An intelligent vehicle terminal based on edge computing and multi-sensor interface

Abstract

The present invention relates to the field of intelligent vehicle terminal technology, specifically an intelligent vehicle terminal based on edge computing and multi-sensor interface, which is composed of an embedded board, a sensor module, a wireless connection module, a driver interaction module, a peripheral module and an edge computing module, wherein the sensor module, the wireless connection module, the driver interaction module, the peripheral module and the edge computing module are all electrically connected to the embedded board, and the embedded board receives multi-source data information from the sensor module, the wireless connection module, the driver interaction module and the peripheral module for fusion. The present invention can effectively solve the problems of insufficient interfaces of multiple sensors or multiple module devices installed later, inability to perform real-time edge computing on data collected by multiple sensors or multiple modules, inability to perform data fusion on data of multiple source sensors or multiple source modules, and inability to autonomously control the vehicle with the result of data fusion of multiple source sensors or multiple source modules.

Description

[0001] A smart vehicle-mounted terminal based on edge computing and multi-sensor interface Technical field

[0002] The present invention and the technical field of intelligent vehicle-mounted terminals specifically relate to an intelligent vehicle-mounted terminal based on edge computing and a multi-sensor interface. Background Art

[0003] At present, there are many types of known vehicle-mounted terminal devices, such as video surveillance vehicle-mounted terminal devices for commercial vehicles, muck trucks, and sanitation vehicles, which can support multi-channel video surveillance and have satellite positioning and remote wireless transmission of video data. However, in these vehicle-mounted terminal devices, it is still difficult to solve the problem of insufficient interfaces for multiple sensor devices installed later; the problem of being unable to perform real-time edge computing on data collected by multiple sensors; the problem of being unable to perform data fusion on multi-sensor data, and the problem of being unable to autonomously control the vehicle with the results of multi-source data fusion. Summary of the invention

[0004] The purpose of this patent is to provide an intelligent vehicle-mounted terminal based on edge computing and multi-sensor interface to solve the problems of insufficient interfaces for multiple sensors or multiple module devices installed later, inability to perform real-time edge computing on data collected by multiple sensors or multiple modules, inability to perform data fusion on multi-source sensor or multi-source module data, and inability to autonomously control the vehicle with the results of multi-source sensor or multi-source module data fusion.

[0005] The technical solution adopted by the present invention to solve the technical problem is: an intelligent vehicle-mounted terminal based on edge computing and multi-sensor interface, characterized in that it is composed of an embedded board, a sensor module, a wireless connection module, a driver interaction module, a peripheral module and an edge computing module, and the sensor module, the wireless connection module, the driver interaction module, the peripheral module and the edge computing module are all electrically connected to the embedded board; the embedded board is composed of an stm32 series embedded processor chip with an ARM core and a chip peripheral circuit, the IO port of the embedded processor chip is extended to lead to a GPIO interface, a serial port and an ADC interface, and a plurality of peripheral circuit interfaces are provided on the chip peripheral circuit, and specifically are a USB interface, a LAN interface, a CAN interface, an SMA interface, an FFC interface, an RS485 interface, an RJ45 interface, a MIC interface, an Amplifer interface, an OBD interface, SD interface, SIM interface, DC interface, A / V interface, IPC interface, Panel interface and GPIO / Sensor interface; the embedded processor chip is connected to the chip peripheral circuit through the IO port, and the GPIO interface extension leads to the LAN interface, SMA interface, FFC interface, RJ45 interface, Amplifer interface, SD interface, SIM interface, DC interface, IPC interface, GPIO / Sensor interface connected to the chip peripheral circuit; the serial port extension leads to the USB interface, CAN interface, RS485 interface, OBD interface, IIC interface and Panel interface connected to the chip peripheral circuit; the ADC interface extension leads to the MIC interface and A / V interface; the DC interface of the embedded board is externally connected to a DC power supply with a voltage range of 8V-36V.

[0006] Furthermore, the wireless connection module includes a Bluetooth module and a WiFi module, both of which are electrically connected to the embedded board through a USB interface. Some sensors in the sensor module are electrically connected to the embedded board using a Bluetooth module. Due to the differences in vehicles, the different spaces for installing sensors, and the different sensor installation and connection methods used, some sensors are connected to the embedded board using wireless Bluetooth. The Bluetooth module is used to connect a variety of sensor devices; the WiFi module is used for debugging and data viewing on the mobile phone APP of the smart vehicle terminal.

