Intelligent traffic accident early warning system based on ESP32-S3 chip

Abstract

The invention relates to the technical field of intelligent traffic, and discloses an intelligent traffic accident early warning system based on an ESP32-S3 chip, which comprises a traffic system, and the traffic system comprises a road monitoring subsystem, an analysis and calculation subsystem and a vehicle early warning subsystem. According to the intelligent traffic accident early warning system based on the ESP32-S3 chip, a traffic system is arranged, the system functions comprise monitoring, tracking, analysis, early warning and braking, firstly, the monitoring function is used for carrying out all-weather monitoring on accident-prone sections, data are obtained through various sensors on different vehicles, secondly, the tracking function and the analysis function are inseparable, and the monitoring function is used for monitoring the accident-prone sections in an all-weather mode. Data acquired by various sensors on a vehicle are firstly transmitted to a computer of the system and are analyzed and judged by the computing system, normal or abnormal marking is carried out on the vehicle or pedestrians on a road, and the system emphatically monitors the motion condition of the abnormal marking, acquires information such as the speed and the like of the abnormal marking and displays the information on the road. These information is then transmitted into a computing system for special processing.

Description

Technical Field

[0001] The present invention relates to the field of intelligent traffic technology, and in particular to an intelligent traffic accident warning system based on an ESP32-S3 chip. Background Art

[0002] According to statistics from the World Health Organization, approximately 1.19 million people die in road traffic accidents each year, and another 20 to 50 million people suffer non-fatal injuries due to traffic accidents, many of whom become disabled. Road traffic accidents have become the leading cause of death for children and young people aged 5 to 29. 92% of road deaths worldwide occur in low-income or middle-income countries, while the number of vehicles in these countries only accounts for about 60% of the world's total. Among all road traffic deaths, more than half are vulnerable road users, including pedestrians, cyclists and motorcyclists. The economic losses caused by traffic accidents account for 3% of the GDP of most countries. This shows that the potential hazards of traffic accidents are huge. In addition, due to its large population and number of motor vehicles, my country has become one of the countries with the most complex traffic conditions in the world.

[0003] With the development of society and the advancement of science and technology, people are working hard to reduce the harm of traffic accidents in all aspects. At present, visual blind spots are still one of the important causes of traffic accidents. In addition, there are many uncertain emergencies in road traffic, such as pedestrians or electric vehicles suddenly rushing out from behind obstacles, making it difficult for vehicles to brake in time even with intelligent auxiliary control, different vehicles being unable to avoid in time due to obstructions, or people or objects being in the space occupied by trucks turning, etc., which are all difficult problems for current intelligent driving to solve. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the shortcomings of the prior art, the present invention provides an intelligent traffic accident warning system based on the ESP32-S3 chip, which has the advantages of fully grasping the road conditions and facilitating analysis, calculation and warning, and solves the problem that visual blind spots are still one of the important causes of traffic accidents. In addition, there are many uncertain emergencies in road traffic, such as pedestrians or electric vehicles suddenly rushing out from behind obstacles, making it difficult for vehicles to brake in time even with intelligent auxiliary control, different vehicles being unable to avoid in time due to obstructions, or people or objects being in the space occupied by trucks when turning.

[0006] (II) Technical solution

[0007] In order to achieve the above-mentioned purpose of comprehensively understanding road conditions and facilitating analysis, calculation and early warning, the present invention provides the following technical solutions: an intelligent traffic accident early warning system based on the ESP32-S3 chip, comprising a traffic system, wherein the traffic system comprises a road monitoring subsystem, an analysis and calculation subsystem and a vehicle early warning subsystem;

[0008] The road monitoring subsystem: Modern vehicles are equipped with a variety of on-board sensors: such as cameras, radars and lidars. As part of the entire system, the sensors on the moving vehicle can be connected through the ESP chip and transmit the data to the computing system for comprehensive analysis;

[0009] The analysis and calculation subsystem: The 5G network of the road can cover the entire road section through multiple routers to ensure communication. In this subsystem, the calculation and analysis are performed by a powerful computer;

[0010] The vehicle warning subsystem: the vehicle's sensors can be integrated into the road monitoring subsystem, and the on-board sensors can be connected to the on-board screen to visualize and display the predicted road conditions on the screen, reminding the driver to avoid in advance.

[0011] Preferably, the road monitoring subsystem benefits from the low power consumption and high efficiency of the ESP32-S3 chip. The vehicle sensors can transmit real-time information after connecting to the network of the designated road section and send the data to the computing system for processing. The computing system integrates the data obtained from different vehicles and can simulate the road conditions more completely.

