Vehicle central high-performance computing control unit and vehicle system

Abstract

Provided are a vehicle central high-performance computing control unit (10) and a vehicle system. The vehicle central high-performance computing control unit (10) is integrated with an autonomous driving domain control module (12), a central vehicle domain control module (11), an intelligent cockpit domain control module (13) and a data interaction module (14), wherein the autonomous driving domain control module (12), the central vehicle domain control module (11) and the intelligent cockpit domain control module (13) are connected to each other and are all connected to the data interaction module (14). Therefore, data sharing is realized between different domains, and data sharing and transmission between the domains are better coordinated.

Description

Vehicle central supercomputing control unit and vehicle system

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of Chinese Patent Application No. 202311429520.X filed on October 31, 2023, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present application relates to the field of automobile control technology, and in particular to a vehicle central supercomputing control unit and a vehicle system. Background Art

[0004] As the automotive industry develops from internal combustion engines and mechanization to electrification and intelligence, automotive electronic technology is gradually applied to automobiles. Early distributed ECUs (Electronic Control Units) have gradually replaced traditional mechanical control and hydraulic control systems due to their low computing power, high reliability, and good safety. Since then, with the emergence of intelligent driving needs, high-computing-power domain controllers have gradually become mainstream. However, with the advancement of technology and the increase in user demand, the application of electronic control technology in automobiles has become more and more extensive. In order to solve the problem of distributed computing power, the chip industry in the automotive industry has developed rapidly. With the increase in demand for self-driving cars, the amount of data they need to process will become larger, and the computing power requirements for chips will continue to increase.

[0005] A typical milestone is the classic five-domain architecture, which functionally divides the entire vehicle into five control systems (body control system, power control system, chassis control system, intelligent cockpit control system, and autonomous driving control system). Each system is controlled by its own domain controller. However, with the advancement of domain integration, computing power and control are further centralized, and the five domains have evolved into three domains (vehicle control system, intelligent cockpit control system, and autonomous driving control system). The development of domain controllers is highly dependent on core processor chips. For example, the body domain controller, chassis domain controller, and power domain controller rely on core processor chips with better real-time performance and higher functional safety levels, while the autonomous driving domain controller relies on SOC (System on Chip) chips with higher computing power and convolutional neural network computing. The intelligent cockpit domain controller relies on SOC chips with high computing power graphics accelerators, large-capacity memory processing, and support for complex operating systems for user interaction.

[0006] For chips, there are still technical engineering difficulties in integrating the above-mentioned features into a SOC. Even though multi-core heterogeneous SOCs already exist, the computing power is still difficult to meet the large computing power requirements of autonomous driving and smart cockpits. Therefore, using only a single core chip to meet the needs of HPC (High Performance Computing) has become a current research difficulty.

[0007] Application Contents

[0008] The purpose of this application is to provide a vehicle central supercomputing control unit and a vehicle system to achieve product-level integration through domain fusion to realize a more integrated computing platform.

[0009] In order to solve the above technical problems, the present application provides a vehicle central supercomputing control unit and a vehicle system.

[0010] In a first aspect, the present application provides a vehicle central supercomputing control unit, which integrates an autonomous driving domain control module, a central vehicle domain control module, an intelligent cockpit domain control module, and a data interaction module. The autonomous driving domain control module, the central vehicle domain control module, and the intelligent cockpit domain control module are interconnected and connected to the data interaction module.

[0011] The autonomous driving domain control module is configured to generate autonomous driving control information using the acquired target position information and information acquired by multiple driving sensing devices configured on the vehicle;

[0012] The central vehicle domain control module is used to control the vehicle body according to the automatic driving control information sent by the automatic driving domain control module;

[0013] The intelligent cockpit domain control module is used to process and display the acquired vehicle information; the vehicle information includes vehicle surrounding environment information, vehicle occupant behavior monitoring information, vehicle driving information and multimedia information;

[0014] The data interaction module is used to control the internal and external data interaction of the autonomous driving domain control module, the central vehicle domain control module and the intelligent cockpit domain control module.

[0015] In a further embodiment, the autonomous driving domain control module includes two autonomous driving domain master control modules, the two autonomous driving domain master control modules being a first autonomous driving domain master control module and a second autonomous driving domain master control module;

[0016] The first autonomous driving domain master control module is configured to use multiple driving perception devices configured on the vehicle to obtain environmental information around the vehicle, and obtain target position information generated by the second autonomous driving domain master control module, and use a deep learning neural network model to process the environmental information around the vehicle and the target position information, perceive the vehicle's surrounding environment, and plan local or global driving paths.

[0017] In a further embodiment, the second autonomous driving domain master control module is configured to generate automatic driving control information based on the local or global driving path and current road condition information sent by the first autonomous driving domain master control module; wherein the automatic driving control information includes vehicle power control information and vehicle-mounted inertial navigation positioning information;

[0018] Furthermore, the received ultrasonic target detection signals, millimeter wave target detection signals and laser target detection signals are filtered and classified to obtain target position information.

[0019] In a further embodiment, the central vehicle domain control module includes a central vehicle domain control master module;

[0020] The central vehicle domain control main control module is used to perform body control and vehicle power control based on the received automatic driving control information, and generate vehicle opening and closing parts status control information based on the acquired vehicle opening and closing parts status information and the pre-set vehicle opening and closing parts control strategy; and control the mutual communication between different CAN bus interfaces and Ethernet interfaces configured on the vehicle.

[0021] In a further embodiment, the central vehicle domain control master module is further used to control the power circuit relay and generate a power distribution management signal based on a preset battery management strategy and vehicle power demand;

[0022] And, managing and controlling the vehicle air conditioning system according to the pre-acquired temperature control demand and the automatic air conditioning control strategy, wherein the temperature control demand is generated by the pre-set thermal management strategy.

[0023] In a further embodiment, the smart cockpit domain control module is connected to a plurality of driving perception devices configured on the vehicle via a low voltage differential signal video transmission line, and the smart cockpit domain control module includes a smart cockpit domain master control module;

[0024] The smart cockpit domain main control module is used to use the multiple driving perception devices to obtain the surrounding video images of the vehicle and the vehicle personnel behavior monitoring information, compress and encode the surrounding video images of the vehicle and display them, use the deep learning neural network model to identify the vehicle personnel behavior monitoring information, and obtain vehicle personnel behavior identification information; and monitor the working status of the multiple driving perception devices connected to the smart cockpit domain control module.

[0025] In a further embodiment, the intelligent cockpit domain control module is connected to an in-vehicle entertainment human-computer interaction system configured on the vehicle;

[0026] The smart cockpit domain main control module is also used to compress and encode the acquired vehicle power information and store it, and output the compressed and encoded vehicle power information to the in-vehicle entertainment human-computer interaction system.

[0027] In a further embodiment, the data interaction module includes an internal connection unit, a vehicle external connection unit, a primary power supply unit, and an interface protection unit;

[0028] The internal connection unit is used to establish internal communication connections with the autonomous driving domain control module, the central vehicle domain control module, and the smart cockpit domain control module;

[0029] The vehicle external connection unit is used to control the communication between the vehicle external device and the autonomous driving domain control module, the central vehicle domain control module and the smart cockpit domain control module;

[0030] The primary power supply unit is used to control the vehicle battery supply voltage;

[0031] The interface protection unit is used for power supply reverse connection protection and overvoltage surge protection.