[0007] Furthermore, the driver interaction module includes a display screen and an alarm button, the display screen is electrically connected to the embedded board through a Panel interface, and the alarm button is electrically connected to the embedded board through a GPIO / Sensor interface.

[0008] Furthermore, the display screen displays the current vehicle status information collected by the intelligent vehicle terminal. The alarm button is used for emergency alarm of the driver in a dangerous rescue state. After the driver triggers the alarm button, the alarm command information is wirelessly transmitted to the remote monitoring center through the SIM card in the peripheral module. The remote monitoring center can view the image information of the alarm vehicle in real time through the multi-channel cameras in the peripheral module, and monitor the audio information of the alarm vehicle in real time through the microphone in the peripheral module.

[0009] Furthermore, the peripheral module includes multiple cameras, a microphone, an information broadcast screen, a GPS / Beidou dual-mode positioning module, a speaker, an OBD parser, a SIM card and a memory.

[0010] Furthermore, the multi-channel camera includes an analog camera and an IP network camera, the analog camera is electrically connected to the embedded board through an A / V interface, and the IP network camera is electrically connected to the embedded board through an IPC interface; the image information monitored by the multi-channel camera can be displayed in real time through a display screen and stored in a memory in real time; at the same time, the remote monitoring center uses a SIM card to monitor the images of the multi-channel cameras in real time through a mobile wireless network, and the mobile phone APP establishes a wireless connection with the smart vehicle terminal through a WiFi module to retrieve the image information of the multi-channel cameras in real time.

[0011] Furthermore, the multi-channel camera also includes a camera for monitoring driver behavior, a front vehicle collision avoidance camera, a passenger flow monitoring camera, and a reversing image camera. After the camera is turned on, the camera for monitoring driver behavior, the front vehicle collision avoidance camera, the passenger flow monitoring camera, and the reversing image camera transmit the monitoring video data in real time to the edge computing module through the embedded board for real-time analysis.

[0012] Furthermore, the microphone is electrically connected to the embedded board through a MIC interface, and the microphone collects audio data together with multiple cameras in real time and stores the audio data in a memory.

[0013] Furthermore, the information broadcast screen is electrically connected to the embedded board through the FFC interface, and the embedded board displays the vehicle location information, arrival information and advertising information in the memory collected by the GPS / Beidou dual-mode positioning module on the information broadcast screen in real time. The information broadcast screen can adopt standard multi-size LED, LCD, OLED display screens.

[0014] Furthermore, the GPS / Beidou dual-mode positioning module is electrically connected to the embedded board through an SMA interface. The GPS / Beidou dual-mode positioning module adopts GPS and Beidou dual-mode positioning chips, and transmits the collected vehicle position information in real time to the display screen, information broadcast screen and embedded board.

[0015] Furthermore, the speaker is electrically connected to the embedded board through an Amplifer interface, and the driver can broadcast voice information in real time through a microphone and simultaneously play it out through the speaker.

[0016] Furthermore, the OBD parser is electrically connected to the embedded board through the OBD interface. The OBD parser reads the vehicle's OBD data through the vehicle's CAN bus, stores the read OBD data in a memory through the embedded board, and transmits it to the edge computing module at the same time.

[0017] Furthermore, the SIM card is electrically connected to the embedded board through a SIM card slot interface, the SIM card is a 4G SIM card and is backward compatible, and transmits data through a mobile wireless cellular network.

[0018] Furthermore, the memory is electrically connected to the embedded board through a USB or SD interface, and the memory is one of a hard disk, an SD card or a U disk, and is used to store sensor module data, video and audio data, OBD data, and location information.

[0019] Furthermore, the sensor module includes at least a temperature and humidity sensor, an angle sensor, a vibration sensor, a multi-axis gyroscope, an acceleration sensor, a hazardous chemicals sensor, a tire pressure sensor, a rainfall sensor and a distance sensor;

[0020] Furthermore, the temperature and humidity sensor adopts the DHT11 temperature and humidity sensor module of Risym Company, and the data interface of the temperature and humidity sensor is electrically connected to the GPIO / Sensor interface of the embedded board; the temperature and humidity sensor is used to collect the internal working environment status data of the vehicle terminal. When the ambient temperature exceeds 60 degrees Celsius and / or the humidity exceeds 90%, the vehicle terminal issues a danger alarm to remind the driver that the terminal equipment may fail and should be handled in time.