[0012] Preferably, the analysis and computing subsystem uses a large amount of labeled data to train the computer, and through deep learning, it has the ability to judge abnormal road traffic conditions. Relying on the strong computing power of the computer, when an abnormal situation occurs, the system will make real-time predictions and simulations of the "abnormality". In addition, the analysis and computing subsystem is also the hub of information transmission, and through the ESP32-S3 chip connected to the 5G network, efficient communication with the vehicle is achieved.

[0013] Preferably, the ESP chip in the vehicle warning subsystem can also be connected to key parts of the vehicle such as the steering wheel and brakes. When receiving the warning information and the driver fails to respond in time, the ESP chip can perform emergency braking on the vehicle through pre-burned instructions to avoid accidents.

[0014] Preferably, the road monitoring subsystem includes hardware connection, program control and data transmission;

[0015] The hardware connection first requires physically connecting the ESP32-S3 chip to the vehicle's communication interface and determining the vehicle's network communication protocol, including connecting to the vehicle's CAN bus, Ethernet or other communication bus;

[0016] The program control first needs to develop or obtain a suitable driver so that the ESP32-S3 chip can communicate. At the same time, it is necessary to integrate necessary middleware, such as TCP / IP stack, CANopen stack, etc., to support network communication protocols. Then, the application code is written. Appropriate debugging tools are used to test and verify the communication function to implement the specific logic of communicating with the vehicle network. To ensure communication security, the communication needs to be encrypted. The ESP32-S3 chip supports Flash encryption and secure boot based on the RSA algorithm, which can protect the vehicle from malicious attacks and data tampering.

[0017] During the data transmission, the ESP32-S3 chip is first connected to the analysis and computing subsystem in the cloud through the vehicle network. Since different sensors obtain different types of data, such as cameras obtaining photos and images, lidar obtaining three-dimensional point clouds, infrared sensors obtaining target thermal image information, etc., the obtained data is packaged according to different needs. Common formats such as JSON, XML, etc. contain sensor data and related identifiers or metadata.

[0018] Preferably, the analysis and calculation subsystem includes vehicle big data processing and storage, road anomaly detection model construction and intelligent matching of driving risk avoidance plans;

[0019] When the vehicle big data is processed and stored, the data uploaded first will be marked and processed first, and data from different sources will be marked differently, and the location of the information source will be determined in combination with GPS or Beidou hazard positioning system. These data will be continuously updated within the monitoring range of the system, and will be stored in the storage space as a record after leaving the monitoring range. The storage space also needs to record the operation log of the system and enter the monitoring data into the storage space. Due to limited storage space, the stored data needs to be cleaned or exported every once in a while;

[0020] The road anomaly detection model construction is a machine learning method that achieves prediction purposes by training artificial intelligence. The output is predicted by training artificial intelligence. Artificial intelligence simulates the neural network structure of the biological brain through deep learning so that the computer can perform specific tasks.

[0021] The driving risk avoidance solution intelligently matches various types of data uploaded by sensors on different vehicles, based on different locations, and relies on powerful computing power for traceability. The subsystem can simulate the conditions of the entire road, establish a real-time 3D model, and simulate road conditions in real time. The trained artificial intelligence is responsible for monitoring the model.

[0022] Preferably, the vehicle warning subsystem includes intelligent warning and emergency braking;

[0023] The intelligent warning is as follows: under normal driving conditions, the vehicle screen will display the recommended braking distance, the distance to the vehicle in front, and the braking warning information. The ESP32-S3 chip connects to the cloud computer through the 5G network to achieve rapid transmission and reception of real-time calculation results. When the analysis and calculation subsystem conveys the warning information, the on-board ESP32-S3 chip will connect to the on-board sensor, combine the computing power of the cloud computer to simulate the real-time road conditions, and project it onto the on-board screen. At the same time, other sensors outside the vehicle connected to the ESP32-S3 chip will also transmit the collected data back to the ESP32-S3 chip, which will be visualized after being processed by the chip and displayed on the vehicle screen.

[0024] During emergency braking, the ESP32-S3 chip compares the size with three thresholds through a pre-written program, triggering different warning effects in different situations. Among the data transmitted, there are two key data: one is the braking distance required for normal braking, which is obtained through the friction and other sensors equipped on the vehicle and calculated by the ESP32-S3 chip. This data is set as the first threshold and displayed on the vehicle display. When the distance between the vehicle and the abnormal element is lower than this value, the first level of alarm will be triggered, with a slight sound and driving advice broadcast to prompt the driver to respond. The alarm will continue until the collision with the abnormal element can be completely avoided. The other data is the calculated real-time minimum braking distance.