[0032] In the second aspect, the present application provides a vehicle system, which includes: a vehicle body, a vehicle central supercomputing control unit as described above, which is arranged in the vehicle body, and several regional controllers arranged around the vehicle body, and the several regional controllers are connected to the vehicle central supercomputing control unit through an Ethernet interface in a ring network manner.

[0033] In a further implementation scheme, the regional controller is used to detect the status of vehicle opening and closing parts and the power distribution status, and after receiving the vehicle control signal from the vehicle central supercomputing control unit, it controls the driving sensing device and power circuit relay configured on the vehicle body according to the vehicle control signal.

[0034] The present application provides a vehicle central supercomputing control unit and a vehicle system, wherein the vehicle central supercomputing control unit integrates an autonomous driving domain control module, a central vehicle domain control module, an intelligent cockpit domain control module, and a data interaction module; the autonomous driving domain control module generates automatic driving control information using multiple driving sensing devices configured on the vehicle and the acquired target position information; the central vehicle domain control module controls the vehicle body according to the automatic driving control information; the intelligent cockpit domain control module is used to process and display the acquired vehicle information; and the data interaction module is used to control the internal and external data interaction of the autonomous driving domain control module, the central vehicle domain control module, and the intelligent cockpit domain control module. The vehicle central supercomputing control unit forms a highly integrated control architecture by integrating the central vehicle domain control module, the autonomous driving domain control module, and the intelligent cockpit domain control module, which can provide more accurate and faster vehicle control, improve driving safety and driving experience, and the highly integrated vehicle central supercomputing control unit has high-performance computing technology and more powerful processing capabilities, realizing more advanced driving assistance and autonomous driving functions, improving driving safety and driving experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] FIG1 is a block diagram of a vehicle central supercomputing control unit provided in an embodiment of the present application;

[0036] FIG2 is a schematic diagram of the upper and lower stacking of the vehicle central supercomputing control unit provided in an embodiment of the present application;

[0037] FIG3 is a functional block diagram of an autonomous driving domain control module provided in an embodiment of the present application;

[0038] FIG4 is a functional block diagram of a central vehicle domain control module provided in an embodiment of the present application;

[0039] FIG5 is a functional block diagram of a smart cockpit domain control module provided in an embodiment of the present application;

[0040] FIG6 is a schematic diagram of the structure of a data interaction module provided in an embodiment of the present application;

[0041] FIG7 is a block diagram of a vehicle system provided by an embodiment of the present application;

[0042] FIG8 is a schematic diagram of the electrical architecture of the entire vehicle provided in an embodiment of the present application. DETAILED DESCRIPTION

[0043] The following describes the implementation methods of the present application in detail with reference to the accompanying drawings. The examples are provided for illustrative purposes only and should not be construed as limiting the present application. The accompanying drawings are provided for reference and illustration purposes only and do not constitute a limitation on the scope of patent protection of the present application, because many changes can be made to the present application without departing from the spirit and scope of the present application.

[0044] Referring to Figure 1, an embodiment of the present application provides a vehicle central supercomputing control unit. As shown in Figure 1, the vehicle central supercomputing control unit 10 integrates a central vehicle domain control module 11 (VDC, Vehicle Domain Controller), an autonomous driving domain control module 12 (ADC, ADAS\AD Domain Controller), an intelligent cockpit domain control module 13 (CDC, Cockpit Domain Controller) and a data interaction module 14, wherein the central vehicle domain control module VDC, the autonomous driving domain control module ADC, and the intelligent cockpit domain control module CDC are all connected to the data interaction module to realize external data interaction through the data interaction module; the autonomous driving domain control module ADC, the intelligent cockpit domain control module CDC and the central vehicle domain control module VDC establish a communication connection through a high-speed serial interface PCIE, and its communication bandwidth is more than 3 times that of 10G Ethernet. This embodiment adopts the PCIE interface to make its bandwidth higher and the communication cost lower, thereby improving the communication efficiency between various components of the vehicle and reducing information transmission delay.

[0045] In this embodiment, the central vehicle domain control module 11, the autonomous driving domain control module 12, and the intelligent cockpit domain control module 13 use Hirose's high-speed floating connector for communication, ensuring the reliability of the connection in a vibration environment. In order to minimize the vehicle central supercomputer (HPC, High Performance 2, this embodiment integrates the central vehicle domain control module 11, the autonomous driving domain control module 12, and the intelligent cockpit domain control module 13 in an up-and-down stacking design, which can achieve a higher degree of integration to minimize the space occupied by the device. In FIG2, CAN is a CAN bus interface, LIN is a LIN bus interface, and ETH is an Ethernet interface. This embodiment improves the integration and reliability of the device by adopting this up-and-down stacking design, while further reducing the volume of the device. The projected area of ​​the device is effectively controlled, making the device more suitable for installation in a narrow car space. At the same time, this embodiment installs the data interaction module vertically, and all external interfaces are arranged on the data interaction module. The data interaction module design adopted in this embodiment can not only ensure that the external interface is located on one side of the device, taking into account both installation convenience and aesthetics, but also reduce the stress damage of the high-precision core board caused by the plugging and unplugging of cables, while reducing the external electrostatic interference directly entering the core board, thereby improving the anti-static interference performance of the device.

[0046] This embodiment adopts an up-and-down stacking design for the central vehicle domain control module 11, the autonomous driving domain control module 12, and the intelligent cockpit domain control module 13, which can integrate multiple domain control modules (central vehicle domain control module, autonomous driving domain control module, and intelligent cockpit domain control module) together, and arrange the domain control modules and computing units more compactly, thereby achieving higher functional integration in a limited space and improving space utilization. At the same time, compared with the traditional horizontal arrangement design, this embodiment integrates multiple domain control modules in an up-and-down stacking manner, which not only makes the heat dissipation more uniform and reduces the volume and cost of the heat dissipation equipment, but also makes the communication distance between each domain control module shorter, thereby reducing communication delay and improving communication efficiency, which helps to improve the safety and performance of the vehicle.

[0047] This embodiment integrates the central vehicle domain control module 11, the autonomous driving domain control module 12, the intelligent cockpit domain control module 13, and the data interaction module 14 to form a vehicle central supercomputing control unit, realizing a highly integrated computing platform. Structurally, this embodiment merges the original three domain control modules into a single vehicle central supercomputing control unit. This not only reduces the number of components on the structure and significantly reduces the manufacturing cost of the vehicle, but also the design of the central supercomputing unit proposed in this embodiment only requires a separate HPC upgrade to upgrade the ADC, VDC, and CDC applications, making upgrades of ADC, VDC, and CDC applications much simpler and eliminating the need to develop a complex OTA Master protocol, thus simplifying the upgrade process. Furthermore, through the introduction of the vehicle central supercomputing control unit, the functions originally required for the three domain control modules are integrated into a single vehicle central supercomputing control unit, making the vehicle's computing power more concentrated, which is beneficial for improving the vehicle's performance and response speed. Furthermore, the design of the vehicle central supercomputing control unit reduces the number of components within the vehicle, making the product more compact and easier to produce and install.

[0048] In this embodiment, the autonomous driving domain control module ADC includes at least two interconnected autonomous driving domain master control modules. The autonomous driving domain control module ADC is configured to generate automatic driving control information using multiple driving perception devices configured on the vehicle and acquired target position information. The multiple driving perception devices 40 include, but are not limited to, the lidar, front millimeter-wave radar, corner radar, ultrasonic radar, and the like as shown in FIG3 . For ease of illustration, this embodiment is described in detail using the example of an autonomous driving domain control module ADC comprising two interconnected autonomous driving domain master control modules, namely, a first autonomous driving domain master control module 121 and a second autonomous driving domain master control module 122. The first autonomous driving domain master control module 121 is configured to utilize the multiple driving perception devices 40 configured on the vehicle to acquire information about the vehicle's surrounding environment and target position information generated by the second autonomous driving domain master control module 122. The first autonomous driving domain master control module 121 processes the vehicle's surrounding environment information and target position information using a deep learning neural network model to perceive the vehicle's surrounding environment and plan a local or global driving path.

