acquired image data in a DMA buffer area preset on the RISC-V hardware computing module, and the DMA buffer area is set to a fixed size;

[0008] Step S2: using the mmap mechanism to map the image data in the DMA buffer area to the address space of the process on the RISC-V hardware computing module;

[0009] Step S3: pre-process the image data using the RVV instruction;

[0010] Step S4: After loading the YOLOv5n target detection model parameters for the first time, the preprocessed image data is cached into the heterogeneous shared memory using the cache consistency protocol supported by the RISC-V hardware computing module, model inference is performed, and perception information is obtained;

[0011] Step S5: Post-process the model inference results through a custom post-processing function and obtain display information.

[0012] The present invention also provides a multi-scenario target detection system for Internet of Vehicles based on RISC-V architecture, comprising:

[0013] A data acquisition module, used for capturing peripheral image data of the vehicle;

[0014] The object detection module is used to process the captured image data to obtain display information;

[0015] A data communication module is used to obtain the display information of the own vehicle and share it with other vehicles, and to obtain the display information from other vehicles;

[0016] A user interaction module is used to obtain and present display information from the vehicle itself and other vehicles;

[0017] The target detection module and the user interface interaction module are both set on the RISC-V hardware computing module;

[0018] The target detection module performs target detection according to the method described above.

[0019] The beneficial effects of the present invention are as follows: the present invention presets a fixed uppercase DMA buffer area on the RISC-V hardware computing module, which can reduce the repeated application and release of memory, thereby improving the real-time frame rate of target detection and the overall performance of the system; step S2 realizes the mapping of virtual addresses to physical addresses, avoiding repeated mapping and memory reallocation during each frame of data processing; in this way, address mapping can be reused, thereby reducing the overhead of memory management and improving memory access efficiency. In addition, by utilizing the low power consumption characteristics of the RISC-V architecture development board, the method can continue to operate efficiently under limited energy conditions and reduce energy consumption; in addition, the flexible scalability of the RISC-V architecture development board ensures the high adaptability and upgradeability of the method, providing excellent performance and practicality for multi-scenario target detection in the Internet of Vehicles, and improving its portability and flexibility.

[0020] Compared with step S4, the previous method will reload the YOLOv5 model when processing each frame of the image. Given that the number of parameters of the YOLOv5 model can reach millions or even tens of millions, this not only increases the processing time, but also leads to unnecessary memory usage. In step S4 of the present method, the cache consistency protocol supported by the RISC-V hardware computing module ensures data consistency in the cache between multi-core processors, thereby avoiding repeated loading of model parameters when processing each frame of the image, effectively reducing memory usage and improving the overall response speed of the system to which the present method is applied.

[0021] Compared with step 5, the previous method is to save the model inference result to a file after the model inference is completed, then read the model inference result from the file, and finally render its bounding box result to the original image. This method not only adds additional I / O operations, but also causes processing delays; while this method directly post-processes the image data with the model inference result through a custom post-processing function after the model inference is completed, thereby reducing unnecessary I / O operations and improving the real-time performance and processing efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a structural diagram of a multi-scenario target detection system for an Internet of Vehicles based on a RISC-V architecture in an embodiment of the present invention;

[0023] Figure 2 This is a flowchart of a multi-scenario target detection method for Internet of Vehicles based on RISC-V architecture in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The following describes the embodiments of the present invention through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention.

[0025] It should be noted that the vehicles in the following embodiments are divided into own vehicles and other vehicles. "Own vehicle" specifically refers to the vehicle used as a reference, while "other vehicles" refers to any other vehicles except the own vehicle. The above-mentioned own vehicle and other vehicles are switched in name according to the changes of the reference vehicle.

[0026] This embodiment provides a multi-scenario target detection method for Internet of Vehicles based on RISC-V architecture. Figure 1 The system implementation shown includes:

[0027] The data acquisition module is used to capture the surrounding image data of the vehicle; in this example, the data acquisition module is composed of vehicle-mounted sensors, which include but are not limited to vehicle-mounted cameras, and the surrounding image data of the vehicle is captured through the vehicle-mounted sensors.

[0028] The object detection module is used to process the captured image data to obtain display information;

[0029] A data communication module is used to obtain the display information of the own vehicle and share it with other vehicles, and to obtain the display information from other vehicles;

[0030] A user interaction module is used to obtain and present display information from the vehicle itself and other vehicles;

[0031] The target detection module and the user interface interaction module are both set on the RISC-V hardware computing module. The target detection module includes a convolutional neural network (CNN) accelerator and a matrix operation unit. The image data obtained by the data acquisition module is pre-processed using RVV instructions, and the target detection model is used to perform detection reasoning and obtain perception information. The model reasoning results are output through the RISC-V hardware accelerator, and then the model reasoning results are post-processed through a custom post-processing function to obtain display information.