[0021] Furthermore, the multi-axis gyroscope and acceleration sensor both use InvenSense's MPU6050 chip as the core chip, and the MPU6050 chip and the embedded board are electrically connected and communicate data through the IIC interface; the MPU6050 chip contains a 3-axis gyroscope and a 3-axis acceleration sensor. The multi-axis gyroscope and acceleration sensor can monitor the vehicle's driving status information in real time, and transmit the information to the edge computing module through the embedded board, and perform information fusion analysis as part of the multi-source sensor data to determine the vehicle status and fault status.

[0022] Furthermore, the vibration sensor adopts a piezoelectric vibration sensor SEN0209, which is electrically connected to the GPIO / Sensor interface of the embedded board to collect the analog voltage output by the piezoelectric vibration sensor SEN0209, collect the vibration state of the vehicle terminal and the large-amplitude vibration state of the vehicle in real time, and transmit the information to the edge computing module through the embedded board, and perform information fusion analysis as part of the multi-source sensor data to judge the vehicle status and fault status.

[0023] Furthermore, the hazardous chemicals sensor adopts the MQ-2-135-3-7-9 hazardous gas sensor, which is electrically connected to the GPIO / Sensor interface of the embedded board. When the gas sensor detects hazardous gas, it outputs a digital signal. The embedded board receives and transmits the signal output by the hazardous chemicals sensor and transmits it to the display screen, while giving the driver a danger warning through the horn.

[0024] Furthermore, the tire pressure sensor uses the SP40 tire pressure sensor chip of Infineon Technologies, and is connected to the embedded board through a Bluetooth module. The collected data is displayed on the display in real time, and an alarm is given through the speaker. The tire pressure data is stored in the memory in real time through the embedded board. The tire pressure data is transmitted to the edge computing module through the embedded board within a certain period for modeling and analysis, and the maintenance data and maintenance status of the tire can be obtained to provide intelligent maintenance decisions.

[0025] Furthermore, the distance sensor is composed of an ultrasonic distance sensor and is electrically connected to the embedded board through the GPIO / Sensor interface. The distance sensor is installed at the front and rear of the vehicle to monitor the distance of objects in front and behind the vehicle in real time, and transmits the monitoring data to the edge computing module through the embedded board for multi-sensor data information fusion analysis. When the vehicle is traveling at a low speed of less than 30km / h, the distance sensor will issue a real-time alarm for objects within 1 meter, and for objects within 2 meters of the front vehicle with a vehicle speed greater than 30km / h. Based on the edge computing module's fusion analysis of multi-sensor data, the vehicle can be emergency braked through the CAN bus of the embedded board for an impending front vehicle collision accident.

[0026] Furthermore, the rain sensor is electrically connected to the RS485 interface on the embedded board. The rain sensor can adjust the operating speed of the wiper action by analyzing the rain precipitation and water intensity data information, effectively reducing the driver's driving troubles and improving the reliability of safe driving. The rain sensor can also monitor the start and end time of precipitation. The rain sensor here is existing technology and will not be elaborated on.

[0027] Furthermore, the embedded board receives data information output by the sensor module, the wireless connection module, the driver interaction module and the peripheral module, and integrates the data information.

[0028] Furthermore, the edge computing module uses Intel's NCS neural computing stick, which has a built-in OpenVINO SDK software development kit. The edge computing module is electrically connected to the embedded board via a USB or / and RS485 or / and FFC interface. The edge computing module can be configured with one or more NCS neural computing sticks according to the needs of the system computing power.

[0029] Furthermore, the edge computing module analyzes the multi-source module data information received by the embedded board, and the analysis results are stored in the memory by the embedded board. The edge computing module can extract part of the OBD data for data fusion calculation according to the needs of data fusion.

[0030] Furthermore, after analyzing the multi-source module data information in the embedded board, the edge computing module can communicate with the driver through the display screen and speaker of the intelligent vehicle terminal according to the analysis results and for specific application scenarios, and can also take emergency braking of the vehicle through the CAN bus of the embedded board. The edge computing module can perform real-time edge computing on the data collected by multiple sensors, and fuse the multi-source sensor data according to the application scenario, set data weights, and clean invalid data.

[0031] Furthermore, the multi-source sensor data is comprehensive data information obtained by each sensor in the sensor module.

[0032] Furthermore, the multi-source module data is comprehensive data information obtained by the sensor module, the wireless connection module, the driver interaction module and the peripheral module.

[0033] Furthermore, the calculation method of the NCS neural computing stick is a prior art and will not be elaborated here.

