Vehicle-mounted multi-source heterogeneous device

Abstract

The utility model provides a vehicle-mounted multi-source heterogeneous device. The vehicle-mounted multi-source heterogeneous device comprises a vehicle head calculation module, a vehicle head sensing module, a vehicle tail calculation module and a vehicle tail sensing module. A single vehicle-mounted sensing mode is upgraded to a plurality of sensing modes, and outputs of a plurality of sensing devices are gathered through the fusion unit and then uniformly output to the calculation unit, so that wiring between a sensing side and a calculation side is greatly simplified. In addition, the fusion unit shares part of sensing data of the vehicle head and the vehicle tail through a data bus, the comprehensive judgment capacity of the whole vehicle is further enhanced, the sensing capacity of the all-weather, all-scene and all-parameter rail transit industry can be improved, and rail transit operation safety is guaranteed. Besides, in order to guarantee the sensing safety of the intelligent traffic industry, sensing side isomerism and computing side isomerism are designed, the reliability of the product is guaranteed, and the availability of the product is improved.

Description

Technical Field

[0001] This application relates to the field of train safety protection technology, and in particular to an on-board multi-source heterogeneous device. Background Technology

[0002] The train's autonomous sensing system is an intelligent system combining advanced electronic and computer technologies. It can comprehensively monitor and control the train's operation to ensure railway traffic safety. In recent years, with the rapid development of the railway industry and continuous technological updates, the train's active sensing system has become an important means of ensuring railway traffic safety.

[0003] The workflow of a train active sensing system mainly consists of three parts: information acquisition, information processing, and command issuance. The system collects relevant information about train operation through various sensing devices, such as train position, speed, and foreign object intrusion. This information is then transmitted to the computing unit for analysis. Based on pre-set operating rules and degradation strategies, multi-level operational control commands are triggered to prevent collisions. However, traditional technologies use directly connected sensing network elements to the computing module for information acquisition, resulting in numerous cables, increased interface failure points, limited deployment within the vehicle, and ineffective information sharing. Utility Model Content

[0004] The purpose of this application is to at least address one of the aforementioned technical deficiencies, particularly the problems of numerous cables, increased interface failure points, and limited deployment within vehicles for sensing network elements in the prior art.

[0005] This application provides an in-vehicle multi-source heterogeneous device, including a front computing module, a front sensing module, a rear computing module, and a rear sensing module;

[0006] The front-end perception module includes at least two first perception modules, each first perception module including a first fusion unit and multiple first perception units connected to the first fusion unit;

[0007] The vehicle front computing module includes a first computing module that corresponds to the first sensing module. The first computing module is connected to the corresponding first fusion unit via a network cable. The first fusion unit is used to package all received sensing signals and transmit them to the corresponding first computing module.

[0008] The rear-end perception module includes at least two second perception modules, each of which includes a second fusion unit and multiple second perception units connected to the second fusion unit.

[0009] The rear computing module includes a second computing module that corresponds one-to-one with the second sensing module. The second computing module is connected to the corresponding second fusion unit via a network cable. The second fusion unit is used to package all received sensing signals and transmit them to the corresponding second computing module.

[0010] The first fusion unit and the second fusion unit are also connected to the first computing module and the second computing module respectively via a data bus, for transmitting the target type sensing signal to the first computing module and the second computing module.

[0011] In one embodiment, the first sensing unit includes a near-focus camera, a far-focus camera, a lidar, a millimeter-wave radar, and an inertial navigation unit.

[0012] In one embodiment, the millimeter-wave radar is a 4D millimeter-wave radar.

[0013] In one embodiment, the second sensing unit includes a near-focus camera, a far-focus camera, a lidar, a millimeter-wave radar, and an inertial navigation unit.

[0014] In one embodiment, the data bus includes a CAN bus.

[0015] In one embodiment, the vehicle head computing module further includes a first decision module, which is connected to each of the first computing modules. The first computing module is used to preprocess the received sensor signals and output them to the first decision module. The first decision module is used to make control decisions based on the preprocessed sensor signals.