[0032] Specifically, the data communication module uses the Internet of Vehicles technology to realize the transmission of display information data between vehicles, and the V2V communication mode or the V2I communication mode can be selected; the user interaction module is used to present the display information. The user interaction module includes an on-board display screen and a visual interface software system. The display information includes the category, location, size and confidence of potential obstacles such as pedestrians and vehicles ahead, as well as information such as traffic lights and road signs, so as to facilitate the drivers of their own vehicles and other vehicles to make corresponding decisions according to the following different scenario requirements:

[0033] Scenario 1: Blind-spot obstacle detection: The vehicle observes potential obstacles such as pedestrians and vehicles in its own blind spot in real time through the user interface interaction module, so as to take measures in advance to avoid traffic collision accidents such as "ghost heads";

[0034] Scenario 2: Traffic sign recognition: The vehicle observes the shared detected traffic sign information and traffic light status in real time through the user interface interaction module. If the vehicle is about to violate traffic regulations, such as speeding or not following specific sign instructions, a warning can be immediately issued to the driver, prompting the driver to slow down or follow the instructions of specific signs;

[0035] Scenario three, lane line detection: The vehicle observes the detected lane line information shared by other vehicles in real time through the user interface interaction module. In situations where the field of vision is limited, such as curves and tunnel entrances, it can ensure that the vehicle is driving in the correct lane and prevent unintentional lane changes.

[0036] The multi-scenario target detection method of the Internet of Vehicles based on the RISC-V architecture in this embodiment is applied to a vehicle equipped with a RISC-V architecture development board as a central control unit. The RISC-V architecture development board constructs a RISC-V hardware computing module. The RISC-V hardware computing module is connected to the vehicle-mounted sensor of the data acquisition module, the Internet of Vehicles device OBU of the data communication module, and the vehicle-mounted display screen of the user interface interaction module through its own USB, HDMI, Ethernet and other different interfaces; at the same time, the YOLOv5n target detection model of the target detection module and the visual interface software system of the user interface interaction module are compiled and set on the RISC-V architecture development board. Preferably, the YOLOv5n target detection model is transplanted to the RISC-V hardware computing module through cross-compilation. The flowchart of this method is basically as follows: Figure 2 As shown, including:

[0037] Step S1: storing the acquired image data in a DMA (direct memory access) buffer area preset on the RISC-V hardware computing module, and the DMA buffer area is set to a fixed size;

[0038] Step S2: using the mmap mechanism to map the image data in the DMA buffer area to the address space of the process on the RISC-V hardware computing module;

[0039] Step S3: pre-processing the image data using RVV instructions (RISC-V Vector Extension);

[0040] Step S4: After the target detection model parameters are loaded for the first time, the preprocessed image data is cached into the heterogeneous shared memory using the cache consistency protocol (such as MESI or MOESI protocol) supported by the RISC-V hardware computing module, model inference is performed, and perception information is obtained;

[0041] Step S5: Post-process the model inference results through a custom post-processing function and obtain display information.

[0042] In the prior art, since each frame of image data usually occupies a large memory space, the previous method repeatedly applies for and releases a large amount of memory when processing each frame of data, resulting in performance degradation. The DMA buffer area preset on the RISC-V hardware computing module of the present invention can reduce the repeated application and release of memory, thereby improving the real-time frame rate of target detection and the overall performance of the system. In addition, considering that the image size output by the camera is fixed, the DMA buffer area of ​​the present invention is set to a fixed size to avoid memory waste. Step S2 realizes the mapping of virtual addresses to physical addresses, avoiding repeated mapping and memory reallocation when processing each frame of data; in this way, address mapping can be reused, thereby reducing the overhead of memory management and improving memory access efficiency. In addition, by utilizing the low power consumption characteristics of the RISC-V architecture development board, the present method can continue to operate efficiently under limited energy conditions and reduce energy consumption; in addition, the flexible scalability of the RISC-V architecture development board ensures the high adaptability and upgradeability of the present method, providing excellent performance and practicality for multi-scenario target detection in the Internet of Vehicles. At the same time, the present invention accelerates the implementation of AI products based on the RISC-V architecture in the field of Internet of Vehicles, further enriches its application scenarios in different fields, and improves its portability and flexibility.