[0049] The second autonomous driving domain main control module 122 is used to generate automatic driving control information based on the local or global driving path and current road condition information sent by the first autonomous driving domain main control module 121, and filter and classify the received ultrasonic target detection signals, millimeter wave target detection signals and laser target detection signals to obtain corresponding target position information, wherein the automatic driving control information includes vehicle power control information and vehicle-mounted inertial navigation positioning information.

[0050] In a specific embodiment, the autonomous driving domain control module ADC is equipped with a Horizon J5 chip (Journey5, an automotive-grade edge computing chip) and an NXP S32G chip that are interconnected. As shown in Figure 3, the autonomous driving domain control module ADC is mainly responsible for implementing the autonomous driving domain control function. The autonomous driving domain control module ADC has the following external interfaces:

[0051] Two LVDS interfaces for connecting two ADAS cameras respectively. The two ADAS cameras are a wide-angle ADAS camera and a telephoto ADAS camera. The specification of the wide-angle ADAS camera is 8MP@120HFOV (resolution 8MP, field of view 120 degrees), and the specification of the telephoto ADAS camera is 8MP@30HFOV (resolution 8MP, field of view 30 degrees); the LVDS interface for connecting the rearview camera is 2MP@60HFOV (resolution 2MP resolution, 60-degree field of view); four LVDS interfaces for connecting four panoramic cameras, all with the same specification of 2MP@120HFOV (resolution of 2MP, 120-degree field of view); four LVDS interfaces for connecting four fisheye cameras, all with the same specification of 2MP@195HFOV (resolution of 2MP, 195-degree field of view); LVDS output interface, specification of 2MP@30FPS (resolution of 2MP, 30FPS frame rate); four 1000M Ethernet interface, specification is 1000BASE-T1, one Ethernet interface is reserved for SOA service, and the other three Ethernet interfaces are used to connect to the lidar; six CAN interfaces, specification is CAN-FD, one CAN interface is connected to the forward millimeter-wave radar, four CAN interfaces are used to connect to the four corner radars, and the last CAN interface is used to connect to the ultrasonic radar controller; one PCIE interface, the PCIE interface is responsible for communication with the VDC; it should be noted that in Figure 3, PMIC is a power management integrated chip, EMMC is an embedded multimedia memory (non-volatile memory), FLASH is a flash memory (non-volatile memory), LPDDR4 is a low-power fourth-generation dynamic random access memory (volatile memory), Switch is an Ethernet switch, HFOV is a horizontal field of view, LRR is a long-range millimeter-wave radar, MRR is a medium-range millimeter-wave radar, USS is an ultrasonic radar, ADR is an on-board inertial navigation module, AVM is a panoramic image monitoring system, SOA (Service-Oriented Architecture) is a service-oriented architecture (here refers to the services under the architecture), and DVR is a digital video surveillance system.

[0052] In this embodiment, the Horizon J5 chip mainly deploys the perception functions of two ADAS cameras, the perception of the surround-view camera, the perception of the rear-view camera, and the perception of the surround-view camera. In addition, the Horizon J5 chip is also responsible for parking fusion algorithms, parking mapping and positioning functions, parking planning, driving fusion algorithms, driving mapping and positioning, driving planning, DVR driving records and diagnosis functions; while the NXP S32G chip is mainly responsible for driving control, parking control, ADR / inertial navigation, OTA, ultrasonic radar acquisition and algorithm, lidar acquisition and algorithm, millimeter-wave radar acquisition and algorithm, CAN communication, Ethernet communication, and diagnosis functions. For ease of understanding, the functions of the Horizon J5 chip and NXP's S32G chip are described in detail below. Among them, the functions of the Horizon J5 chip are specifically as follows:

[0053] Regarding the perception functions of the ADAS cameras, surround-view cameras, rearview cameras, and surround-view cameras, this embodiment uses the LVDS interface to acquire digital video signals from the two ADAS cameras in front of the vehicle, the four surround-view cameras installed around the vehicle, the rearview camera installed at the rear of the vehicle, and the surround-view cameras installed around the vehicle. These digital signals are decoded and restored, and then output to the NPU processor in the Horizon J5 chip. The NPU processor uses a deep learning neural network algorithm to identify objects in front of the vehicle, such as motor vehicles, pedestrians, lane markings, stop or speed limit signs, traffic lights, road guardrails, green belts, and parking spaces. This embodiment primarily identifies parking space targets based on the digital video signals captured by the rearview and surround-view cameras. After identifying various objects around the vehicle, this embodiment uses the S32G chip to acquire target position information detected by the ultrasonic radar. This information is then combined with the digital video signals from the surround-view and surround-view cameras and processed using a parking fusion algorithm to derive target information containing characteristic information such as target type and location.

[0054] The parking mapping, positioning, and parking path planning functions of the Horizon J5 chip are primarily used during the parking process. Their mission is to provide accurate positioning information for parking operations by perceiving and modeling the vehicle's surroundings, ensuring smooth and safe parking. This embodiment utilizes sensors such as lidar and cameras to collect characteristic information about the surrounding environment. While the vehicle is in motion, the vehicle's motion trajectory is inferred from the captured characteristic information about the vehicle's surroundings, thereby creating a high-precision map for use in parking scenarios. This embodiment performs a repositioning operation during the parking process, matching the characteristic information about the surrounding environment at the time of the vehicle's stop with the information in the map to determine the vehicle's actual position in the map. This helps the vehicle determine its own position more accurately, enabling more accurate path planning and operation during parking. Parking path planning is then performed. Specifically, after the user selects the desired target parking space information through the human-computer interaction interface, the target parking space information is imported into the target parking point in the high-precision map. A reasonable parking path is planned by combining the current vehicle positioning information and the target parking space information.

[0055] The driving mapping, positioning and driving planning functions of the Horizon J5 chip are mainly used in the driving process of the vehicle. Its mission goal is to provide accurate positioning information for the vehicle's driving by perceiving and modeling the vehicle's surrounding environment to ensure driving safety and accuracy. In this process, this embodiment obtains target detection information from sensors such as ultrasonic waves, radars, millimeter-wave radars and lidars through the S32G chip. At the same time, it combines the video digital signals of the forward ADAS camera, the surround-view camera and the surround-view camera, and uses the driving fusion algorithm to perform data fusion processing to obtain the surrounding driving environment target information containing characteristic information such as target type and position. When the vehicle is in motion, the system will capture the surrounding driving environment target information and The motion trajectory is calculated, and then a high-precision map of the vehicle's driving path is constructed. The embodiment performs a repositioning operation during driving, matches the surrounding environment feature information when the vehicle is driving with the information in the map, and then calculates the actual position of the vehicle in the map, thereby realizing the driving mapping and positioning function. After the user selects the target location in the navigation system through the human-computer interaction interface, the system will construct a global path planning. During driving, the system performs reasonable local planning of the driving path based on the information source of the high-precision map and the current road conditions (including pedestrians, lanes, motor vehicles, stop signs / speed limit signs, traffic lights, zebra crossings, etc., which are detected and identified through a fusion algorithm). At this point, the driving planning is completed.