[0025] (III) Beneficial effects

[0026] Compared with the prior art, the present invention provides an intelligent traffic accident warning system based on the ESP32-S3 chip, which has the following beneficial effects:

[0027] The intelligent traffic accident warning system based on the ESP32-S3 chip sets up a traffic system. The system functions include monitoring, tracking, analysis, warning and braking. First, the monitoring function is to conduct all-weather monitoring of accident-prone areas and obtain data through various sensors on different vehicles. Secondly, the tracking and analysis functions are inseparable. The data obtained by various sensors on the vehicle are first transmitted to the system's computer, and the computing system analyzes and judges it, and marks the vehicles or pedestrians on the road as "normal" or "abnormal". The system focuses on monitoring the movement of the "abnormal" marks and obtains information such as their speed. This information is then transmitted to the computing system for special processing and combined with other vehicles on the entire road section. The computer analyzes and predicts the movement trajectory of these information objects and future road conditions, that is, it predicts and simulates possible accidents such as collisions, generates relevant warning information and sends it in time. The warning information is transmitted through the ESP32-S3 chip connected to the vehicle. When there is sufficient reaction time, the warning information will also be transmitted to the driver through the on-board screen and audio. If the driver cannot react in time, the system can perform emergency braking. Thanks to its rich peripheral interfaces, the ESP32-S3 chip can pre-burn relevant instructions and link and control key components of the vehicle (such as brakes and steering wheels), so that it can react in time when receiving the transmission message to avoid accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a schematic diagram of the system of the present invention;

[0029] Figure 2 It is a flow chart of the system of the present invention. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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.

[0031] See also Figure 1-2 , an intelligent traffic accident warning system based on the ESP32-S3 chip, including a traffic system, which includes a road monitoring subsystem, an analysis and calculation subsystem, and a vehicle warning subsystem;

[0032] Road monitoring subsystem: Modern vehicles are equipped with a variety of on-board sensors: such as cameras, radars and lidars. As part of the entire system, the sensors on the moving vehicle can be connected through the ESP chip and transmit the data to the computing system for comprehensive analysis;

[0033] Analysis and computing subsystem: It plays the role of the brain. The 5G network of the road can cover the entire road section through multiple routers to ensure communication. In this subsystem, the computing analysis is performed by a powerful computer.

[0034] Vehicle warning subsystem: The vehicle's sensors can be integrated into the road monitoring subsystem. The assisted driving function is also a key component. Considering the driver's needs, the on-board sensors can be connected to the on-board screen to visualize the predicted road conditions and display them on the screen, reminding the driver to avoid them in advance.

[0035] In the case implementation, the road monitoring subsystem benefits from the low power consumption and high efficiency of the ESP32-S3 chip. Vehicle sensors can transmit real-time information after connecting to the network of the designated road section and send the data to the computing system for processing. The computing system integrates the data obtained by different vehicles and can simulate the road conditions more completely. When the data obtained by the road sensors can be integrated at the same time, the system's capabilities will be further enhanced.

[0036] In the case implementation, the analysis and computing subsystem uses a large amount of labeled data to train the computer, and through deep learning, it has the ability to judge abnormal road traffic conditions. Relying on the strong computing power of the computer, when an abnormal situation occurs, the system will make instant predictions and simulations of the "abnormality". In addition, the analysis and computing subsystem is also the hub of information transmission, and it realizes efficient communication with the vehicle through the ESP32-S3 chip connected to the 5G network.

[0037] In the case implementation, although the vehicle's sensors can be integrated into the road monitoring subsystem, the assisted driving function is also a key component. Considering the driver's needs, the on-board sensors can be connected to the on-board screen to visualize and display the predicted road conditions on the screen, reminding the driver to avoid them in advance. In addition, the ESP chip can also be connected to key parts of the vehicle such as the steering wheel and brakes. When receiving warning information and the driver fails to respond in time, the ESP chip can perform emergency braking on the vehicle through pre-burned instructions to avoid accidents.