[0043] Compared with step S4, the previous method will reload the target detection model when processing each frame of the image. Generally, the number of parameters of the target detection model can reach millions or even tens of millions, which not only increases the processing time, but also leads to unnecessary memory usage. In step S4 of the present method, the cache consistency protocol supported by the RISC-V hardware computing module ensures the data consistency of the cache between multi-core processors, avoids repeated loading of model parameters when processing each frame of the image, effectively reduces memory usage, and at the same time improves the overall response speed of the system to which the present method is applied.

[0044] Compared with step 5, the previous method is to save the model inference result to a file after the model inference is completed, then read the model inference result from the file, and finally render its bounding box result to the original image. This method not only adds additional I / O operations, but also causes processing delays; while this method directly post-processes the image data with the model inference result through a custom post-processing function after the model inference is completed, thereby reducing unnecessary I / O operations and improving the real-time performance and processing efficiency of the system.

[0045] In this example, the post-processing is to render the bounding box contained in the perception information onto the corresponding image.

[0046] The acquired image data includes but is not limited to photographed blind spot obstacles, traffic signs, lane lines, etc., which are all objects of target detection.

[0047] In this example, before step S1, step S0 is also included: the target detection module set on the RISC-V hardware computing module calls the open function to obtain input image data. The image data is the vehicle surrounding image data captured by the data acquisition module. Calling the open function to obtain the input image data avoids the additional overhead caused by using a third-party library, because when obtaining the input image, the previous method usually obtains the camera image by calling a third-party library (such as OpenCV, etc.), which generates additional overhead.

[0048] In this example, the preprocessing in step S3 includes scaling and filtering operations on the image data.

[0049] The RVV instruction provides a vectorized calculation method. The preprocessing in step S3 uses vector load and store instructions to access image data, reducing data access overhead. In addition, the RVV instruction can calculate the interpolation values ​​of multiple target pixels at the same time, further shortening the operation time. In this example, the following main RVV instructions are used to implement bilinear interpolation image scaling and 3×3 convolution filtering operations:

[0050] First, the "vsetvlmax_e8m1" instruction is used to determine the vector length of the current batch processing, so that the program can automatically adjust the number of elements processed in parallel according to the actual hardware resources;

[0051] Second, use the load instruction "vle8_v_i8m1" to load the 2x2 neighborhood pixels of the original image to provide basic data for bilinear interpolation;

[0052] Third, use the multiplication instruction "vwmul_vx_i16m2" and the accumulation instruction "vadd_vv_i16m2" to calculate the weights of horizontal interpolation and vertical interpolation, and use the reduction instruction "vnclip_wx_i8m1" to reduce the intermediate results;

[0053] Fourth, a 3x3 convolution filtering operation is performed on the calculated interpolation result through multiple multiplication instructions "vwmul_vx_i16m2" and accumulation instructions "vadd_vv_i16m2";

[0054] Fifth, use the storage instruction "vse8_v_i8m1" to store the final processing result into the output buffer.

[0055] The target detection model loaded in step S4 is a YOLOv5 model, but is not limited thereto and may also be other detection models.

[0056] In the model inference stage, the NPU (neural network processing unit) in the RISC-V hardware computing module is used to process multi-channel convolution and matrix multiplication in parallel. Compared with the traditional CPU inference method, the NPU is specially optimized for deep learning operations. With its parallel computing capability, it significantly improves the inference speed, optimizes the real-time performance of target detection, and further reduces the delay of model inference.

[0057] Preferably, steps S3 to S5 are processed in parallel using multiple computing core units through a pipeline mechanism. In the prior art, the previous methods usually adopt a serial execution method in the preprocessing, model reasoning and post-processing stages of image data processing, which makes it impossible to fully utilize its hardware resources, resulting in limited overall processing speed. However, this method uses multiple computing core units through a pipeline mechanism and parallel processing, and image processing tasks of different frames can execute different processing stages at the same time, such as Figure 2 As shown, specifically, in the same time period, while the first frame is performing post-processing, the second frame is performing model inference, and the third frame is performing pre-processing; this structure can improve processing efficiency and enable the system to achieve higher throughput when processing multiple consecutive frames of images.

[0058] In this example, the use of Hart (hardware thread) in the RISC-V hardware computing module enables collaborative work in a multi-core environment. The preprocessing, model reasoning, and post-processing tasks are divided into multiple threads, each thread is processed by a Hart, and Harts can execute their respective tasks simultaneously on different core units.

