Healthy driving system and method based on Internet of Vehicles

Abstract

The invention relates to a healthy driving system and method, in particular to a healthy driving system and method based on the Internet of Vehicles. The problems that an existing anti-fatigue safe driving early warning instrument is single in function and can only mechanically conduct voice reminding are solved. The system comprises a driver drunk driving condition acquisition and analysis module, a fatigue driving condition acquisition and analysis module, a position acquisition module and a data visualization module. The invention belongs to the technical field of intelligent driving.

Description

Technical Field

[0001] The present invention relates to a healthy driving system and method, belonging to the technical field of intelligent driving. Background Art

[0002] Through the new generation of information and communication technology, the Internet of Vehicles realizes all-round network connection between vehicles and cloud platforms, vehicles and vehicles, vehicles and roads, vehicles and people, and vehicles, mainly realizing the "three-network integration", that is, integrating the in-vehicle network, inter-vehicle network and in-vehicle mobile Internet. The Internet of Vehicles uses sensor technology to perceive the status information of vehicles, and uses wireless communication networks and modern intelligent information processing technologies to realize intelligent management of traffic, intelligent decision-making of traffic information services and intelligent control of vehicles.

[0003] The current vehicle healthy driving system is limited to the interior of the car and simply interacts with the driver. For example, driving comfort can be improved by monitoring the driver's posture and behavior habits. The in-car intelligent health monitoring system can provide corresponding driving adjustment suggestions to help the driver maintain a good posture and attention and reduce fatigue. The concept of vehicle networking has not been introduced, resulting in untimely handling of driver fatigue and increasing the risk of traffic accidents.

[0004] After market research, there are currently three main methods for checking drunk driving in China, breath alcohol testing, blood alcohol testing and infrared alcohol testing. Among them, breath alcohol testing is the most commonly used method at the scene of drunk driving, but the accuracy rate is low, and it is the primary method for detecting drunk driving; blood alcohol testing is the main method for detecting drunk driving, but the testing time is long and the operation is complicated, usually It takes seven working days to get the results; infrared alcohol testing has been widely used abroad, but the cost is relatively high.

[0005] These detection methods have a common problem: personnel are required to intercept and inspect cars one by one on the main routes of vehicles. This not only makes it impossible to detect some drivers with dangerous driving styles, but also makes drivers take chances and have a low cost of breaking the law, which is also the main cause of traffic accidents.

[0006] It is difficult to determine whether a driver is driving fatigued based on a certain basis. Through research, we learned that for their own safety, many drivers hope to have a device that can help them solve the problem of driving fatigued. However, the common problem of products such as anti-fatigue safety driving warning devices available on the market is that the functions are too single. They can only mechanically give voice reminders. After a long time, the driver gets used to the reminder tone, and the reminder function is not very practical.

[0007] Defects and shortcomings of existing technology:

[0008] 1. There are three main methods for drunk driving detection technology: breath alcohol test, blood alcohol test and infrared alcohol test. Among them, breath alcohol test is the most commonly used method at the scene of drunk driving, but the accuracy rate is low, and it is the primary method for detecting drunk driving; blood alcohol test is the main method for detecting drunk driving, but the test time is long and the operation is complicated, usually it takes seven working days to get the result; infrared alcohol test has been widely used abroad, but the cost is relatively high;

[0009] 2. The existing drunk driving detection requires personnel to intercept cars one by one on the main driving routes for inspection. This not only makes some drivers with dangerous driving styles fail to be detected, but also makes drivers have a fluke mentality, which is also the main cause of traffic safety accidents.

[0010] 3. Currently, it is difficult to determine whether a driver is driving while fatigued, which leads to a high probability that the driver may be in danger. Summary of the invention

[0011] The present invention aims to solve the problem that the existing anti-fatigue safety driving warning instrument has a single function and can only provide mechanical voice reminders, and further proposes a healthy driving system and method based on the Internet of Vehicles.