[0034] The beneficial effects of the intelligent vehicle-mounted terminal based on edge computing and multi-sensor interface of the present invention are:

[0035] (1) The setting of the peripheral circuit interface of the embedded board can effectively solve the problem of insufficient interfaces for multiple sensors or multiple module devices installed later;

[0036] (2) The embedded board uses an STM32 series embedded processor chip with an ARM core to collect and fuse multi-source sensor data or multi-source module data, solving the problem that the prior art cannot fuse multi-source sensor data or multi-source module data;

[0037] (3) The edge computing module uses the NCS neural computing stick, which can perform real-time edge computing on the fused data collected by multi-source sensors or multi-source modules. The edge computing module can be configured with one or more NCS neural computing sticks according to the computing power requirements of the system, which effectively solves the problem in the prior art that it is impossible to perform real-time edge computing on the data collected by multi-source sensors or multi-source modules, and solves the problem that the data fusion results of multi-source sensors or multi-source modules cannot be used for autonomous control of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0039] Attached Figure 1 It is a schematic diagram of an intelligent vehicle-mounted terminal based on edge computing and multi-sensor interface described in the present invention.

[0040] Attached Figure 1 Explanation of the numbers in the figure: 1-embedded board, 2-sensor module, 3-wireless connection module, 4-driver interaction module, 5-peripheral module, 6-edge computing module, 201-temperature and humidity sensor, 202-angle sensor, 203-vibration sensor, 204-multi-axis gyroscope, 205-acceleration sensor, 206-hazardous chemicals sensor, 207-tire pressure sensor, 208-rainfall sensor, 209-distance sensor, 301-Bluetooth module, 302-WiFi module, 401-display screen, 402-alarm button, 501-multi-channel camera, 502-microphone, 503-information broadcast screen, 504-GPS / Beidou dual-mode positioning module, 505-speaker, 506-OBD parser, 507-SIM card, 508-memory. DETAILED DESCRIPTION

[0041] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] Example 1

[0043] Figure 1 Shown

[0044] In this embodiment, an intelligent vehicle terminal based on edge computing and multi-sensor interface is characterized in that it is composed of an embedded board 1, a sensor module 2, a wireless connection module 3, a driver interaction module 4, a peripheral module 5 and an edge computing module 6, wherein the sensor module 2, the wireless connection module 3, the driver interaction module 4, the peripheral module 5 and the edge computing module 6 are all electrically connected to the embedded board 1; the embedded board 1 is composed of an stm32 series embedded processor chip with an ARM core and a chip peripheral circuit, the IO port of the embedded processor chip is extended to lead to a GPIO interface, a serial port and an ADC interface, and a plurality of peripheral circuit interfaces are provided on the chip peripheral circuit, specifically a USB interface, a LAN interface, a CAN interface, an SMA interface, an FFC interface, an RS485 Interface, RJ45 interface, MIC interface, Amplifer interface, OBD interface, SD interface, SIM interface, DC interface, IIC interface (not shown in the figure), A / V interface, IPC interface, Panel interface and GPIO / Sensor interface; the embedded processor chip is connected to the chip peripheral circuit through the IO port, and the GPIO interface extension leads to the LAN interface, SMA interface, FFC interface, RJ45 interface, Amplifer interface, SD interface, SIM interface, DC interface, IPC interface, GPIO / Sensor interface connected to the chip peripheral circuit; the serial port extension leads to the USB interface, CAN interface, RS485 interface, OBD interface, IIC interface (not shown in the figure) and Panel interface connected to the chip peripheral circuit; the ADC interface extension leads to the MIC interface and the A / V interface; the DC interface of the embedded board 1 is externally connected to a DC power supply with a voltage range of 8V-36V.

[0045] In this embodiment, a plurality of backup interfaces can be provided on the peripheral circuit of the embedded board 1 according to requirements.

[0046] The embedded board 1 in this embodiment can realize the integration of comprehensive data information obtained by each sensor in the sensor module 2; the embedded board 1 can also realize the integration of comprehensive data information obtained by the sensor module 2, the wireless connection module 3, the driver interaction module 4 and the peripheral module 5.

[0047] The sensor module 2 in this embodiment includes at least a temperature and humidity sensor 201 , an angle sensor 202 , a vibration sensor 203 , a multi-axis gyroscope 204 , an acceleration sensor 205 , a hazardous chemicals sensor 206 , a tire pressure sensor 207 , a rainfall sensor 208 and a distance sensor 209 .

[0048] The temperature and humidity sensor 201 in this embodiment adopts the DHT11 temperature and humidity sensor module of Risym Company. The data interface of the temperature and humidity sensor 201 is electrically connected to the GPIO / Sensor interface of the embedded board 1; the temperature and humidity sensor 201 is used to collect the internal working environment status data of the vehicle terminal. When the ambient temperature exceeds 60 degrees Celsius and / or the humidity exceeds 90%, the vehicle terminal issues a danger alarm to remind the driver that the terminal equipment may fail and should be handled in time.