[0016] In one embodiment, the vehicle-mounted multi-source heterogeneous device further includes a front alarm device, which is used to issue an alarm signal under the control of the first decision module.

[0017] In one embodiment, the rear-end computing module further includes a second decision module connected to each of the second computing modules. The second computing module is used to preprocess the received sensor signals and output them to the second decision module. The second decision module is used to make control decisions based on the preprocessed sensor signals.

[0018] In one embodiment, the vehicle-mounted multi-source heterogeneous device further includes a rear-end alarm device, which is used to issue an alarm signal under the control of the second decision module.

[0019] In one embodiment, the first fusion unit and the second fusion unit include FPGA units.

[0020] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0021] The onboard multi-source heterogeneous device in this embodiment includes a front-end computing module, a front-end sensing module, a rear-end computing module, and a rear-end sensing module. It upgrades the onboard sensing method from a single sensing method to multiple sensing methods. A fusion unit aggregates the outputs from multiple sensing devices and then outputs them uniformly to the computing unit, greatly simplifying the wiring between the sensing and computing sides. The fusion unit also shares some sensing data from the front and rear of the vehicle via a data bus, further enhancing the overall judgment capability of the entire vehicle. This improves the all-weather, all-scenario, and all-parameter sensing capabilities of the rail transit industry, ensuring the safety of rail transit operations. Furthermore, to ensure sensing safety in the intelligent transportation industry, heterogeneous sensing and computing sides are designed, ensuring product reliability and improving product availability. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of the structure of an in-vehicle multi-source heterogeneous device provided in one embodiment of this application;

[0024] Figure 2 This is a schematic diagram of the structure of a vehicle-mounted multi-source heterogeneous device provided in another embodiment of this application. Detailed Implementation

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0027] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0028] In the description of the embodiments of this utility model, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings, or the orientation or positional relationships commonly used when the utility model product is in use. These are only for the convenience of describing the utility model and simplifying the description, and do not indicate or imply that the device or component referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the utility model. Furthermore, the terms "horizontal," "vertical," and "suspended" do not mean that the component is required to be absolutely horizontal or suspended, but rather that it can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," not that the structure must be completely horizontal, but can be slightly tilted. In the description of the embodiments of this utility model, "multiple" means at least two.

[0029] This application provides an in-vehicle multi-source heterogeneous device. Please refer to [link / reference]. Figure 1 It includes a front-end computing module 100, a front-end perception module, a rear-end computing module 300, and a rear-end perception module.

[0030] The front-end sensing module is located at the front of the vehicle and serves as the front-end for information collection. The first sensing module 200 is its basic unit. The front-end sensing module will include at least two first sensing modules 200, with each module serving as a heterogeneous backup to ensure the stability of the rail transit sensing equipment and prevent a single unit's malfunction from affecting the judgment.

[0031] The first sensing module 200 includes a first fusion unit 220 and various types of first sensing units 210 connected to the first fusion unit 220. The first sensing units 210 can be different types of sensors, such as lidar sensors, camera sensors, millimeter-wave radar sensors, etc., each sensing information about the vehicle's surrounding environment from different dimensions. The first fusion unit 220 is responsible for integrating and packaging the sensor signals acquired by these different types of first sensing units 210. Different first sensing units 210 collect data based on their own physical characteristics and working principles. For example, lidar determines the distance and contour information of target objects by emitting laser beams and receiving reflected light, while cameras capture image information through optical imaging principles. The first fusion unit 220 integrates this data from different sensing units according to a certain format and rules for subsequent transmission to the vehicle head computing module 100. Specifically, the various first sensing units 210 are rationally arranged at different positions on the front of the vehicle to obtain the optimal sensing field of view. For example, a camera is installed at a suitable position above the windshield, and lidar is installed near the front bumper. The first fusion unit 220 can be a data processing chip with multiple interfaces. It is connected to each first sensing unit 210 through a dedicated data transmission line, such as a high-speed serial data line. After receiving data from each sensing unit, it packages the data according to a predetermined communication protocol and data structure, and then transmits the packaged data to the corresponding vehicle head computing module 100 through a network cable.