[0012] The technical solution adopted by the present invention to solve the above problems is: a healthy driving system based on the Internet of Vehicles described in the present invention includes:

[0013] Driver's drunk driving situation acquisition and analysis module, fatigue driving situation acquisition and analysis module, location acquisition module and data visualization block;

[0014] The driver's drunk driving situation acquisition and analysis module consists of an MQ3 alcohol concentration sensor, an ADS1115 digital-to-analog converter, and a Raspberry Pi 4B. The MQ3 sensor senses the alcohol concentration information and generates an analog voltage. The ADS1115 converts the analog voltage into a digital signal and transmits it to the Raspberry Pi. The voltage value is compared with the voltage threshold during drunk driving through a python file to obtain the driver's drunk driving probability and store it in a document in real time and refresh it in real time.

[0015] The fatigue driving condition acquisition and analysis module uses a USB camera to acquire images, with a maximum resolution of 1080P and a maximum frame rate of 30FPS. The camera is connected to the USB port of the Raspberry Pi and placed above the dashboard inside the car. On the Raspberry Pi, the Python imutil package is used to capture real-time video streams, and then each frame of the video stream is acquired for image processing;

[0016] The location acquisition module obtains the real-time location information of the vehicle through the GS05B GPS locator;

[0017] In the data visualization module, the Raspberry Pi is connected to the Alibaba Cloud IoT platform through the MQTT protocol and transmits the acquired and processed data to the Alibaba Cloud IoT platform.

[0018] The steps of a healthy driving method based on vehicle networking described in the present invention include:

[0019] Step 1. The MQ3 sensor senses the alcohol concentration and generates an analog voltage. The ADS1115 converts the analog voltage into a digital signal and transmits it to the Raspberry Pi. The voltage value is compared with the voltage threshold during drunk driving through the python file to obtain the driver's drunk driving probability and store it in the document in real time and refresh it in real time.

[0020] Step 2: Raspberry Pi connects to Alibaba Cloud IoT platform through MQTT protocol, transmits the acquired and processed data to Alibaba Cloud IoT platform, and purchases Centos cloud server and domain name;

[0021] Step 3: Use Alibaba Cloud IOT Studio to develop a visualization webpage to display the drunk driving situation, fatigue driving situation, and vehicle location of each vehicle in real time.

[0022] Furthermore, in step 1, a USB camera is used to acquire images, and the driver's facial image is processed. The eye and mouth opening degrees are detected in real time by calculating the EAR and MAR parameters. The driver's fatigue state is judged by the number of times the driver closes his eyes and yawns in multiple consecutive frames of images, and the data is uploaded in real time.

[0023] Furthermore, in step 1, the real-time location information of the vehicle is obtained through the GS05B GPS locator and uploaded to the cloud in real time.

[0024] Furthermore, in step 3, the cloud server is connected to the IoT platform to obtain data, and a WeChat applet is developed through the JavaScript language so that the driver's information can be displayed on the mobile terminal in real time. When the probability of the driver driving under the influence of alcohol or fatigue is high, the applet will issue a warning.

[0025] Furthermore, in step 3, the API of the locator is called to realize the real-time display and tracking of the location information on the web page and the applet. At the same time, the locator can detect the speed of the car and determine whether the drunk driving vehicle is speeding.

[0026] The beneficial effects of the present invention are:

[0027] 1. The present invention can detect drivers in a targeted manner through experimental simulation, eliminate the interference of passengers drinking, and does not force the car to be locked. Instead, the driver's information is uploaded to a webpage through the Internet of Vehicles to facilitate targeted police dispatch, which will not affect the travel experience and normal life order of the driver and passengers;

[0028] 2. Improve efficiency: The present invention combines automobile driving with the Internet of Things, and can intuitively obtain the drunk driving situation of vehicles in the city through the web page so as to dispatch police in a targeted manner. Compared with the previous method of stopping vehicles at intersections for monitoring, this project plan reduces the waste of human resources.