[0049] The multi-axis gyroscope 204 and acceleration sensor 205 in this embodiment both use the MPU6050 chip of InvenSense as the core chip. The MPU6050 chip and the embedded board 1 are electrically connected and communicate data via an IIC interface (not shown in the figure); the MPU6050 chip contains a 3-axis gyroscope and a 3-axis acceleration sensor. The multi-axis gyroscope 204 and acceleration sensor 205 can monitor the vehicle's driving status information in real time, and transmit the information to the edge computing module 6 through the embedded board 1, and perform information fusion analysis as part of the multi-source sensor data to determine the vehicle status and fault status.

[0050] The vibration sensor 203 in this embodiment adopts a piezoelectric vibration sensor SEN0209, which is electrically connected to the GPIO / Sensor interface of the embedded board 1, and is used to collect the analog voltage output by the piezoelectric vibration sensor SEN0209, collect the vibration state of the vehicle terminal and the large-amplitude vibration state of the vehicle in real time, and transmit the information to the edge computing module 6 through the embedded board 1, and perform information fusion analysis as part of the multi-source sensor data to judge the vehicle status and fault status.

[0051] The hazardous chemical sensor 206 in this embodiment adopts the MQ-2-135-3-7-9 type hazardous gas sensor, which is electrically connected to the GPIO / Sensor interface of the embedded board 1. When the gas sensor detects hazardous gas, it outputs a digital signal. The embedded board 1 receives the signal output by the hazardous chemical sensor and transmits it to the display screen 401, and at the same time gives the driver a danger warning through the speaker 505.

[0052] The tire pressure sensor 207 in this embodiment uses the SP40 tire pressure sensor chip of Infineon Technologies, and is connected to the embedded board 1 through the Bluetooth module 301 in the wireless connection module 3, and the collected data is displayed in real time on the display screen 401, and an alarm is given through the speaker 205. The tire pressure data is stored in real time in the memory 508 through the embedded board 1. The tire pressure data is transmitted to the edge computing module 6 through the embedded board 1 within a certain period for modeling and analysis, and the maintenance data and maintenance status of the tire can be obtained to provide intelligent maintenance decisions.

[0053] The rain sensor 208 in this embodiment is electrically connected to the RS485 interface on the embedded board 1. The rain sensor 208 can adjust the operating speed of the wiper action by analyzing the rain precipitation and water intensity data information, effectively reducing the driver's driving troubles and improving the reliability of safe driving. The rain sensor 208 can also monitor the start and end time of precipitation. The rain sensor 208 here is existing technology and will not be repeated.

[0054] The distance sensor 209 in this embodiment is composed of an ultrasonic distance sensor and is electrically connected to the embedded board 1 through the GPIO / Sensor interface. The distance sensor 209 is installed at the front and rear of the vehicle to monitor the distance of objects in front and behind the vehicle in real time, and transmits the monitoring data to the edge computing module 6 through the embedded board 1 for multi-sensor data information fusion analysis. When the vehicle is traveling at a low speed of less than 30km / h, the distance sensor 209 will issue a real-time alarm for objects within 1 meter, and for objects within 2 meters of the front vehicle with a vehicle speed greater than 30km / h. Based on the multi-sensor data fusion analysis by the edge computing module 6, the vehicle can be emergency braked through the CAN bus of the embedded board 1 for an impending front vehicle collision accident.

[0055] The wireless connection module 3 in this embodiment includes a Bluetooth module 301 and a WiFi module 302, both of which are electrically connected to the embedded board 1 through a USB interface. Some sensors in the sensor module 2 are electrically connected to the embedded board 1 by means of the Bluetooth module 301. Due to the differences in vehicles, the different spaces for installing sensors, and the different sensor installation and connection methods used, some sensors are connected to the embedded board 1 by means of wireless Bluetooth. The Bluetooth module 301 is used to connect a variety of sensor devices. The WiFi module 302 is used for debugging and data viewing on the mobile phone APP side of the smart vehicle terminal.