[0056] At the same time, this embodiment also uses the Horizon J5 chip to encode and process the videos of the ADAS camera, surround-view camera, panoramic camera and rearview camera, and then compress and store them, realizing the DVR driving recording function of the driving process. At the same time, the Horizon J5 chip will also check the working status of the camera and diagnose whether there are any abnormal conditions, realizing diagnostic functions, such as: diagnosing abnormal conditions such as damage or occlusion.

[0057] In this embodiment, the functions of NXP's S32G chip are specifically as follows: during the driving process of the vehicle, the driving control system in the S32G chip outputs vehicle power information based on the planned driving route and real-time road condition information; during the parking process of the vehicle, the parking control system in the S32G chip outputs vehicle power information based on the planned parking route and real-time road condition information, where the vehicle power information includes gear, torque, steering and braking information; when the vehicle enters an area with weak or missing GPS signals, the ADR / inertial navigation module in the S32G chip calculates the real-time positioning information of the subsequent vehicle in the map based on information such as the vehicle's current driving speed, acceleration and steering angle.

[0058] At the same time, the S32G chip receives the target detection signal of the ultrasonic radar sensor through the CAN interface, and filters, classifies and processes the signal to obtain stable target information (including target and distance), realizing the ultrasonic radar acquisition and calculation process. At the same time, the S32G chip receives the target detection signal of the millimeter wave radar sensor through the CAN interface, and filters, tracks and clusters the signal to obtain stable target information (including target quantity, category and distance). The S32G chip receives the target detection signal of the lidar sensor through the Ethernet interface, and filters, classifies and processes the signal to obtain stable target information (including all point cloud information). In this embodiment, the S32G chip has basic CAN interface and Ethernet interface communication functions, and can complete remote air upgrades (OTA, Over-the-Air Technology), including self-upgrade and OTA Master functions. In addition, the S32G chip can detect the working status of the external radar sensor and diagnose whether there is any abnormality.

[0059] In this embodiment, the central vehicle domain control module VDC is used to control the vehicle body according to the automatic driving control information sent by the automatic driving domain control module.

[0060] The central vehicle domain control module includes a central vehicle domain control main control module 111, which is used to perform body control and vehicle power control based on the received automatic driving control information, and generate vehicle opening and closing parts status control information based on the acquired vehicle opening and closing parts status information and the pre-set vehicle opening and closing parts control strategy; and control the mutual communication between different CAN bus interfaces and Ethernet interfaces configured on the vehicle.

[0061] The central vehicle domain control main control module 111 is also used to control the power circuit relay and generate a power distribution management signal based on a preset battery management strategy and vehicle power demand; and to manage and control the vehicle air-conditioning system according to pre-acquired temperature control requirements and automatic air-conditioning control strategies, where the temperature control requirements are generated by a preset thermal management strategy.

[0062] In a specific embodiment, the central vehicle domain control module VDC is mainly responsible for vehicle control, central gateway and SOA vehicle control services. As shown in Figure 4, the central vehicle domain control module VDC has the following external interfaces:

[0063] Four high-side drivers (HS Drive) and eight low-side drivers (LS Drive) for driving external devices, and with reserved interfaces for accessing relay control signals; five analog inputs and ten digital inputs for receiving external analog and digital signals, and reserved interfaces for accessing sensor signals such as the accelerator pedal and brake pedal to facilitate subsequent expansion applications; 12 CAN FD interfaces, which are respectively connected to the body domain, power domain, chassis domain, cockpit domain, intelligent driving domain, TBOX (Telematics Box, Internet of Vehicles system), OBD (On-Board Diagnostics, on-board automatic diagnostic system), calibration and other circuits; four LIN interfaces, which are respectively connected to the vehicle's small battery, compressor, radar and other equipment; Ethernet interface: five channels are provided for external communication with the ZCU (Zone Control Unit) through the switch circuit. Unit), five 1000BASE-T1 interfaces for connecting to TBOX, cockpit controller, self-driving controller, etc., and one 100BASE-TX interface for communicating with the OBD interface; two PCIE interfaces, communicating with the ADC core board and the CDC core board respectively. It should be noted that in Figure 4, SBC is the system basis chip, EEPROM is the electrically erasable programmable read-only memory, HS Drive is the high-side drive, LS Drive is the bottom-side drive, CAN PHY is the PHY chip of the CAN interface, LIN PHY is the PHY chip of the LIN interface, 100BASE-T1 is a 100Mbps vehicle Ethernet interface, 1000BASE-T1 is a 1000Mbps vehicle Ethernet interface, 100BASE-TX is an industrial Ethernet interface with a rate of 100Mbps, and OBD is a vehicle diagnostic interface.

[0064] The central vehicle domain control master module includes NXP's S32G274 chip. The main functions deployed by the central vehicle domain control master module include body control, vehicle power control, battery management, power distribution management, automatic air conditioning, network routing, network management, gateway configuration, information security, functional safety, OTA Master, diagnosis, and SOA. The functions of the central vehicle domain control master module are as follows:

[0065] The central vehicle domain control master module mainly realizes the body control function by controlling the body parts such as lights, door locks, windows, rearview mirrors, wipers, and tailgate. Its control logic is as follows: the regional controller ZCU detects the control switch status of the vehicle and the status of the body parts, and sends these detection results to the central vehicle domain control master module through the CAN network. The central vehicle domain control master module will output control signals for lights, door locks and other components to the regional controller according to the current vehicle status, the detected control switch status and the status of the body parts according to the predetermined control strategy. After receiving these control signals, the regional controller will drive the corresponding body parts to respond.

[0066] The central vehicle domain control main control module mainly realizes the power control function by controlling the vehicle power system's power on and off, torque, driving mode, gear, energy and thermal management. Its control logic is as follows: the central vehicle domain control main control module obtains the gear signal, acceleration signal and deceleration signal respectively collected by the gear sensor, accelerator pedal and brake pedal through the chassis domain CAN network, and outputs the torque control signal according to the preset control strategy in combination with the working status of the vehicle, and then controls the motor controller to generate the corresponding torque. At the same time, according to the power demand of the vehicle, combined with the motor controller status, on-board charger status, DCDC (Direct Current-Direct Current Converter) status and battery management system status obtained by CAN signal, the vehicle's high-voltage components are driven and controlled according to the preset control strategy.

[0067] In addition, the central vehicle domain control main control module monitors and manages the charging and discharging process of the vehicle, and realizes the battery management function by controlling the power circuit relay according to the battery management strategy. At the same time, the central vehicle domain control main control module also undertakes the power distribution management task of controlling the power distribution for all electrical equipment in the vehicle. The control logic of the power distribution management is as follows: the regional controller ZCU drives and controls the relays of the power supply equipment and detects the status of the power supply equipment. The central vehicle domain control main control module is connected to the regional controller ZCU through the CAN network, and outputs the power distribution management signal of all electrical equipment according to the power demand of the whole vehicle. After receiving the power distribution management signal, the regional controller controls the power distribution or power-off of the corresponding electrical equipment.

[0068] The central vehicle domain control main control module realizes the automatic air-conditioning control function by managing and controlling the air-conditioning system. The specific control logic is: the central vehicle domain control main control module obtains the passenger compartment temperature and the battery pack temperature through the CAN network, and then outputs the temperature control requirements according to the preset thermal management strategy. According to the temperature control requirements and the preset automatic air-conditioning control strategy, the network control signals of the compressor, blower and other equipment are output to the regional controller ZCU. After receiving these network control signals, the regional controller ZCU will drive the compressor and blower to work, thereby realizing the control of the air-conditioning system.