[0029] 3. Reduce the probability of dangerous driving: The present invention reduces dangerous driving behaviors through family constraints. This project provides the driver's status to his family through the WeChat applet. On the one hand, when the driver drives dangerously, his family can remind the driver to stop the dangerous behavior. On the other hand, the driver will restrain his behavior because of his sense of family responsibility, thereby reducing the occurrence of dangerous driving behaviors;

[0030] 4. High accuracy of drunk driving monitoring: The present invention simulates the movement of alcohol molecules through SPH smooth particle fluid dynamics method and Blender modeling software to obtain the diffusion law of alcohol vapor, thereby determining the most reasonable sensor placement position, eliminating the interference of non-driver drinking on the drunk driving monitoring system;

[0031] 5. The present invention processes and analyzes facial data through opencv, and calculates the eye closure and mouth closure of the human body through the 68 facial feature points in dlib to judge the fatigue driving situation with high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a flow chart of the healthy driving system based on the Internet of Vehicles;

[0033] Figure 2 This is a simulation of alcohol particle emission;

[0034] Figure 3 This is a model diagram of alcohol particles diffusing in a car;

[0035] Figure 4 It is a graph showing the number of emitted particles from the driver's seat and the co-pilot's seat changing with time;

[0036] Figure 5 It is a website page map. DETAILED DESCRIPTION

[0037] Specific implementation method 1: Figure 1 As shown, a healthy driving system based on the Internet of Vehicles includes a driver's drunk driving situation acquisition and analysis module, a fatigue driving situation acquisition and analysis module, a location acquisition module, and a data visualization module;

[0038] The driver's drunk driving situation acquisition and analysis module consists of an MQ3 alcohol concentration sensor, an ADS1115 digital-to-analog converter, and a Raspberry Pi 4B. The MQ3 sensor senses the alcohol concentration information and generates an analog voltage. The ADS1115 converts the analog voltage into a digital signal and transmits it to the Raspberry Pi. The voltage value is compared with the voltage threshold during drunk driving through a python file to obtain the driver's drunk driving probability and store it in a document in real time and refresh it in real time.

[0039] The fatigue driving condition acquisition and analysis module uses a USB camera to acquire images, with a maximum resolution of 1080P and a maximum frame rate of 30FPS. The camera is connected to the USB port of the Raspberry Pi and placed above the dashboard inside the car. On the Raspberry Pi, the Python imutil package is used to capture real-time video streams, and then each frame of the video stream is acquired for image processing;

[0040] The location acquisition module obtains the real-time location information of the vehicle through the GS05B GPS locator;

[0041] In the data visualization module, the Raspberry Pi is connected to the Alibaba Cloud IoT platform through the MQTT protocol and transmits the acquired and processed data to the Alibaba Cloud IoT platform.

[0042] Among them, the movement of alcohol molecules is simulated by SPH smooth particle fluid dynamics method and Blender modeling software to obtain the diffusion law of alcohol vapor, so as to determine the most reasonable sensor placement position;

[0043] The Smoothed Particle Hydrodynamics (SPH) method is used to obtain the diffusion law of alcohol vapor through the movement of alcohol molecules, and the 3D modeling software Blender is used to simulate the scene; Blender provides a particle system and related solvers, and the operation is relatively simple. With other functions, it can intuitively show the gas concentration at each position;

[0044] The following simulation experiment is conducted: in a rectangular container, the rectangular container is strip-shaped and contains a particle emission source; at the same time, several "detectors" are set up, each of which can detect the number of particles within a certain radius around it (instead of concentration), such as Figure 2As shown in the figure, all particle motion parameters are consistent with those of gas molecules. All particles are released at once (the more the total number, the more accurate the result, in this case 50,000). The relevant data are recorded and sorted out, and the following conclusions are drawn: For the detector closer to the emission source, its concentration curve shows a "low-high-low" trend. This is because the concentration in the container has not reached a state of equilibrium in the early stage, resulting in a high detection result; the value is stable in the later stage. Then, a simulation is performed in a car model (Mercedes Benz GLC Class 2016). In order to simulate the breathing of a drunk driver, the particle emission source is now changed to a plane, placed in front of the driver's seat, and set to periodically "spray" a certain number of particles. The total number of particles is temporarily set to 50,000. The first 600 frames (20s) are simulated. The effect diagram is as follows Figure 3 As shown, the detectors are placed in front of the driver and co-pilot seats respectively, so that the concentration at each position can be intuitively seen. To avoid unnecessary errors, the detection ranges of the two are consistent and both are suspended. The recorded data are as follows Figure 4 As shown by Figure 4 The following conclusions were drawn: (a) When the driver was drunk, the curve showed a sharp upward trend at around 10 seconds; at the same time, it was found that the readings of the detectors on the driver and the co-pilot were quite different at 10 seconds to 20 seconds; (b) When the co-pilot was drunk, the readings of the two detectors were opposite to the above situation; (c) It was observed that in 15-20 seconds, the red / purple curve had a stable trend; the green / blue curve had a rapid upward trend. Combined with the first experiment, the trend of the curve after 20 seconds can be inferred. The above experiments proved that placing the alcohol sensor on the steering wheel can effectively determine whether the driver is driving under the influence of alcohol and will not be disturbed by the drinking of others, which also verified the feasibility of this project.