[0056] The driver interaction module 4 in this embodiment includes a display screen 401 and an alarm button 402. The display screen 401 is electrically connected to the embedded board 1 through a Panel interface. The display screen 401 displays the current vehicle status information collected by the intelligent vehicle terminal; the alarm button 402 is electrically connected to the embedded board 1 through a GPIO / Sensor interface. The alarm button 402 is used for emergency alarming of the driver in a dangerous rescue state. After the driver triggers the alarm button 402, the alarm 402 command information is wirelessly transmitted to the remote monitoring center through the SIM card 507 in the peripheral module 5. The remote monitoring center views the image information of the alarm vehicle in real time through the multi-channel camera 502 in the peripheral module 5, and monitors the audio information of the alarm vehicle in real time through the microphone 502 in the peripheral module 5.

[0057] The peripheral module 5 in this embodiment includes a multi-channel camera 501, a microphone 502, an information broadcast screen 503, a GPS / Beidou dual-mode positioning module 504, a speaker 505, an OBD parser 506, a SIM card 507 and a memory 508.

[0058] The multi-channel camera 501 in this embodiment includes an analog camera and an IP network camera. The analog camera 501 is electrically connected to the embedded board 1 through an A / V interface, and the IP network camera is electrically connected to the embedded board 1 through an IPC interface; the image information monitored by the multi-channel camera 501 can be displayed in real time through the display screen 401 and stored in real time in the memory 508; at the same time, the remote monitoring center uses the SIM card 507 to monitor the image of the multi-channel camera 501 in real time through the mobile wireless network, and the mobile phone APP establishes a wireless connection with the smart vehicle terminal through the WiFi module 302 to retrieve the image information of the multi-channel camera 501 in real time. The multi-channel camera 501 also includes a camera for monitoring driver behavior, a front vehicle anti-collision camera, a passenger flow monitoring camera, and a reversing image camera. After the camera is turned on, the camera for monitoring driver behavior, the front vehicle anti-collision camera, the passenger flow monitoring camera, and the reversing image camera transmit the monitoring video data in real time to the edge computing module 6 through the embedded board 1 for real-time analysis.

[0059] The microphone 502 in this embodiment is electrically connected to the embedded board 1 via a MIC interface. The microphone 502 collects audio data together with the multi-channel cameras 501 in real time and stores the audio data in the memory 508 .

[0060] The information broadcast screen 503 in this embodiment is electrically connected to the embedded board 1 via an FFC interface. The embedded board 1 displays the vehicle location information, arrival information and advertising information in the memory 508 collected by the GPS / Beidou dual-mode positioning module 504 in real time on the information broadcast screen 503. The information broadcast screen 503 can adopt standard multi-size LED, LCD, and OLED display screens.

[0061] The GPS / Beidou dual-mode positioning module 504 in this embodiment is electrically connected to the embedded board 1 via an SMA interface. The GPS / Beidou dual-mode positioning module 504 adopts GPS and Beidou dual-mode positioning chips, and transmits the collected vehicle position positioning information to the display screen 401, the information broadcast screen 503 and the embedded board 1 in real time.

[0062] The speaker 505 in this embodiment is electrically connected to the embedded board 1 via the Amplifer interface, and the driver can broadcast voice information in real time through the microphone 502 and play it outward through the speaker 505 at the same time.

[0063] The OBD parser 506 in this embodiment is electrically connected to the embedded board 1 via the OBD interface. The OBD parser 506 reads the vehicle's OBD data through the vehicle CAN bus, and stores the read OBD data in the memory 508 through the embedded board 1, and transmits it to the edge computing module 6 at the same time. The edge computing module 6 can extract part of the OBD data for data fusion calculation according to data fusion needs.

[0064] The SIM card 507 in this embodiment is electrically connected to the embedded board 1 via a SIM card slot interface. The SIM card is a 4G SIM card and is backward compatible, and transmits data via a mobile wireless cellular network.

[0065] The memory in this embodiment is electrically connected to the embedded board 1 via a USB or SD interface. The memory 508 is a hard disk, an SD card or a USB flash drive and is used to store sensor module 2 data, video and audio data, OBD data, and location information.

[0066] Example 2

[0067] The edge computing module 6 in this embodiment uses Intel's NCS neural computing stick, which has a built-in OpenVINO SDK software development kit. The edge computing module 6 is electrically connected to the embedded board 1 via a USB or / and RS485 or / and FFC interface. The edge computing module 6 can be configured with one or more NCS neural computing sticks according to the needs of the system computing power.

[0068] The edge computing module 6 in this embodiment analyzes the multi-source module data information received by the embedded board 1, and the analysis results are then stored in the memory 508 by the embedded board 1.