[0038] In the case implementation, the road monitoring subsystem includes hardware connection, program control and data transmission;

[0039] Hardware connection First, the ESP32-S3 chip must be physically connected to the vehicle's communication interface and the vehicle's network communication protocol must be determined, including the connection to the vehicle's CAN bus, Ethernet or other communication bus. For example, if the vehicle is based on the CAN bus, the ESP32-S3 chip needs to be connected to the CAN bus through a CAN driver chip (such as MCP2515). Next, the WiFi settings of the ESP32-S3 chip are performed so that it can connect to the vehicle's wireless network, thereby achieving the connection between the vehicle network and the 5G network, benchmarking network communication conditions, and ensuring smooth communication. In addition, it is also necessary to ensure that the ESP32-S3 chip has been reliably connected to each sensor, including power management, communication interface, and external drive circuit design. In order to more accurately simulate road conditions, the sensors on the vehicle need to be evenly and widely distributed. The cameras should be distributed in the front, side, and rear of the vehicle to monitor the surrounding environment in real time. In order to deal with special In the case of a collision, the vehicle also needs to be equipped with laser radar, infrared and other sensors to ensure that there are no blind spots around the vehicle. At the same time, the vehicle should also be equipped with friction and other sensors so that it can predict the braking trajectory and other situations when an early warning is generated. In terms of hardware connection, it is necessary to ensure that all lines are connected correctly and firmly. For example, the camera can be connected to the SPI interface of the ESP32-S3 chip through the SPI interface. When configuring the ESP32-S3 chip, it is necessary to set the SPI interface on the ESP32-S3 chip, set the correct clock frequency and data bit order (MSB or LSB), etc. To initialize the camera, set the pixel format, resolution and other parameters. To initialize the program, collect image data from the camera sensor. At this point, the camera connection is completed. Finally, the ESP32-S3 chip performs necessary processing on the collected image data, such as denoising and brightness adjustment, and the camera data can be transmitted. The connection method of other sensors is similar.

[0040] Program control first needs to develop or obtain a suitable driver so that the ESP32-S3 chip can communicate. At the same time, it is necessary to integrate necessary middleware, such as TCP / IP stack, CANopen stack, etc., to support network communication protocols. Then, the application code is written. Appropriate debugging tools should be used to test and verify the communication function to implement the specific logic of communicating with the vehicle network. To ensure communication security, the communication needs to be encrypted. The ESP32-S3 chip supports Flash encryption and secure boot based on the RSA algorithm, which can protect the vehicle from malicious attacks and data tampering. When writing the code, it is necessary to initialize the communication interface of the sensor. For example, if the sensor is connected via SPI, it is necessary to initialize the communication interface in the code. Set the SPI rate, host mode and other parameters in ; if the sensor is connected via I2C, you need to set the I2C bus and address, etc., and then read the sensor data. Once the communication interface is set up, you can write the code to read the data, such as using analogread to read the analog or digital signal of the sensor. The data format and reading method of different types of sensors may be different. Some sensors provide registers for direct data reading, while other sensors may need to obtain data through interrupts or command sequences. The data obtained by the sensor needs to be processed, such as filtering, calibration and unit conversion of certain data, and then some sensor monitored data can be converted into more accurate physical quantities through program writing;

[0041] During data transmission, the ESP32-S3 chip first connects to the analysis and computing subsystem in the cloud through the vehicle network. Since different sensors obtain different types of data, such as photos and images obtained by cameras, three-dimensional point clouds obtained by lidar, and target thermal image information obtained by infrared sensors, the obtained data is packaged according to different needs. Common formats such as JSON and XML contain sensor data and related identifiers or metadata. In the code of the ESP32-S3 chip, it is necessary to comply with the logic of writing data transmission to ensure that the data can be successfully transmitted to the analysis and computing subsystem in the cloud. After the code is deployed, it is also necessary to monitor and adjust it. Similarly, data transmission also requires the use of appropriate communication protocols (such as HTTP, MQTT, CoAP, etc.) to upload data from the ESP32-S3 chip to the cloud through the 5G network. In addition, it is necessary to write corresponding server-side programs to receive and process relevant data. Through these steps, the efficient operation of the entire system in terms of hardware connection, program control and data transmission can be ensured, and ultimately comprehensive monitoring and early warning of road conditions can be achieved.

[0042] In the case implementation, the analysis and computing subsystem includes vehicle big data processing and storage, road anomaly detection model construction, and intelligent matching of driving risk avoidance plans;

[0043] When processing and storing vehicle big data, the first uploaded data will be marked and processed. Data from different sources will be marked differently and combined with GPS or Beidou hazard positioning system to determine the location of the information source. These data will be continuously updated within the system monitoring range and stored in the storage space as records after leaving the monitoring range. The storage space also needs to record the system operation log and enter the monitoring data into the storage space. Due to limited storage space, the stored data needs to be cleaned or exported at regular intervals. In addition, the cloud database stores various driving avoidance plans so that plans can be matched in subsequent warnings.