[0032] The vehicle front computing module 100 is one of the core processing units in the vehicle-mounted multi-source heterogeneous device, located in the vehicle's front area. It is a collection of modules specifically designed to process and compute the data acquired by the vehicle front sensing module. The vehicle front computing module 100 includes first computing modules 110, each corresponding to a first sensing module 200. Each first computing module 110 works in conjunction with its corresponding first sensing module 200, connecting via a network cable to a first fusion unit 220 within that module, and receiving sensor signals packaged by the first fusion unit 220. These sensor signals contain various types of information, such as image data captured by a camera, distance and speed data detected by radar, etc. The first computing module 110 can preprocess the raw sensor information, including point cloud fusion of the sensor data. For example, it can fuse data obtained from lasers, millimeter waves, and cameras to form a point cloud containing RGB color, distance, speed, temperature, and other sensing data. The first computing module 110 can also perform deep learning-based artificial intelligence judgments on point clouds. The judgments include information such as foreign object intrusion, foreign object type, foreign object distance, and foreign object volume. Specifically, the primary judgment may use a camera (including near-focus and far-focus cameras) as the first judgment condition for foreign object recognition, with laser as an auxiliary judgment; the secondary judgment may use a lidar as the judgment condition, with a camera as an auxiliary judgment. Each first computing module 110 serves as a backup for the others, ensuring the safety and stability of the obstacle avoidance system. The first computing module 110 and the first fusion unit 220 can interact via the TCP / IP protocol.

[0033] The rear-end sensing module is installed at the rear of the vehicle to collect environmental information behind the vehicle. The second sensing module 400 is its basic component. The rear-end sensing module will include at least two second sensing modules 400, which serve as heterogeneous backups to ensure the stability of the rail transit sensing equipment and prevent a single unit's malfunction from affecting the judgment.

[0034] The second sensing module 400 includes a second fusion unit 420 and various types of second sensing units 410 connected to the second fusion unit 420. The second sensing units 410 can be different types of sensors, such as lidar sensors, camera sensors, millimeter-wave radar sensors, etc., each sensing information about the vehicle's surrounding environment from different dimensions. The second fusion unit 420 is responsible for integrating and packaging the sensor signals acquired by these different types of second sensing units 410. Different second sensing units 410 collect data based on their own physical characteristics and working principles. For example, lidar determines the distance and contour information of target objects by emitting laser beams and receiving reflected light, while cameras capture image information through optical imaging principles. The second fusion unit 420 integrates this data from different sensing units according to a certain format and rules for subsequent transmission to the vehicle head computing module 100. Specifically, the various second sensing units 410 are rationally arranged at different positions on the front of the vehicle to obtain the optimal sensing field of view. For example, cameras are installed at a suitable position above the windshield, and lidar is installed near the front bumper. The second fusion unit 420 can be a data processing chip with multiple interfaces. It is connected to each second sensing unit 410 through a dedicated data transmission line. After receiving data from each sensing unit, it packages the data according to a predetermined communication protocol and data structure, and then transmits the packaged data to the corresponding vehicle head computing module 100 through a network cable.