[0069] One of the central gateway functions of the central vehicle domain control main control module is network routing, which is responsible for achieving intercommunication between different networks. The main task of network routing is to achieve communication between different CAN buses on the vehicle, such as: body domain CAN, power domain CAN, chassis domain CAN, cockpit domain CAN, intelligent driving domain CAN and TBOX CAN. In addition, network routing also needs to complete communication between Ethernet networks; in addition, another important function of the central gateway is port isolation, firewall and other network security protection functions, which block external malicious attacks in accordance with network security standards, thereby ensuring the security of the vehicle network.

[0070] During the OTA upgrade process, the server sends upgrade packages for components other than the vehicle's HPC to the base station via the wireless network (4G / 5G). The base station transfers the upgrade packages to the on-board TBOX. The on-board TBOX sends the upgrade packages to the central vehicle domain control module of the HPC via Ethernet or CAN network. The central vehicle domain control module sends the upgrade packages to the corresponding upgrade device via the CAN bus for software refresh based on the preset upgrade strategy.

[0071] In terms of SOA services, the central vehicle domain control master module encapsulates the vehicle's lighting control, door lock control, window control and other services, and publishes them to the vehicle's LAN service bus through the Ethernet interface. The client on the vehicle's LAN service bus can call the vehicle's lighting control, door lock control, window control and other services through Ethernet.

[0072] In this embodiment, the intelligent cockpit domain control module CDC is used to process and display the acquired vehicle information, compress and encode the acquired vehicle power information and store it, and output the compressed and encoded vehicle power information to the in-vehicle entertainment human-computer interaction system 30; wherein, the vehicle information includes vehicle surrounding environment information, vehicle personnel behavior monitoring information, vehicle driving information and multimedia information; the intelligent cockpit domain control module is connected to multiple driving sensing devices configured on the vehicle through a low-voltage differential signal video transmission line, and the intelligent cockpit domain control module is also connected to the in-vehicle entertainment human-computer interaction system 30 configured on the vehicle.

[0073] The smart cockpit domain control module includes a smart cockpit domain main control module 131, which is used to use the driving perception device to obtain the surrounding video images of the vehicle and the vehicle personnel behavior monitoring information, and compress, encode and display the surrounding video images of the vehicle, and use the deep learning neural network model to identify the vehicle personnel behavior monitoring information to obtain vehicle personnel behavior identification information; and monitor the working status of the driving perception device connected to the smart cockpit domain control module.

[0074] In a specific embodiment, the intelligent cockpit domain control module CDC mainly assumes the intelligent cockpit control function. As shown in FIG5 , the intelligent cockpit domain control module CDC has the following external interfaces:

[0075] Four LVDS interfaces for connecting to four surround-view fisheye cameras, with specifications of 2MP@195HFOV (resolution of 2MP and field of view of 195 degrees); one LVDS interface for connecting to a DMS camera, with specifications of 2MP@60HFOV (resolution of 2MP and field of view of 60 degrees); one LVDS interface for connecting to an OMS camera, with specifications of 2MP@120HFOV (resolution of 2MP and field of view of 120 degrees); two LVDS interfaces for connecting to two CMS cameras, with specifications of 2MP@60FPS (resolution of 2MP and field of view of 60 degrees); one LVDS interface for connecting to a TOF camera, with specifications of 2MP@7 0HFOV (resolution of 2MP, field of view of 70 degrees); one DP output for connecting to the entertainment screen, the specification of the entertainment screen is 500MP@60FPS (resolution of 500MP, frame rate of 60FPS); one LVDS output for connecting to the central control screen, the specification of the central control screen is 200MP@30FPS (resolution of 200MP, frame rate of 30FPS); two LVDS outputs for connecting to the electronic rearview mirror screen, the specification of the electronic rearview mirror screen is 2MP@30FPS (resolution of 2MP, frame rate of 30FPS); one A2B protocol bus for connecting to the audio system for digital audio communication; two 1000M Ethernet interfaces, reserved for RSE system communication and SOA service, with specification of 1000BASE-T1; three CAN FD interface, respectively connected to the instrument display, HUD, and CPD sensor; one PCIE interface, the PCIE interface is used to communicate with the VDC core board; it should be noted that in Figure 5, HUD is the head-up display system, DMS is the driver monitoring system, OMS is the occupant detection system, CPD is the child monitoring system, Face ID is the facial recognition system, and TOF is the time of flight technology (here refers to the technology of calculating distance based on the difference in flight time).

[0076] The smart cockpit domain master control module includes a combined chip formed by combining the Qualcomm 8155 chip and the Infineon TC397 chip. In this embodiment, the Qualcomm 8155 chip mainly deploys functions such as AVM (Around View Monitoring), HUD (Head-Up Display), electronic rearview mirror control, instrument display, entertainment screen display, rear-seat entertainment, AI voice recognition, rear-seat gesture recognition, OMS (Occupant Monitoring System) detection, and CPD (Child Passenger Detection). At the same time, the Qualcomm 8155 chip also deploys algorithms such as Face ID algorithm, DMS algorithm, OMS algorithm, gesture recognition algorithm, AI voice recognition algorithm, CPD detection algorithm, and DVR, OTA, and diagnostic functions; the Infineon TC397 chip mainly implements CAN communication, diagnosis, information security, and functional safety. For ease of understanding, the functions of the Qualcomm 8155 chip and the Infineon TC397 chip are described in detail below. Among them, the functions of the Qualcomm 8155 chip are specifically as follows:

[0077] AVM is a 360-degree panoramic monitoring audio and video system. It connects to four external surround-view cameras through the LVDS interface to obtain video images in the four directions of the vehicle. The AVM stitching algorithm is used to stitch the video images in the four directions into a 360-degree bird's-eye view. The stitched 360-degree bird's-eye view is then transmitted to the vehicle's central control screen for display through the LVDS interface. The AVM panoramic surround-view monitoring function provided in this embodiment can provide a panoramic view around the vehicle, helping the driver to better understand the surrounding environment. The Qualcomm 8155 chip can also judge the working status of external devices such as cameras connected to the intelligent cockpit domain control module CDC to diagnose whether there are any abnormal conditions.

[0078] HUD is a head-up display system, which projects key information (such as vehicle speed, gear position and navigation information) onto an optical display in front of the driver's line of sight. This key information is output to the HUD optical display by the intelligent cockpit domain control module CDC through the LIN interface, and then this key information is projected in front of the driver's line of sight, so that the driver can obtain important driving information while keeping his line of sight forward, avoiding the driver's attention being distracted by frequently looking down at the dashboard, thereby increasing driving safety.

[0079] DMS is a driver behavior monitoring system. The intelligent cockpit domain control module CDC is connected to the external DMS camera through the LVDS interface. The DMS camera captures the driver's image and transmits the video encoding and compression to the intelligent cockpit domain control module CDC. After receiving the video, the intelligent cockpit domain control module CDC decodes it and then runs the deep learning neural network algorithm through the NPU processor to output the driver's driving behavior results, such as smoking, drowsiness, making phone calls, distraction, etc. When it is detected that the driver's driving behavior is abnormal, the vehicle horn will issue an early warning to correct the driver's behavior.