[0045] Among them, the OpenCV computer vision library, Dlib machine learning library and Python development language are used to process the driver's facial image. The EAR and MAR parameters are calculated to detect the eye and mouth opening degrees in real time, and the driver's fatigue state is judged by the number of times the driver closes his eyes and yawns in multiple consecutive frames of images, and the data is uploaded in real time; the eye opening degree is measured using the EAR parameter, and the eye opening state is judged by calculating the value of the EAR; when the human eyes are open, the EAR fluctuates around a certain value, and when the eyes are closed, the EAR drops rapidly, and theoretically it will be close to zero, but the face detection model is not so accurate; so we believe that when the EAR is below a certain threshold, the eyes are in a closed state; to detect blinking times, you need to set the number of consecutive frames of the same blink; the blinking speed is relatively fast, usually 1 to 3 frames to complete the blinking action; by setting the thresholds of the number of eye closings and eye closing time within 1 minute, determine whether the person is tired; (set the threshold to 0.2, but each person's eye size ratio is different, the average value calculation method should be adopted - the threshold is obtained by: first collect 30 data, and take the average value as the default value; for the accuracy of the data, you should look straight at the camera when collecting data); double threshold method yawn detection - that is, detect the inner contour: combine the mouth opening degree and the mouth opening time to determine whether yawning (when the mouth opening degree in more than 3 consecutive frames of images is greater than 0.5, it is judged as a yawn), when the number of yawns within 5 minutes reaches the threshold, the system determines fatigue.

[0046] Among them, by calling the open API of the locator's open website Autohome Cloud Platform, real-time display and tracking of location information can be achieved on web pages and mini-programs. At the same time, the locator can detect the speed of the car and determine whether the drunk driving vehicle is speeding.

[0047] Among them, we purchased Centos cloud servers and domain names. We developed a visual webpage through Alibaba Cloud IOT Studio to display the drunk driving situation, fatigue driving situation and vehicle location of each vehicle in real time for easy viewing; we connected the cloud server to the IoT platform to obtain data, and developed WeChat mini-programs through JavaScript language so that the driver's information can be displayed in real time on the mobile terminal. When the probability of a driver driving under the influence of alcohol or fatigue is high, the mini-program will remind the driver's family.

[0048] Specific implementation method 2: Figures 1 to 5 As shown, a healthy driving method based on vehicle networking includes the following specific steps:

[0049] Step 1. The MQ3 sensor senses the alcohol concentration information and generates an analog voltage. The ADS1115 converts the analog voltage into a digital signal and transmits it to the Raspberry Pi. The voltage value is compared with the voltage threshold during drunk driving through the python file to obtain the driver's drunk driving probability and store it in the document and refresh it in real time. Use a USB camera to obtain images, process the driver's face image, calculate the EAR and MAR parameters to detect the eye and mouth opening degrees in real time, and judge the driver's fatigue state by the number of times the driver closes his eyes and yawns in multiple consecutive frames of images, and upload the data in real time. Obtain the real-time location information of the vehicle through the GS05B GPS locator and upload it to the cloud in real time.

[0050] Step 2: Raspberry Pi connects to Alibaba Cloud IoT platform through MQTT protocol, transmits the acquired and processed data to Alibaba Cloud IoT platform, and purchases Centos cloud server and domain name;

[0051] Step 3: Develop a visualization web page through Alibaba Cloud IOT Studio to display the drunk driving situation, fatigue driving situation, and vehicle location of each vehicle in real time; connect the cloud server to the IoT platform to obtain data, and develop a WeChat applet through JavaScript language so that the driver's information can be displayed in real time on the mobile terminal. When the probability of the driver driving under the influence or driving while fatigued is high, the applet will issue a warning; call the locator's API to realize real-time display and tracking of location information on the web page and applet. At the same time, the locator can detect the speed of the car and determine whether the drunk driving vehicle is speeding.