[0035] The rear-end computing module 300 is one of the core processing units in the vehicle-mounted multi-source heterogeneous device, located in the rear area of ​​the vehicle. It is a collection of modules specifically designed to process and compute the data acquired by the rear-end sensing module. The rear-end computing module 300 includes second computing modules 310, each corresponding to a second sensing module 400. Each second computing module 310 works in conjunction with its corresponding second sensing module 400, connecting via a network cable to the second fusion unit 420 within the corresponding second sensing module 400 to receive the sensor signals packaged by the second fusion unit 420. These sensor signals contain various types of information, such as image data captured by cameras, distance and speed data detected by radar, etc. The second computing module 310 can preprocess the raw sensor information, including point cloud fusion of the sensor data. For example, it can fuse data obtained from lasers, millimeter waves, and cameras to form a point cloud containing RGB color, distance, speed, temperature, and other sensing data. The second computing module 310 can also perform deep learning-based artificial intelligence judgments on point clouds. The judgments include information such as foreign object intrusion, foreign object type, foreign object distance, and foreign object volume. Specifically, the primary judgment may use a camera (including near-focus and far-focus cameras) as the secondary judgment condition for foreign object recognition, with laser as an auxiliary judgment; the secondary judgment may use a lidar as the judgment condition, with a camera as an auxiliary judgment. Each of the second computing modules 310 serves as a backup for the others, ensuring the safety and stability of the obstacle avoidance system. The second computing module 310 and the second fusion unit 420 can interact via the TCP / IP protocol.

[0036] In addition to packaging and transmitting data to the corresponding computing modules, the first fusion unit 220 and the second fusion unit 420 can also share data between different computing modules, enabling the sharing of sensor data between the front and rear of the vehicle. Each first fusion unit 220 is connected to all first computing modules 110 and second computing modules 310 via a data bus (the figure only shows the first fusion unit 220 in the upper left corner as an example, illustrating the connection between the first fusion unit and all first computing modules 110 and second computing modules 310 via the data bus, which is represented by the bold black line in the figure). Each second fusion unit 420 is also connected to all first computing modules 110 and second computing modules 310 via a data bus, enabling data sharing between the front and rear of the vehicle. The type of sensor data to be shared is the target category. Target category sensor data generally uses data with a lower frame rate. By sharing low-rate data, the goals of saving sensors, ensuring data heterogeneity, and improving system reliability can be achieved.

[0037] The onboard multi-source heterogeneous device in this embodiment includes a front-end computing module 100, a front-end sensing module, a rear-end computing module 300, and a rear-end sensing module. It upgrades the onboard sensing method from a single sensing method to multiple sensing methods. A fusion unit aggregates the outputs of multiple sensing devices and then outputs them uniformly to the computing unit, greatly simplifying the wiring between the sensing and computing sides. The fusion unit also shares some sensing data from the front and rear of the vehicle via a data bus, further enhancing the overall judgment capability of the entire vehicle. This improves the all-weather, all-scenario, and all-parameter sensing capabilities of the rail transit industry, ensuring the safety of rail transit operations. Furthermore, to ensure the sensing safety of the intelligent transportation industry, heterogeneous sensing and computing sides are designed, ensuring product reliability and improving product availability.

[0038] In one embodiment, the first sensing unit 210 and the second sensing unit 410 include a near-focus camera, a far-focus camera, a lidar, a millimeter-wave radar, and an inertial navigation unit. The near-focus camera and the far-focus camera are two types of cameras with different focal lengths. The near-focus camera primarily focuses on capturing detailed information about objects close to the vehicle; specifically, it can be positioned near the edge of the front of the vehicle to obtain the optimal near-field field of view. The far-focus camera observes scenes at a greater distance from the vehicle; specifically, it can be mounted high on the front of the vehicle, such as in a central position above the windshield, ensuring its field of view is not significantly obstructed and providing a wide-angle view. The lidar detects the distance, orientation, height, and speed of a target object by emitting a laser beam and receiving the laser signal reflected back from the target object. Lidar is typically mounted on the top of the front of the vehicle to ensure its scanning range covers most of the area in front of the vehicle. Millimeter-wave radar is a radar technology that utilizes the millimeter-wave frequency band. Compared to traditional radar, millimeter-wave radar has a higher frequency and a shorter wavelength, which allows it to provide higher resolution and more accurate measurement data. Specifically, a 4D millimeter-wave radar can be selected as the millimeter-wave radar, which is an extension of the traditional millimeter-wave radar. In terms of terminology, in addition to detecting the three-dimensional information of a target object (distance, velocity, and angle), it also adds the ability to detect the height information of the target object, thus outputting a point cloud with height information. An inertial navigation unit (INS) is a device that uses inertial sensors (such as accelerometers and gyroscopes) to measure the vehicle's acceleration, angular velocity, and other motion parameters, and calculates the vehicle's attitude, position, and velocity based on these parameters. When the vehicle enters environments with poor GPS signals, such as tunnels, the INS can continuously provide relatively accurate position and attitude information based on its own measurement data, ensuring the stability and safety of the vehicle's operation. In one embodiment, the millimeter-wave radar is a 4D millimeter-wave radar.