[0080] Regarding the electronic rearview mirror control function of the Qualcomm 8155 chip, the intelligent cockpit domain control module CDC is connected to the CMS cameras on the left and right sides through the LVDS interface. The CMS cameras transmit real-time video images of the left and right sides of the vehicle to the intelligent cockpit domain control module CDC. The intelligent cockpit domain control module CDC then decodes, crops and stretches the video, and outputs it to the electronic rearview mirror screens on the left and right sides for display. Technical personnel in this field can also allow the driver to adjust parameters such as the angle and brightness of the rearview mirror through touch screen or voice control to improve driving safety.

[0081] In this embodiment, the Qualcomm 8155 chip sends the acquired vehicle driving information to the LCD instrument display unit through the CAN interface for display. The LCD instrument display unit can adopt an entertainment display unit of a pure Android system. The Qualcomm 8155 chip can transmit the acquired video, music and other entertainment services to the entertainment display unit via DP signals for display. In addition, the Qualcomm 8155 chip can be connected to the rear seat entertainment human-computer interaction system RSE through a 1000M Ethernet interface, which can provide passengers with a variety of entertainment content such as video, audio, and games to relieve fatigue during the journey.

[0082] The Qualcomm 8155 chip can also realize AI voice recognition function. The pickup unit transmits the collected digital audio signal in the form of a daisy chain via the A2B bus to the intelligent cockpit domain control module CDC. The intelligent cockpit domain control module CDC decodes the digital audio signal and transmits it to the Qualcomm 8155 chip, so that the Qualcomm 8155 chip uses the AI ​​voice algorithm engine for semantic analysis and outputs voice commands, so that the driver can operate various functions of the car through voice control and improve driving safety.

[0083] The intelligent cockpit domain control module CDC is connected to the DMS camera, external TOF camera, and external OMS camera through the LVDS interface. The DMS camera is used to capture the driver's facial image and use the FACE ID algorithm to extract the driver's facial feature points. The driver's facial feature points are transmitted to the memory of the intelligent cockpit domain control module CDC so that the driver can directly call the FACE ID algorithm when logging in and authenticating later. ID algorithm performs driver facial comparison and outputs the driver facial comparison result; the external TOF camera is used to capture the gestures of rear-seat passengers in the car, and transmits the gestures of rear-seat passengers in the car to the Qualcomm 8155 chip of the smart cockpit domain control module CDC, so that the Qualcomm 8155 chip uses the gesture recognition algorithm to recognize the gestures of rear-seat passengers in the car and outputs operation instructions, thereby improving the convenience of passengers; the external OMS camera is used to capture the real-time images of rear-seat passengers, and encode the real-time images of rear-seat passengers and transmit them to the Qualcomm 8155 chip of the smart cockpit domain control module CDC, so that the Qualcomm 8155 chip uses the OMS occupant detection algorithm to identify the behavior of the occupants (such as seat belt wearing detection, driver interference detection, etc.). Those skilled in the art can also apply it to the gesture recognition of all passengers in the car, not limited to the embodiments of the present application.

[0084] The intelligent cockpit domain control module CDC is connected to the external CPD detection unit through the CAN interface. The external CPD detection unit includes a MEMS ultrasonic radar sensor, which is used to detect the presence of child passengers and feed back the child passenger detection results to the external CPD detection unit, so that the external CPD detection unit can analyze the child passenger detection results through a clustering algorithm to obtain child behavior information, and send the child behavior information to the Qualcomm 8155 chip of the intelligent cockpit domain control module CDC, so that the Qualcomm 8155 chip uses pre-set reminder rules to send reminder information to the user, reminding the driver or other passengers to place child safety seats in time to ensure the safety of child passengers.

[0085] During the OTA upgrade process, the server sends the remote upgrade package to the on-board TBOX via the wireless network (4G / 5G). The on-board TBOX sends the remote upgrade package to the central vehicle domain control module VDC via Ethernet. The central vehicle domain control module VDC transmits it to the intelligent cockpit domain control module CDC through the PCIE interface. The intelligent cockpit domain control module CDC writes the remote upgrade package to the memory, overwriting the original application file.

[0086] It should be noted that the DMS algorithm, OMS algorithm, gesture recognition algorithm, and AI speech recognition algorithm are all trained and processed by deep learning neural network algorithms on the transmitted audio and video materials. Among them, the DMS algorithm is used to identify the driver's driving behaviors such as smoking, making phone calls, distraction, drowsiness, closing eyes, yawning, and wearing sunglasses; the OMS algorithm is used to identify behaviors such as passengers not wearing seat belts and interfering with the driver; the gesture recognition algorithm is used to identify passengers' gestures; and the AI ​​speech recognition algorithm is used to recognize speech semantics.

[0087] In this embodiment, the data interaction module is used to control the internal and external data interaction of the autonomous driving domain control module, the central vehicle domain control module and the smart cockpit domain control module; the data interaction module includes an internal connection unit 141, a vehicle external connection unit 142, a primary power supply unit 143 and an interface protection unit 144; the internal connection unit is used to establish internal communication connections with the autonomous driving domain control module, the central vehicle domain control module and the smart cockpit domain control module; the vehicle external connection unit is used to control the communication between the vehicle external device and the autonomous driving domain control module, the central vehicle domain control module and the smart cockpit domain control module; the primary power supply unit is used to control the vehicle battery supply voltage; the interface protection unit is used for power reverse connection protection and overvoltage surge protection.

[0088] In a specific embodiment, the data interaction module is mainly responsible for the functions of communication signal conversion, external interface, and whole-machine power supply. As shown in Figure 6, the internal connection unit is connected to the autonomous driving domain control module, the central vehicle domain control module and the intelligent cockpit domain control module through the Hirose floating electronic connector. The vehicle external connection unit has the following external interfaces: 16-channel LVDS input interface, 4-channel LVDS output interface, one DP output interface, 11-channel 1000BASE-T1 interface, five-channel 100BASE-T1 interface, one 100BASE-TX interface, 21-channel CAN FD interface, 4-channel LIN interface, 1-channel A2B interface, and 27-channel hard-wired signal interface.

[0089] The data interaction module also includes a primary power supply unit and an interface protection unit. The main functions of the primary power supply unit and the interface protection unit are to realize power conversion and provide interface anti-static, anti-surge and anti-electromagnetic interference functions, which play an important role in ensuring the normal operation of electronic equipment. Specifically, the primary power supply unit is responsible for adjusting the normal supply voltage of the vehicle battery and using TVS (transient diode) for voltage surge protection. In this embodiment, the interface protection unit includes an anti-reverse connection circuit, a common-mode inductor circuit and a buck-boost circuit. Among them, the anti-reverse connection circuit is used to prevent current backflow, the common-mode inductor circuit is used for filtering, and the buck-boost circuit is used to convert the voltage to a stable 12V state. The data interaction module also adds an electrostatic protection circuit at all external interfaces. The electrostatic protection circuit is used to filter out external electrostatic interference to prevent electrostatic interference from entering the machine, avoiding damage to internal devices and circuits or abnormal operation.

[0090] In an embodiment of the present application, the vehicle central supercomputing control unit 10 is a supercomputing control device comprising multiple core boards. The data interaction module 14 is an interface board comprising multiple communication interfaces and multiple processors or controllers. The internal connection unit 141 and the vehicle external connection unit 142 are multiple communication interfaces, respectively. The primary power supply unit 143 is a primary power supply circuit, and the interface protection unit 144 is an interface protection circuit.