[0052] In step 1, a USB camera is used to acquire images, and the driver's facial image is processed. The eye and mouth opening degrees are detected in real time by calculating the EAR and MAR parameters. The driver's fatigue state is judged by the number of times the driver closes his eyes and yawns in multiple consecutive frames of images, and the data is uploaded in real time. The real-time location information of the vehicle is obtained through the GS05B GPS locator and uploaded to the cloud in real time.

[0053] The above is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent replacement and improvement made to the above embodiments without departing from the content of the technical solution of the present invention, based on the technical essence of the present invention, within the spirit and principles of the present invention, still fall within the protection scope of the technical solution of the present invention.

Claims

1. A healthy driving system based on Internet of Vehicles, characterized in that: It includes the driver's drunk driving situation acquisition and analysis module, fatigue driving situation acquisition and analysis module, location acquisition module and data visualization module; The driver's drunk driving situation acquisition and analysis module consists of an MQ3 alcohol concentration sensor, an ADS1115 digital-to-analog converter, and a Raspberry Pi 4B. The MQ3 sensor senses the alcohol concentration information and generates an analog voltage. The ADS1115 converts the analog voltage into a digital signal and transmits it to the Raspberry Pi. The voltage value is compared with the voltage threshold during drunk driving through a python file to obtain the driver's drunk driving probability and store it in a document in real time and refresh it in real time. The fatigue driving condition acquisition and analysis module uses a USB camera to acquire images, with a maximum resolution of 1080P and a maximum frame rate of 30FPS. The camera is connected to the USB port of the Raspberry Pi and placed above the dashboard inside the car. On the Raspberry Pi, the Python imutil package is used to capture real-time video streams, and then each frame of the video stream is acquired for image processing; The location acquisition module obtains the real-time location information of the vehicle through the GS05B GPS locator; In the data visualization module, the Raspberry Pi is connected to the Alibaba Cloud IoT platform through the MQTT protocol and transmits the acquired and processed data to the Alibaba Cloud IoT platform.

2. A healthy driving method based on Internet of Vehicles, characterized in that: The specific steps include: Step 1. The MQ3 sensor senses the alcohol concentration and generates an analog voltage. The ADS1115 converts the analog voltage into a digital signal and transmits it to the Raspberry Pi. The voltage value is compared with the voltage threshold during drunk driving through the python file to obtain the driver's drunk driving probability and store it in the document in real time and refresh it in real time. Step 2: Raspberry Pi connects to Alibaba Cloud IoT platform through MQTT protocol, transmits the acquired and processed data to Alibaba Cloud IoT platform, and purchases Centos cloud server and domain name; Step 3: Use Alibaba Cloud IOT Studio to develop a visualization webpage to display the drunk driving situation, fatigue driving situation, and vehicle location of each vehicle in real time.

3. A healthy driving method based on vehicle networking according to claim 2, characterized in that: In step 1, a USB camera is used to acquire images, and the driver's facial image is processed. The eye and mouth opening degrees are detected in real time by calculating the EAR and MAR parameters. The driver's fatigue state is judged by the number of times the driver closes his eyes and yawns in multiple consecutive frames of images, and the data is uploaded in real time.

4. A healthy driving method based on vehicle networking according to claim 2 or 3, characterized in that: In step 1, the real-time location information of the vehicle is obtained through the GS05B GPS locator and uploaded to the cloud in real time.

5. The healthy driving method based on the Internet of Vehicles according to claim 2, characterized in that: In step 3, the cloud server is connected to the IoT platform to obtain data, and a WeChat applet is developed using JavaScript so that the driver's information can be displayed in real time on the mobile terminal. When the probability of the driver driving under the influence of alcohol or fatigue is high, the applet will issue a warning.

6. A healthy driving method based on vehicle networking according to claim 2 or 5, characterized in that: In step 3, the API of the locator is called to realize the real-time display and tracking of the location information on the web page and the applet. At the same time, the locator can detect the speed of the car and determine whether the drunk driving vehicle is speeding.

Images