[0039] In one embodiment, the data bus includes a CAN bus.

[0040] In one embodiment, please refer to Figure 2 The vehicle head computing module 100 also includes a first decision module 120, which is connected to each of the first computing modules 110. The first computing module 110 is used to preprocess the received sensor signals and output them to the first decision module 120. The first decision module 120 is used to make control decisions based on the preprocessed sensor signals.

[0041] It is understandable that the first decision module 120 is a key component of the vehicle front computing module 100. Located at the back end of the entire data processing chain, it is responsible for receiving pre-processed sensor signals from each first computing module 110 and making corresponding control decisions based on the information contained in these signals. The sensor signals received by each first computing module 110 come from diverse sources. After preprocessing by the first computing module 110, such as performing data point cloud fusion and preliminary judgments on information such as foreign object intrusion, foreign object type, foreign object distance, and foreign object size, these signals, carrying crucial information about the environment in front of the vehicle, are transmitted to the first decision module 120. The first decision module 120 internally runs specific decision algorithms, which are often constructed based on a large amount of experimental data, traffic rules, and vehicle control theory. The first decision module 120, according to pre-set logic, weights and judges the results output by multiple first computing modules 110 to determine whether the current situation is a dangerous scenario requiring emergency braking or evasive action, and then makes corresponding control decisions.

[0042] In one embodiment, the vehicle-mounted multi-source heterogeneous device also includes a front-end warning device, which is used to issue an alarm signal under the control of the first decision module 120. The front-end warning device is a device capable of issuing intuitive and conspicuous alarm signals based on the control decisions made by the first decision module 120, aiming to alert the driver and other occupants of potential hazards. For example, if the first decision module 120 determines that an obstacle suddenly appears in front of the vehicle and the distance is within a dangerous range, it will send a command to the front-end warning device, which will then activate alarm measures such as flashing lights and emitting a rapid warning sound.

[0043] In one embodiment, the rear-end computing module 300 further includes a second decision module 320, which is connected to each of the second computing modules 310. The second computing module 310 is used to preprocess the received sensor signals and output them to the second decision module 320. The second decision module 320 is used to make control decisions based on the preprocessed sensor signals.

[0044] It is understandable that the second decision module 320 is a key component of the rear-end computing module 300. Located at the back end of the entire data processing chain, it is responsible for receiving pre-processed sensor signals from each second computing module 310 and making corresponding control decisions based on the information contained in these signals. The sensor signals received by each second computing module 310 come from diverse sources. After preprocessing by the second computing module 310, such as performing data point cloud fusion and preliminary judgments on information such as foreign object intrusion, foreign object type, foreign object distance, and foreign object size, these signals, carrying crucial information about the environment in front of the vehicle, are transmitted to the second decision module 320. The second decision module 320 internally runs specific decision algorithms, which are often constructed based on a large amount of experimental data, traffic rules, and vehicle control theory. The second decision module 320, according to pre-set logic, weights and judges the results output by multiple second computing modules 310 to determine whether the current situation is a dangerous scenario requiring emergency braking or evasive action, and then makes corresponding control decisions.