[0091] This embodiment adopts the design of a vehicle central supercomputing unit including a data interaction module (interface board), which not only makes the overall interface of the product more centralized, reduces the complexity of interface management, and improves the overall stability of the vehicle, but also makes the interface more protective, which can better protect the internal circuits and components of the vehicle and improve the safety of the vehicle. This embodiment enhances the reliability of the entire vehicle through design optimization such as reducing the number of controllers and concentrating interfaces. At the same time, due to the improvement in communication efficiency, the safety of the vehicle can also be better ensured.

[0092] The vehicle central supercomputing control unit proposed in this embodiment is a highly integrated system achieved through domain fusion at the vehicle architecture level. The central vehicle domain control module, the autonomous driving domain control module, and the intelligent cockpit domain control module are all managed by the vehicle central supercomputing control unit. This architecture not only enables data sharing and improves data utilization efficiency, for example, when both the autonomous driving domain control module and the intelligent cockpit domain control module need to obtain vehicle status information, this information can be shared between different domains through the management of the vehicle central supercomputing control unit, avoiding repeated collection and processing, thereby improving the vehicle's response speed and execution efficiency. It also enables the vehicle central supercomputing control unit to more centrally control and manage various vehicle functions such as control, driving, and cockpit, better implementing mutual backup and redundant design of functions, thereby better achieving optimal resource allocation and improving the vehicle's functional safety performance. In addition, this embodiment adopts a domain-centralized electronic architecture design, which can reduce the number and complexity of wiring harnesses or electronic components, reducing vehicle costs. At the same time, the design of the central supercomputing unit makes vehicle upgrades and iterations more convenient and efficient. For example, when the intelligent driving system needs to be upgraded, only the software of the vehicle central supercomputing control unit needs to be updated, without the need to modify the entire vehicle, simplifying the development and maintenance process and improving vehicle reliability.

[0093] An embodiment of the present application provides a vehicle central supercomputing control unit, which integrates a central vehicle domain control module, an autonomous driving domain control module, an intelligent cockpit domain control module and a data interaction module; the autonomous driving domain control module is used to generate automatic driving control information by using multiple driving perception devices configured on the vehicle and the acquired target position information; the central vehicle domain control module is used to control the vehicle body according to the automatic driving control information sent by the autonomous driving domain control module; the intelligent cockpit domain control module is used to process and display the acquired vehicle information; the data interaction module is used to control the internal and external data interaction of the autonomous driving domain control module, the central vehicle domain control module and the intelligent cockpit domain control module. The vehicle central supercomputing control unit proposed in this embodiment is a highly integrated system achieved through domain fusion at the level of the entire vehicle architecture. The entire vehicle control and computing parts are completed by the vehicle central supercomputing control unit, which can better coordinate data sharing and control signal transmission between various domains. Moreover, this vehicle central supercomputing control unit design based on highly integrated domain fusion enables the entire vehicle control and computing parts of the vehicle to have higher integration and more powerful processing capabilities. It can handle complex algorithms, models and large-scale data, realize more advanced driving assistance and autonomous driving functions, and better support the intelligent and networked development of vehicles, thereby achieving more efficient, smarter and safer vehicle control.

[0094] In one embodiment, as shown in Figure 7, an embodiment of the present application provides a vehicle system, which includes a vehicle body 20, a vehicle central supercomputing control unit 10 as described above, which is arranged in the vehicle body, and several regional controllers 30 arranged around the vehicle body. The several regional controllers 30 are connected to the vehicle central supercomputing control unit 10 through an Ethernet interface in a ring network manner, and four regional controllers ZCU are connected to the vehicle central supercomputing control unit to form a closed loop.

[0095] The regional controller 30 is used to detect the status of vehicle opening and closing parts and the power distribution status, and after receiving the vehicle control signal from the vehicle central supercomputing control unit, it controls the driving sensing device and power circuit relay configured on the vehicle body according to the vehicle control signal.

[0096] The electrical architecture of the entire vehicle is shown in Figure 8. In Figure 8, Cameras refers to on-board cameras, Radars refers to millimeter-wave radars, Lidars refers to laser radars, RSE refers to rear seat entertainment systems, and Serdes refers to high-speed serial interfaces. The vehicle's central supercomputing control unit HPC serves as the central brain of the vehicle and is primarily responsible for vehicle control logic, computing, and central gateway functions. Four regional controllers ZCU are arranged around the vehicle and are connected to the vehicle's central supercomputing control unit HPC via Ethernet in a ring network. In this embodiment, the regional controller ZCU is primarily responsible for the perception, control, and execution of the vehicle body and the entire vehicle, while the decision-making strategy is completed by the vehicle's central supercomputing control unit HPC. The vehicle's central supercomputing control unit (HPC) is equipped with driving perception components such as cameras, millimeter-wave radars, ultrasonic radars, and lidars, as well as cockpit interaction components such as central control, instruments, and entertainment screens. These components are directly connected to the vehicle's central supercomputing control unit (HPC). The vehicle's central supercomputing control unit (HPC) also serves as the central gateway for the entire vehicle, as well as the vehicle's Ethernet center node and the center node of the CAN network. The on-board T-BOX (Telematics-BOX, intelligent on-board terminal) is directly connected to the vehicle's central supercomputing control unit (HPC) via 1000M Ethernet. In addition, the vehicle's central supercomputing control unit (HPC) also reserves three Gigabit Ethernet interfaces for connecting to computing power expansion units under the SOA architecture.

[0097] For the specific definition of a vehicle system, please refer to the above-mentioned definition of a vehicle central supercomputing control unit, which will not be repeated here. Those of ordinary skill in the art will appreciate that the various modules and functions described in conjunction with the embodiments disclosed in this application can be implemented in hardware, software, or a combination of both. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0098] An embodiment of the present application provides a vehicle central supercomputing control unit and a vehicle system, wherein a vehicle central supercomputing control unit forms a highly integrated computing platform by integrating a central vehicle domain control module, an autonomous driving domain control module, an intelligent cockpit domain control module and a data interaction module, thereby achieving higher-integration domain fusion at the level of the entire vehicle architecture, so as to better coordinate data sharing and control signal transmission between domains, improve data utilization efficiency, thereby achieving faster data transmission and processing, and improving the vehicle's response speed and execution efficiency. In addition, this highly integrated control architecture can reduce the number and complexity of wiring harnesses, reduce the cost and time of repeated development for different platforms, and reduce development cost and time. The central supercomputing control unit based on high-integration domain fusion provided in this embodiment can better support the intelligent and networked development of vehicles, thereby achieving more efficient, smarter and safer vehicle control, and is one of the important development directions of future automotive electronic architecture.

[0099] The above-described embodiments merely represent several preferred implementations of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art could make several improvements and substitutions without departing from the technical principles of the present application, and such improvements and substitutions should also be considered within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be based on the scope of protection of the claims.

Claims

1. A vehicle central supercomputer control unit, wherein: The vehicle central supercomputing control unit integrates an autonomous driving domain control module, a central vehicle domain control module, an intelligent cockpit domain control module and a data interaction module. The autonomous driving domain control module, the central vehicle domain control module and the intelligent cockpit domain control module are interconnected and are all connected to the data interaction module; The autonomous driving domain control module is used to generate automatic driving control information using the acquired target position information and information acquired by multiple driving sensing devices configured on the vehicle; The central vehicle domain control module is used to control the vehicle body according to the automatic driving control information sent by the automatic driving domain control module; The intelligent cockpit domain control module is used to process and display the acquired vehicle information; the vehicle information includes vehicle surrounding environment information, vehicle personnel behavior monitoring information, vehicle driving information and multimedia information; The data interaction module is used to control the internal and external data interactions among the autonomous driving domain control module, the central vehicle domain control module and the smart cockpit domain control module.