[0044] The construction of road anomaly detection model is a machine learning method that achieves prediction by training artificial intelligence. The output is predicted by training artificial intelligence. Andrew Ng, a professor of computer science and electrical engineering at Stanford University, once pointed out: "Deep learning models are like rocket engines. The fuel is the massive data we can provide to these algorithms." The training methods include supervised learning and unsupervised learning (supervised learning uses data sets marked with input and expected output for training, while unsupervised learning uses unclassified or labeled information for machine learning, allowing the algorithm to operate on information without guidance). Both can train artificial intelligence. In terms of road traffic, supervised learning is more efficient. Artificial intelligence simulates the neural network structure of the biological brain through deep learning so that computers can perform specific tasks. Since the data received by the system is diverse and complicated, it is necessary to train a system that can handle multiple The AI ​​model of modal data is first of all multimodal feature fusion. Deep learning can fuse the features of multiple data modalities. By building a multimodal deep neural network model, the features of different data modalities can be jointly learned and fused to obtain richer and more accurate feature representations. For example, multi-layer perceptron (MLP), convolutional neural network (CNN), recurrent neural network (RNN) and other models can be used for feature fusion. Secondly, multimodal pre-trained models. Deep learning pre-trained models can be used for processing and analysis of multimodal data. Pre-training on large-scale data, such as the Bidirectional Encoder Representation for Images and Text (BERT) and Visual Language Encoder (ViLBERT), can be used. Obtain a model with rich expressive power for multimodal data, which is used for feature extraction, joint modeling and prediction of multimodal tasks. In this system, the multimodal attention mechanism is indispensable and can significantly improve the ability of road anomaly recognition. The attention mechanism in deep learning can be used for the association and alignment of multimodal data. By introducing the attention mechanism, the correlation between different data modalities can be automatically learned, and the important modalities can be weighted, which helps to enhance the joint analysis and understanding of multimodal data. For example, the multimodal attention network (MAN) can be used for joint modeling and analysis of image and text data, followed by a generative model. The generative model is an important part of deep learning and can be used for multimodal data. Generation and synthesis. By training generative models, such as generative adversarial networks (GANs) and variational autoencoders (VAEs), data samples with multimodal features can be generated for tasks such as data enhancement, sample generation, and model evaluation. Finally, anomaly detection and abnormal data processing are equally important. Deep learning can be used for anomaly detection and abnormal data processing of multimodal data. By training models to learn the normal distribution of multimodal data, abnormal data that does not conform to the normal distribution can be detected, thereby improving the accuracy and reliability of data analysis. In order to improve the accuracy of the analytical computing subsystem, its predicted results need to be compared with past results, which requires two elements: higher computing power and a large amount of data.Therefore, we need to collect a large amount of road data and accident data caused by blind spots to train artificial intelligence. First, we need to mark the elements such as vehicles, pedestrians, and non-motor vehicles in the road data to ensure that artificial intelligence can accurately identify key elements. Then, we need to train artificial intelligence on traffic rules so that it can distinguish whether the elements on the road violate traffic rules. Finally, we use a large amount of labeled accident data to train artificial intelligence to improve its ability to distinguish accidents caused by blind spots. After training, artificial intelligence should be able to identify the elements of illegal traffic on the road, identify the dangerous elements in blind spots, and mark these elements.

[0045] The driving risk avoidance scheme intelligently matches various types of data uploaded by sensors on different vehicles, based on different locations and relying on powerful computing power for traceability. The subsystem can simulate the situation of the entire road, establish a real-time 3D model, and simulate the road conditions in real time. The trained artificial intelligence is responsible for monitoring the model. When an abnormal situation occurs (such as a pedestrian running a red light), the system can promptly mark it as an abnormal item, and generate early warning information based on the current road conditions, and quickly transmit it to the vehicle that may collide with it. After the early warning information is transmitted, the artificial intelligence collects the vehicle motion status information on the current road, predicts the vehicle's motion trajectory, and predicts the motion trajectory of abnormal elements based on the real-time transmitted data, simulates the collision position, and packages the information and transmits it to the corresponding vehicle. At the same time, the artificial intelligence can connect to the driving scheme library, match the risk avoidance scheme, and quickly find the optimal solution. The vehicle can receive the best solution through the ESP32-S3 chip. After the originally predicted collision crisis is resolved, the early warning status will be lifted.