[0069] The edge computing module 6 in this embodiment analyzes the multi-source module data information in the embedded board 1, and based on the analysis results and for specific application scenarios, can communicate with the driver through the display screen 401 and the speaker 505 of the intelligent vehicle terminal, and can also take emergency braking of the vehicle through the CAN bus of the embedded board 1. The edge computing module 6 can perform real-time edge computing on the data collected by various sensors, and fuse the multi-source sensor data according to the application scenario, set data weights, and clean invalid data.

[0070] It can be understood that the above specific description of the present invention is only used to illustrate the present invention and is not limited to the technical solutions described in the embodiments of the present invention. Those skilled in the art should understand that the present invention can still be modified or replaced by equivalents to achieve the same technical effects. As long as the needs of use are met, they are all within the protection scope of the present invention.

Claims

1. An intelligent vehicle-mounted terminal based on edge computing and multi-sensor interface, characterized in that: The system is composed of an embedded board (1), a sensor module (2), a wireless connection module (3), a driver interaction module (4), a peripheral module (5), and an edge computing module (6), wherein the sensor module (2), the wireless connection module (3), the driver interaction module (4), the peripheral module (5), and the edge computing module (6) are all electrically connected to the embedded board (1); The embedded board (1) is composed of an STM32 series embedded processor chip with an ARM core and a chip peripheral circuit. The IO port of the embedded processor chip is extended to lead to a GPIO interface, a serial port and an ADC interface. The chip peripheral circuit is provided with a plurality of peripheral circuit interfaces, which are specifically a USB interface, a LAN interface, a CAN interface, an SMA interface, an FFC interface, an RS485 interface, an RJ45 interface, a MIC interface, an Amplifer interface, an OBD interface, an SD interface, a SIM interface, a DC interface, an A / V interface, an IPC interface, a Panel interface and a GPIO / Sensor interface. The embedded processor chip is connected to the chip peripheral circuit through the IO port, and the GPIO interface extension leads to the LAN interface, SMA interface, FFC interface, RJ45 interface, Amplifer interface, SD interface, SIM interface, DC interface, IPC interface, and GPIO / Sensor interface connected to the chip peripheral circuit; the serial port extension leads to the USB interface, CAN interface, RS485 interface, OBD interface, IIC interface, and Panel interface connected to the chip peripheral circuit; the ADC interface extension leads to the MIC interface and A / V interface; The DC interface of the embedded board (1) is externally connected to a DC power supply with a voltage range of 8V-36V; The driver interaction module (4) comprises a display screen (401) and an alarm button (402), wherein the display screen (401) is electrically connected to the embedded board (1) via a panel interface, and the alarm button (402) is electrically connected to the embedded board (1) via a GPIO / Sensor interface; The display screen (401) displays the current vehicle status information collected by the intelligent vehicle terminal. The alarm button (402) is used to issue an emergency alarm to the driver in a dangerous rescue state. After the driver triggers the alarm button (402), the alarm instruction information is wirelessly transmitted to the remote monitoring center through the SIM card (507) in the peripheral module (5). The remote monitoring center views the image information of the alarm vehicle in real time through the multi-channel camera (501) in the peripheral module (5), and monitors the audio information of the alarm vehicle in real time through the microphone (502) in the peripheral module (5); The edge computing module (6) adopts Intel's NCS neural computing stick, which has a built-in OpenVINO SDK software development kit. The edge computing module (6) is electrically connected to the embedded board (1) via a USB or / and RS485 or / and FFC interface.

2. The intelligent vehicle-mounted terminal based on edge computing and multi-sensor interface according to claim 1, characterized in that: The wireless connection module (3) comprises a Bluetooth module (301) and a WiFi module (302), both of which are electrically connected to the embedded board (1) via a USB interface.