[0045] In one embodiment, the in-vehicle multi-source heterogeneous device also includes a rear-end alarm device, which issues an alarm signal under the control of the second decision module 320. The rear-end alarm device is a device that sends corresponding alarm signals to the outside world based on the control decisions made by the second decision module 320. Its main purpose is to promptly alert occupants to various potential hazards appearing at the rear of the vehicle, avoiding possible collisions and other safety accidents. Examples include images of close-range objects captured by a rear-end close-focus camera, distance and orientation information of objects detected by a rear-end lidar, and speed and relative position data of objects fed back by a rear-end millimeter-wave radar. When the second decision module 320 determines, based on these sensor signals, that there is an obstacle at the rear of the vehicle requiring a warning while reversing, it sends corresponding control commands to the rear-end alarm device.

[0046] In one embodiment, the first fusion unit 220 and the second fusion unit 420 include FPGA units. An FPGA unit is a programmable logic device consisting of numerous programmable logic blocks, lookup tables (LUTs), registers, and programmable interconnect resources. In the first fusion unit 220 and the second fusion unit 420, the FPGA unit plays a core data fusion role. Since the first sensing unit 210 and the second sensing unit 410 contain various types of sensors, such as close-range cameras, telephoto cameras, LiDAR, millimeter-wave radar, and inertial navigation units, the data output by these sensors differs in format, rate, timing, etc. The FPGA unit can adapt to this diversity with its high flexibility.

[0047] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0048] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A vehicle-mounted multi-source heterogeneous device, characterized in that, It includes a front-end computing module, a front-end perception module, a rear-end computing module, and a rear-end perception module; The vehicle front perception module includes at least two first perception modules, and each first perception module includes a first fusion unit and multiple first perception units connected to the first fusion unit. The vehicle front computing module includes a first computing module that corresponds to the first sensing module. The first computing module is connected to the corresponding first fusion unit via a network cable. The first fusion unit is used to package all received sensing signals and transmit them to the corresponding first computing module. The rear-end perception module includes at least two second perception modules, each of which includes a second fusion unit and multiple second perception units connected to the second fusion unit. The rear computing module includes a second computing module that corresponds to the second sensing module. The second computing module is connected to the corresponding second fusion unit via a network cable. The second fusion unit is used to package all received sensing signals and transmit them to the corresponding second computing module. The first fusion unit and the second fusion unit are also connected to each of the first computing modules and the second computing modules via a data bus, respectively, for transmitting the target type sensing signal to each of the first computing modules and the second computing modules.

2. The vehicle-mounted multi-source heterogeneous device according to claim 1, characterized in that, The first sensing unit includes a near-focus camera, a far-focus camera, a lidar, a millimeter-wave radar, and an inertial navigation unit.

3. The vehicle-mounted multi-source heterogeneous device according to claim 2, characterized in that, The millimeter-wave radar is a 4D millimeter-wave radar.

4. The vehicle-mounted multi-source heterogeneous device according to claim 1, characterized in that, The second sensing unit includes a near-focus camera, a far-focus camera, a lidar, a millimeter-wave radar, and an inertial navigation unit.

5. The vehicle-mounted multi-source heterogeneous device according to claim 1, characterized in that, The data bus includes the CAN bus.

6. The vehicle-mounted multi-source heterogeneous device according to claim 1, characterized in that, The vehicle head computing module further includes a first decision module, which is connected to each of the first computing modules. The first computing module is used to preprocess the received sensor signals and output them to the first decision module. The first decision module is used to make control decisions based on the preprocessed sensor signals.

7. The vehicle-mounted multi-source heterogeneous device according to claim 6, characterized in that, It also includes a front alarm device, which is used to issue an alarm signal under the control of the first decision module.

8. The vehicle-mounted multi-source heterogeneous device according to claim 1, characterized in that, The rear-end computing module further includes a second decision module, which is connected to each of the second computing modules. The second computing module is used to preprocess the received sensor signals and output them to the second decision module. The second decision module is used to make control decisions based on the preprocessed sensor signals.

9. The vehicle-mounted multi-source heterogeneous device according to claim 8, characterized in that, It also includes a rear-end alarm device, which is used to issue an alarm signal under the control of the second decision module.

10. The vehicle-mounted multi-source heterogeneous device according to claim 1, characterized in that, The first fusion unit and the second fusion unit include FPGA units.

Images