[0046] In the case implementation, the vehicle warning subsystem includes intelligent warning and emergency braking;

[0047] Intelligent warning is: under normal driving conditions, the vehicle screen will display the recommended braking distance, the distance to the vehicle in front, and the braking warning information. The ESP32-S3 chip connects to the cloud computer through the 5G network to achieve rapid transmission and reception of real-time calculation results. When the analysis and calculation subsystem conveys the warning information, the on-board ESP32-S3 chip will connect to the on-board sensor, and combine the computing power of the cloud computer to simulate the real-time road conditions and project them onto the on-board screen. At the same time, other sensors outside the vehicle connected to the ESP32-S3 chip will also transmit the collected data back to the ESP32-S3 chip, which will be visualized and displayed on the vehicle screen after chip processing. The ESP32-S3 chip will collect data from the friction sensor, and then combine factors such as vehicle speed to calculate the braking distance of the current braking situation. When receiving the warning information, the cloud will transmit the motion status information of the abnormal elements in real time. At the same time, the vehicle does not stop transmitting data to the cloud, and the cloud computer will calculate and simulate the vehicle and the abnormal elements. The motion trajectory of the vehicle and abnormal elements is displayed on the vehicle screen through the processing of the ESP32-S3 chip. The motion trajectory of the vehicle and abnormal elements is also rendered in different colors and projected onto the vehicle screen. The motion trajectory of the vehicle and abnormal elements simulated according to the current motion state is displayed in real time. According to the expected collision position marked by the cloud computer, the ESP32-S3 chip will mark the intersection of the two trajectories in red and display the expected collision time, so that the driver can understand the road conditions more intuitively. The cloud computer will also give driving suggestions. The ESP32-S3 chip is connected to the in-car voice system. The driving suggestions combined with the early warning prompts will be quickly broadcast by the in-car voice system after processing. The ESP32-S3 chip will calculate the braking distance in real time according to the built-in program. In an emergency, if the driver cannot react in time, the ESP32-S3 chip will urgently control the vehicle steering wheel and brakes, and perform emergency braking according to the driving suggestions in the cloud.

[0048] During emergency braking, the ESP32-S3 chip compares the size with three thresholds through a pre-written program, triggering different warning effects in different situations. There are two key data in the transmitted data: one is the braking distance required for normal braking, which is obtained through the friction and other sensors equipped on the vehicle and calculated by the ESP32-S3 chip. This data is set as the first threshold and displayed on the vehicle display. When the distance between the vehicle and the abnormal element is lower than this value, the first level of alarm will be triggered, with a slight sound and driving advice broadcast to prompt the driver to respond. The alarm will continue until the collision with the abnormal element can be completely avoided. The other data is the calculated real-time minimum braking distance, and 1.5 times the current minimum braking distance is set as the second threshold. When the vehicle and the abnormal element are close to this threshold, the second level of alarm will be triggered, emitting a sharper alarm to remind the driver to brake urgently until the vehicle and the abnormal element are close. The distance of the abnormal element is greater than the second threshold or the vehicle avoids the abnormal element. The third threshold is the real-time minimum braking distance. When the distance to the abnormal element is less than or equal to 1.1 times the value, the automatic avoidance program will be triggered. When the automatic avoidance program is triggered, the voice system in the car will broadcast to remind the driver that there is an emergency and the automatic avoidance program will be triggered. In the analysis and calculation subsystem, the cloud computing opportunity will analyze and calculate the collision probability under the current situation, and plan multiple plans, compare their collision probabilities (as mentioned in the calculation and analysis subsystem), and finally obtain the optimal solution and transmit it to the vehicle for execution. When the automatic avoidance program is triggered, it will directly feedback and will no longer continue to calculate new plans. The cloud computer will transmit the plan with the lowest current collision probability to the ESP32-S3 chip for execution. Considering that the collision probability cannot be reduced to zero, the ESP32-S3 chip will also control the alarm to emit a special alarm sound to remind passengers to prepare for collision avoidance.

[0049] When implementing, follow these steps:

[0050] 1) First, physically connect the ESP32-S3 chip to the vehicle's communication interface and determine the vehicle's network communication protocol;

[0051] 2) Then in the program control stage, first develop or obtain a suitable driver for the ESP32-S3 chip to communicate. At the same time, integrate the necessary middleware, such as TCP / IP stack, CANopen stack, etc., to support the network communication protocol. Then write the application code and use appropriate debugging tools to test and verify the communication function to implement the specific logic of communicating with the vehicle network.

[0052] 3) Then connect to the analysis and computing subsystem in the cloud through the vehicle network, because different sensors obtain different types of data;

[0053] 4) The uploaded data will be marked and processed first. Data from different sources will be marked differently, and the location of the information source will be determined in combination with GPS or Beidou Hazard Location System;

[0054] 5) Predict outputs by training AI, which should be able to identify elements of illegal traffic on the road, identify dangerous elements with blind spots, and mark these elements;

[0055] 6) Various data uploaded by sensors on different vehicles are traced based on different locations and rely on powerful computing power. The subsystem can simulate the entire road situation, build a real-time 3D model, and simulate the road conditions in real time. The trained artificial intelligence is responsible for monitoring the model.