[0059] In this example, a core A is pre-allocated for pre-processing tasks. Within core A, multiple Harts are assigned to process different image regions or processing steps in parallel. Hart1 is responsible for image scaling, and Hart2 is responsible for filtering.

[0060] Assign a core B to perform model reasoning tasks. In core B, assign multiple Harts to perform different model layers or computing tasks. Hart1 is responsible for convolution operations, Hart2 is responsible for activation function calculations, Hart3 is responsible for feature fusion, and Hart4 is responsible for candidate box decoding and output calculations.

[0061] A core C is assigned to process the work task. Within core C, multiple Harts are assigned to perform different post-processing tasks. Hart1 is responsible for drawing the rendering bounding box, and Hart2 is responsible for result integration.

[0062] The scheduling mechanism of the RISC-V hardware computing module allows Harts to coordinate work through synchronization mechanisms (such as semaphores, mutex locks, and barriers) to ensure smooth docking of tasks at each stage.

[0063] In this embodiment, after the image preprocessing is completed, the Harts responsible for preprocessing will send a signal through a semaphore or barrier to notify the Harts responsible for model reasoning that they can start working; the model reasoning Harts reads the preprocessed image data, performs convolution operations, activation function calculations, feature fusion, candidate box decoding and output calculations. When the model reasoning task is completed, the model reasoning Harts will pass the model reasoning results to the post-processing Harts through shared memory or message passing mechanisms, and release the semaphore or reach the barrier to notify the post-processing Harts to start reading the model reasoning results and perform post-processing tasks.

[0064] The above-mentioned task division and synchronization mechanism ensures that all stages of image processing can be smoothly connected without delays due to resource competition or data inconsistency, thereby significantly improving the real-time performance of the overall system; this method not only optimizes the efficiency of each processing stage, but also improves the resource utilization of the overall system, thereby significantly improving the real-time frame rate performance of target detection.

[0065] In this example, the method also includes step S6: the data communication module obtains the display information of its own vehicle and shares the display information with other vehicles. At the same time, the data communication module obtains the display information of other vehicles; the user interface interaction module set on the RISC-V hardware computing module obtains and presents the display information of its own vehicle and other vehicles.

[0066] Specifically, the data communication module uses the Internet of Vehicles technology to share the displayed information with other vehicles. Preferably, the Internet of Vehicles technology uses C-V2X technology based on cellular networks to achieve real-time data exchange. Specifically, this function can be achieved through either V2V (Vehicle-to-Vehicle) communication or V2I (Vehicle-to-Infrastructure) communication.

[0067] In V2V communication mode, the on-board unit (OBU) of other vehicles encodes the display information detected by their own vehicle into a specific data packet; secondly, the data packet is directly sent to the on-board unit (OBU) of the own vehicle via C-V2X wireless communication technology;

[0068] In the V2I communication mode, the on-board unit (OBU) of other vehicles encodes the detection information into specific data packets and transmits it to the nearest roadside unit (RSU) through C-V2X wireless communication technology. The roadside unit (RSU) parses the received information and forwards it to the on-board unit (OBU) of its own vehicle.

[0069] Both V2V and V2I communication modes can ensure that the driver of the own vehicle can receive and observe the display information of other vehicles after identification and detection in real time, so as to understand the road conditions in advance and make corresponding driving decisions accordingly;

[0070] Furthermore, the user interaction module includes an on-board display screen and a visualization interface software system. The on-board display screen is used to display the visualization interface software system and to display real-time display information after target detection, including at least one of image information or point cloud information, to assist the driver in making corresponding decisions in a timely manner.

[0071] Preferably, the visualization interface software system is written using the cross-platform C++ application development framework Qt5 and ported to the RISC-V hardware computing module for compilation and operation.

[0072] The visualization interface software system includes a display information acquisition function and a network debugging assistant function. The display information acquisition function is used to display the display information received by the vehicle from other vehicles, as well as the display information of the vehicle itself; the network debugging assistant function is used to test the communication connection status between vehicle networking devices to ensure normal communication.