2. A vehicle central supercomputer control unit according to claim 1, wherein: The autonomous driving domain control module includes two autonomous driving domain master control modules, which are respectively a first autonomous driving domain master control module and a second autonomous driving domain master control module; The first autonomous driving domain master control module is used to use multiple driving perception devices configured on the vehicle to obtain the vehicle's surrounding environment information and the target position information generated by the second autonomous driving domain master control module, and use a deep learning neural network model to process the vehicle's surrounding environment information and target position information to perceive the vehicle's surrounding environment to plan a local or global driving path.

3. A vehicle central supercomputer control unit as claimed in claim 2, wherein: The second autonomous driving domain master control module is used to generate automatic driving control information according to the local or global driving path and current road condition information sent by the first autonomous driving domain master control module; wherein the automatic driving control information includes vehicle power control information and vehicle-mounted inertial navigation positioning information; Furthermore, the received ultrasonic target detection signals, millimeter wave target detection signals and laser target detection signals are filtered and classified to obtain target position information.

4. A vehicle central supercomputer control unit according to claim 1, wherein: The central vehicle domain control module includes a central vehicle domain control master control module; The central vehicle domain control main control module is used to perform vehicle body control and vehicle power control according to the received automatic driving control information, and generate vehicle opening and closing part status control information according to the acquired vehicle opening and closing part status information and the pre-set vehicle opening and closing part control strategy; And, control the mutual communication between different CAN bus interfaces and Ethernet interfaces configured on the vehicle.

5. A vehicle central supercomputing control unit as claimed in claim 4, wherein: The central vehicle domain control master control module is also used to control the power circuit relay and generate a power distribution management signal according to the preset battery management strategy and vehicle power demand; And, according to the pre-acquired temperature control requirements and automatic air conditioning control strategies, the vehicle air conditioning system is managed and controlled. The temperature control requirements are generated by a pre-set thermal management strategy.

6. A vehicle central supercomputer control unit according to claim 1, wherein: The smart cockpit domain control module is connected to a plurality of driving sensing devices configured on the vehicle through a low voltage differential signal video transmission line, and the smart cockpit domain control module includes a smart cockpit domain main control module; The smart cockpit domain main control module is used to use the multiple driving perception devices to obtain the vehicle's surrounding video images and vehicle personnel behavior monitoring information, compress, encode and display the vehicle's surrounding video images, use a deep learning neural network model to identify the vehicle personnel behavior monitoring information, and obtain vehicle personnel behavior identification information; and monitor the working status of multiple driving perception devices connected to the smart cockpit domain control module.

7. A vehicle central supercomputer control unit as claimed in claim 6, wherein: The intelligent cockpit domain control module is connected to the in-vehicle entertainment human-computer interaction system configured on the vehicle; The smart cockpit domain main control module is also used to compress and encode the acquired vehicle power information and store it, and output the compressed and encoded vehicle power information to the in-vehicle entertainment human-computer interaction system.

8. A vehicle central supercomputer control unit according to claim 1, wherein: The data interaction module includes an internal connection unit, a vehicle external connection unit, a primary power supply unit and an interface protection unit; The internal connection unit is used to perform internal communication connection with the autonomous driving domain control module, the central vehicle domain control module and the smart cockpit domain control module; The vehicle external connection unit is used to control the vehicle external device to communicate with the autonomous driving domain control module, the central vehicle domain control module and the smart cockpit domain control module; The primary power supply unit is used to control the vehicle battery supply voltage; The interface protection unit is used for power supply reverse connection protection and overvoltage surge protection.

9. A vehicle system, wherein: It comprises a vehicle body, a vehicle central supercomputer control unit as claimed in claim 1 arranged in the vehicle body, and a plurality of regional controllers arranged around the vehicle body, wherein the plurality of regional controllers are connected to the vehicle central supercomputer control unit via an Ethernet interface in a ring network connection manner.

10. A vehicle system, wherein: It comprises a vehicle body, a vehicle central supercomputer control unit as claimed in claim 2 arranged in the vehicle body, and a plurality of regional controllers arranged around the vehicle body, wherein the plurality of regional controllers are connected to the vehicle central supercomputer control unit via an Ethernet interface in a ring network connection manner.

11. A vehicle system, wherein: It comprises a vehicle body, a vehicle central supercomputer control unit as claimed in claim 3 arranged in the vehicle body, and a plurality of regional controllers arranged around the vehicle body, wherein the plurality of regional controllers are connected to the vehicle central supercomputer control unit via an Ethernet interface in a ring network connection manner.

12. A vehicle system, wherein: It comprises a vehicle body, a vehicle central supercomputer control unit as claimed in claim 4 arranged in the vehicle body, and a plurality of regional controllers arranged around the vehicle body, wherein the plurality of regional controllers are connected to the vehicle central supercomputer control unit via an Ethernet interface in a ring network connection manner.

13. A vehicle system, wherein: It comprises a vehicle body, a vehicle central supercomputer control unit as claimed in claim 5 arranged in the vehicle body, and a plurality of regional controllers arranged around the vehicle body, wherein the plurality of regional controllers are connected to the vehicle central supercomputer control unit via an Ethernet interface in a ring network connection manner.

14. A vehicle system, wherein: It comprises a vehicle body, a vehicle central supercomputer control unit as claimed in claim 6 arranged in the vehicle body, and a plurality of regional controllers arranged around the vehicle body, wherein the plurality of regional controllers are connected to the vehicle central supercomputer control unit via an Ethernet interface in a ring network connection manner.

15. A vehicle system, wherein: It comprises a vehicle body, a vehicle central supercomputer control unit as claimed in claim 7 arranged in the vehicle body, and a plurality of regional controllers arranged around the vehicle body, wherein the plurality of regional controllers are connected to the vehicle central supercomputer control unit via an Ethernet interface in a ring network connection manner.

16. A vehicle system, wherein: It comprises a vehicle body, a vehicle central supercomputer control unit as claimed in claim 8 arranged in the vehicle body, and a plurality of regional controllers arranged around the vehicle body, wherein the plurality of regional controllers are connected to the vehicle central supercomputer control unit via an Ethernet interface in a ring network connection manner.

17. A vehicle system as claimed in claim 9, wherein: The regional controller is used to detect the status of vehicle opening and closing parts and the power distribution status, and after receiving the vehicle control signal from the vehicle central supercomputing control unit, it controls multiple driving sensing devices and power circuit relays configured on the vehicle according to the vehicle control signal.

18. A vehicle system as claimed in claim 10, wherein: The regional controller is used to detect the status of vehicle opening and closing parts and the power distribution status, and after receiving the vehicle control signal from the vehicle central supercomputing control unit, it controls multiple driving sensing devices and power circuit relays configured on the vehicle according to the vehicle control signal.

19. A vehicle system as claimed in claim 12, wherein: The regional controller is used to detect the status of vehicle opening and closing parts and the power distribution status, and after receiving the vehicle control signal from the vehicle central supercomputing control unit, it controls multiple driving sensing devices and power circuit relays configured on the vehicle according to the vehicle control signal.

20. A vehicle system as claimed in claim 14, wherein: The regional controller is used to detect the status of vehicle opening and closing parts and the power distribution status, and after receiving the vehicle control signal from the vehicle central supercomputing control unit, it controls multiple driving sensing devices and power circuit relays configured on the vehicle according to the vehicle control signal.

Images