[0056] 7) Under normal driving conditions, the vehicle screen will display the recommended braking distance, the distance to the vehicle in front, and the braking warning information. The ESP32-S3 chip connects to the cloud computer through the 5G network to quickly transmit and receive real-time calculation results. When the analysis and calculation subsystem conveys the warning information, the on-board ESP32-S3 chip will connect to the on-board sensor, combine the computing power of the cloud computer to simulate the real-time road conditions, and project them onto the on-board screen;

[0057] 8) The ESP32-S3 chip compares the size of the three thresholds through a pre-written program, triggering different warning effects in different situations. Among the data transmitted, there are two key data: one is the braking distance required for normal braking, which is obtained through the friction and other sensors equipped on the vehicle and calculated by the ESP32-S3 chip. This data is set as the first threshold and displayed on the vehicle display. When the distance between the vehicle and the abnormal element is lower than this value, the first level of alarm will be triggered, with a slight sound and driving advice broadcast to prompt the driver to respond. The alarm will continue until the collision with the abnormal element can be completely avoided. The other data is the calculated real-time minimum braking distance.

[0058] In summary, the intelligent traffic accident warning system based on the ESP32-S3 chip sets up a traffic system, and the system functions include monitoring, tracking, analysis, warning and braking. First, the monitoring function is to conduct all-weather monitoring of accident-prone areas, and obtain data through various sensors on different vehicles. Secondly, the tracking and analysis functions are inseparable. The data obtained by various sensors on the vehicle are first transmitted to the system's computer, and the computing system analyzes and judges it, and marks the vehicles or pedestrians on the road as "normal" or "abnormal". The system focuses on monitoring the movement of the "abnormal" marks and obtains information such as their speed. This information is then transmitted to the computing system for special processing. Combined with the information of other vehicles and objects in the entire road section, the computer analyzes and predicts the movement trajectory of these information objects and future road conditions, that is, predicts and simulates possible collisions and other accidents, and generates relevant The warning information is transmitted in time through the ESP32-S3 chip connected to the vehicle. When there is sufficient reaction time, the warning information will also be transmitted to the driver through the on-board screen and audio. If the driver cannot react in time, the system can perform emergency braking. Thanks to its rich peripheral interfaces, the ESP32-S3 chip can pre-burn relevant instructions and link and control key components of the vehicle (such as brakes and steering wheels), so as to respond in time when receiving the transmission message to avoid accidents. It solves the problem that visual blind spots are still one of the important causes of traffic accidents. In addition, there are many uncertain emergencies in road traffic, such as pedestrians or electric vehicles suddenly rushing out from behind obstacles, making it difficult for vehicles to brake in time even with intelligent auxiliary control, and different vehicles cannot avoid in time due to obstructions, or people or objects are in the space occupied by trucks when turning.

[0059] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0060] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent traffic accident warning system based on the ESP32-S3 chip, including a traffic system, characterized in that: The traffic system includes a road monitoring subsystem, an analysis and calculation subsystem, and a vehicle warning subsystem; The road monitoring subsystem: Modern vehicles are equipped with a variety of on-board sensors: such as cameras, radars and lidars. As part of the entire system, the sensors on the moving vehicle can be connected through the ESP chip and transmit the data to the computing system for comprehensive analysis; The analysis and calculation subsystem: The 5G network of the road can cover the entire road section through multiple routers to ensure communication. In this subsystem, the calculation and analysis are performed by a powerful computer; The vehicle warning subsystem: the vehicle's sensors can be integrated into the road monitoring subsystem, and the on-board sensors can be connected to the on-board screen to visualize and display the predicted road conditions on the screen, reminding the driver to avoid in advance.

2. The intelligent traffic accident warning system based on the ESP32-S3 chip according to claim 1, characterized in that: The road monitoring subsystem benefits from the low power consumption and high efficiency of the ESP32-S3 chip. Vehicle sensors can transmit real-time information after connecting to the network of a designated road section and send data to the computing system for processing. The computing system integrates data obtained from different vehicles and can simulate road conditions more completely.

3. The intelligent traffic accident warning system based on the ESP32-S3 chip according to claim 1, characterized in that: The analysis and computing subsystem uses a large amount of labeled data to train the computer, and through deep learning, it has the ability to judge abnormal road traffic conditions. Relying on the strong computing power of the computer, when an abnormal situation occurs, the system will make instant predictions and simulations of the "abnormality". In addition, the analysis and computing subsystem is also the hub of information transmission, and through the ESP32-S3 chip connected to the 5G network, efficient communication with the vehicle is achieved.