[0073] This method can cover the following typical application scenarios:

[0074] Blind-spot obstacle detection: In the crossroads scenario, the side area is usually a visual blind spot. Pedestrians or vehicles that suddenly rush out of this area are very likely to cause traffic accidents. In this scenario, sensors such as cameras are used to identify obstacles in the surrounding environment such as pedestrians and vehicles, and the information is shared with other vehicles to prevent traffic collision accidents such as "ghost heads";

[0075] Traffic sign recognition: If a large truck blocks the view in front of the vehicle, the vehicle may not be able to observe the status of the traffic light, and may blindly follow the vehicle in front and mistakenly run a red light. In this scenario, the vehicle in front detects and recognizes the current status of the traffic light and other road signs, and shares them with the vehicle behind, so that the vehicle behind can make the right decision.

[0076] Lane line detection: In situations where vision is limited, such as on bends and at tunnel entrances, it may be difficult for the driver to clearly identify lane markings, which can easily cause the vehicle to deviate from the correct lane and cause unintentional lane change accidents. In this scenario, lane markings are detected in real time through sensors such as cameras to ensure that the vehicle remains in the correct lane, effectively preventing unintentional lane changes.

[0077] The openness of RISC-V allows it to be implemented on a variety of architecture platforms, such as processor cores, microcontrollers, and embedded devices. This high degree of portability enables software developers to write architecture-independent code and easily port it between different RISC-V architectures. The RISC-V architecture is highly customizable due to its modular design, and the necessary extended instruction sets can be added according to specific application scenarios, thereby optimizing the execution efficiency of the target detection algorithm. At the same time, RISC-V can significantly reduce power consumption while maintaining high performance, which is especially important for long-term operation and battery-powered in-vehicle equipment. In addition, the open standard instruction set architecture (ISA) makes the design and manufacturing costs of RISC-V chips relatively low.

[0078] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present invention. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A multi-scenario target detection method for Internet of Vehicles based on RISC-V architecture, characterized in that: include: Step S1: storing the acquired image data in a DMA buffer area preset on the RISC-V hardware computing module, and the DMA buffer area is set to a fixed size; Step S2: using the mmap mechanism to map the image data in the DMA buffer area to the address space of the process on the RISC-V hardware computing module; Step S3: pre-process the image data using the RVV instruction; Step S4: After the target detection model parameters are loaded for the first time, the pre-processed image data is cached into the heterogeneous shared memory using the cache consistency protocol supported by the RISC-V hardware computing module, model inference is performed, and perception information is obtained; Step S5: Post-process the model inference results through a custom post-processing function and obtain display information.

2. The method according to claim 1, characterized in that The loaded target detection model is the YOLOv5 model.

3. The method according to claim 1, characterized in that The preprocessing process in step S3 includes scaling and filtering operations on the image data.

4. The method according to claim 1, characterized in that: The image data is accessed using vector load and store instructions during the preprocessing process in step S3.

5. The method according to claim 1, characterized in that In the model inference in step S4, the NPU in the RISC-V hardware computing module is used to process multi-channel convolution and matrix multiplication in parallel.

6. The method according to claim 1, characterized in that The post-processing includes rendering the bounding box included in the perception information onto the corresponding image.

7. The method according to claim 1, characterized in that The method further includes utilizing a plurality of computing core units to process steps S3 to S5 in parallel through a pipeline mechanism.

8. The method according to claim 7, characterized in that It also includes pre-allocating a core A to perform pre-processing tasks, and in the core A, allocating multiple Harts to process different image regions or processing steps in parallel, wherein Hart1 is responsible for image scaling, and Hart2 is responsible for filtering; Assign a core B to perform model reasoning tasks. In core B, assign multiple Harts to perform different model layers or computing tasks. Hart1 is responsible for convolution operations, Hart2 is responsible for activation function calculations, Hart3 is responsible for feature fusion, and Hart4 is responsible for candidate box decoding and output calculations. A core C is assigned to perform post-processing tasks. Within core C, multiple Harts are assigned to perform different post-processing tasks. Hart1 is responsible for drawing the rendering bounding box, and Hart2 is responsible for result integration.

9. The method according to claim 1, characterized in that: The method also includes step S6: each vehicle shares its own display information with other vehicles, and simultaneously presents the display information from itself and other vehicles.

10. A multi-scenario target detection system for Internet of Vehicles based on RISC-V architecture, characterized in that: include: A data acquisition module, used for capturing peripheral image data of the vehicle; The object detection module is used to process the captured image data to obtain display information; A data communication module is used to obtain the display information of the own vehicle and share it with other vehicles, and to obtain the display information from other vehicles; A user interaction module is used to obtain and present display information from the vehicle itself and other vehicles; The target detection module and the user interface interaction module are both set on the RISC-V hardware computing module; The target detection module performs target detection according to the method according to any one of claims 1 to 9.

Images