3. The intelligent vehicle-mounted terminal based on edge computing and multi-sensor interface according to claim 1, characterized in that: The peripheral module includes a multi-channel camera (501), a microphone (502), an information broadcast screen (503), a GPS / Beidou dual-mode positioning module (504), a speaker (505), an OBD parser (506), a SIM card (507) and a memory (508); The multi-channel camera (501) includes an analog camera and an IP network camera, the analog camera is electrically connected to the embedded board (1) via an A / V interface, and the IP network camera is electrically connected to the embedded board (1) via an IPC interface; the image information monitored by the multi-channel camera (501) can be displayed in real time via a display screen (401) and stored in real time in a memory (508); at the same time, the remote monitoring center uses a SIM card (507) to monitor the images of the multi-channel camera (501) in real time via a mobile wireless network, and a mobile phone APP establishes a wireless connection with an intelligent vehicle-mounted terminal via a WiFi module (302) to retrieve the image information of the multi-channel camera (501) in real time; The multi-channel camera (501) also includes a camera for monitoring driver behavior, a front vehicle collision avoidance camera, a passenger flow monitoring camera, and a reversing image camera. After the camera is turned on, the camera for monitoring driver behavior, the front vehicle collision avoidance camera, the passenger flow monitoring camera, and the reversing image camera transmit the monitoring video data in real time to the edge computing module (6) through the embedded board (1) for real-time analysis; The microphone (502) is electrically connected to the embedded board (1) via a MIC interface; the microphone (502) and the multi-channel cameras (501) collect audio data in real time and store the data in a memory (508); The information broadcast screen (503) is electrically connected to the embedded board (1) via an FFC interface, and the embedded board (1) displays the vehicle position information, arrival information and advertising information in the memory (508) collected by the GPS / Beidou dual-mode positioning module (504) on the information broadcast screen (503) in real time; The GPS / Beidou dual-mode positioning module (504) is electrically connected to the embedded board (1) via an SMA interface, the GPS / Beidou dual-mode positioning module (504) uses GPS and Beidou dual-mode positioning chips, and transmits the collected vehicle position positioning information to the display screen (401), the information broadcast screen (503) and the embedded board (1) in real time; The speaker (505) is electrically connected to the embedded board (1) via an Amplifer interface, and the driver can broadcast voice information in real time through the microphone (502) and simultaneously play it outward through the speaker (505); The OBD parser (506) is electrically connected to the embedded board (1) via an OBD interface, and the OBD parser (506) reads the OBD data of the vehicle via the vehicle CAN bus, stores the read OBD data in a memory (508), and transmits it to the edge computing module (6); The SIM card (507) is electrically connected to the embedded board (1) via a SIM card slot interface; the SIM card (507) is a 4G SIM card and is backward compatible, and transmits data via a mobile wireless cellular network; The memory (508) is electrically connected to the embedded board (1) via a USB or SD interface, and the memory (508) is one of a hard disk, an SD card or a U disk.

4. The intelligent vehicle-mounted terminal based on edge computing and multi-sensor interface according to claim 1, characterized in that: The sensor module (2) comprises at least a temperature and humidity sensor (201), an angle sensor (202), a vibration sensor (203), a multi-axis gyroscope (204), an acceleration sensor (205), a hazardous chemical sensor (206), a tire pressure sensor (207), a rainfall sensor (208) and a distance sensor (209); The temperature and humidity sensor (201) adopts a DHT11 temperature and humidity sensor module of Risym Company, and the data interface of the temperature and humidity sensor (201) is electrically connected to the GPIO / Sensor interface of the embedded board (1); The multi-axis gyroscope (204) and the acceleration sensor (205) both use the MPU6050 chip of InvenSense as the core chip, and the MPU6050 chip and the embedded board (1) are electrically connected and communicate with each other via an IIC interface; The vibration sensor (203) is a piezoelectric vibration sensor SEN0209, which is electrically connected to the GPIO / Sensor interface of the embedded board (1) and is used to collect the analog voltage output by the piezoelectric vibration sensor SEN0209, collect the vibration state of the vehicle terminal and the large-amplitude vibration state of the vehicle in real time, and transmit the data to the edge computing module (6) through the embedded board (1); The hazardous chemical sensor (206) is a MQ-2-135-3-7-9 type hazardous gas sensor, which is electrically connected to the GPIO / Sensor interface of the embedded board (1). When the gas sensor detects hazardous gas, it outputs a digital signal. The embedded board (1) receives and transmits the signal output by the hazardous chemical sensor (206) to the display screen (401), and at the same time gives the driver a danger prompt through the speaker (505); The tire pressure sensor (207) uses an SP40 tire pressure sensor chip from Infineon Technologies, and is connected to the embedded board (1) via a Bluetooth module (301), and displays the collected data in real time on a display screen (401), and gives an alarm via a speaker (505), and the tire pressure data is stored in real time in a memory (508) via the embedded board (1); The distance sensor (209) is composed of an ultrasonic distance sensor and is electrically connected to the embedded board (1) via a GPIO / Sensor interface. The distance sensor (209) is installed at the front and rear of the vehicle to monitor the distance of objects in front and behind the vehicle in real time, and transmits the monitoring data to the edge computing module (6) via the embedded board (1).

5. An intelligent vehicle-mounted terminal based on edge computing and multi-sensor interface as described in any one of claims 1 to 4, characterized in that: The embedded board (1) receives data information output by the sensor module (2), the wireless connection module (3), the driver interaction module (4) and the peripheral module (5), and fuses the data information.

Images