4. The intelligent traffic accident warning system based on the ESP32-S3 chip according to claim 1, characterized in that: The ESP chip in the vehicle warning subsystem can also be connected to key parts of the vehicle such as the steering wheel and brakes. When receiving warning information and the driver fails to respond in time, the ESP chip can perform emergency braking on the vehicle through pre-burned instructions to avoid accidents.

5. The intelligent traffic accident warning system based on the ESP32-S3 chip according to claim 1, characterized in that: The road monitoring subsystem includes hardware connection, program control and data transmission; The hardware connection first requires physically connecting the ESP32-S3 chip to the vehicle's communication interface and determining the vehicle's network communication protocol, including connecting to the vehicle's CAN bus, Ethernet or other communication bus; The program control first needs to develop or obtain a suitable driver so that the ESP32-S3 chip can communicate. At the same time, it is necessary to integrate necessary middleware, such as TCP / IP stack, CANopen stack, etc., to support network communication protocols. Then, the application code is written. Appropriate debugging tools are used to test and verify the communication function to implement the specific logic of communicating with the vehicle network. To ensure communication security, the communication needs to be encrypted. The ESP32-S3 chip supports Flash encryption and secure boot based on the RSA algorithm, which can protect the vehicle from malicious attacks and data tampering. During the data transmission, the ESP32-S3 chip is first connected to the analysis and computing subsystem in the cloud through the vehicle network. Since different sensors obtain different types of data, such as cameras obtaining photos and images, lidar obtaining three-dimensional point clouds, infrared sensors obtaining target thermal image information, etc., the obtained data is packaged according to different needs. Common formats such as JSON, XML, etc. contain sensor data and related identifiers or metadata.

6. The intelligent traffic accident warning system based on the ESP32-S3 chip according to claim 1, characterized in that: The analysis and calculation subsystem includes vehicle big data processing and storage, road anomaly detection model construction and intelligent matching of driving risk avoidance plans; When the vehicle big data is processed and stored, the data uploaded first will be marked and processed first, and data from different sources will be marked differently, and the location of the information source will be determined in combination with GPS or Beidou hazard positioning system. These data will be continuously updated within the monitoring range of the system, and will be stored in the storage space as a record after leaving the monitoring range. The storage space also needs to record the operation log of the system and enter the monitoring data into the storage space. Due to limited storage space, the stored data needs to be cleaned or exported every once in a while; The road anomaly detection model construction is a machine learning method that achieves prediction purposes by training artificial intelligence. The output is predicted by training artificial intelligence. Artificial intelligence simulates the neural network structure of the biological brain through deep learning so that the computer can perform specific tasks. The driving risk avoidance solution intelligently matches various types of data uploaded by sensors on different vehicles, based on different locations, and relies on powerful computing power for traceability. The subsystem can simulate the conditions of the entire road, establish a real-time 3D model, and simulate road conditions in real time. The trained artificial intelligence is responsible for monitoring the model.

7. The intelligent traffic accident warning system based on the ESP32-S3 chip according to claim 1, characterized in that: The vehicle warning subsystem includes intelligent warning and emergency braking; The intelligent warning is as follows: under normal driving conditions, the vehicle screen will display the recommended braking distance, the distance to the vehicle in front, and the braking warning information. The ESP32-S3 chip connects to the cloud computer through the 5G network to achieve rapid transmission and reception of real-time calculation results. When the analysis and calculation subsystem conveys the warning information, the on-board ESP32-S3 chip will connect to the on-board sensor, combine the computing power of the cloud computer to simulate the real-time road conditions, and project it onto the on-board screen. At the same time, other sensors outside the vehicle connected to the ESP32-S3 chip will also transmit the collected data back to the ESP32-S3 chip, which will be visualized after being processed by the chip and displayed on the vehicle screen. During emergency braking, the ESP32-S3 chip compares the size with three thresholds through a pre-written program, triggering different warning effects in different situations. Among the data transmitted, there are two key data: one is the braking distance required for normal braking, which is obtained through the friction and other sensors equipped on the vehicle and calculated by the ESP32-S3 chip. This data is set as the first threshold and displayed on the vehicle display. When the distance between the vehicle and the abnormal element is lower than this value, the first level of alarm will be triggered, with a slight sound and driving advice broadcast to prompt the driver to respond. The alarm will continue until the collision with the abnormal element can be completely avoided. The other data is the calculated real-time minimum braking distance.

Images