Electric vehicle collision signal rapid processing and transmission system based on edge calculation

Abstract

An electric vehicle collision signal rapid processing and transmission system based on edge calculation comprises a collision signal acquisition module, a signal preprocessing module, an edge calculation module, a communication and transmission module, a system operation and maintenance module and a user interface and management module. The signal preprocessing module is used for preprocessing signals, the edge calculation module is used for processing collision signals and judging collision types and occurrence time, the communication and transmission module is used for transmitting processed collision information to a cloud end in real time, and the system operation and maintenance module is used for monitoring and maintaining operation states of all modules of the system. And the user interface and management module is used for providing an interactive interface and managing the interactive interface. According to the electric vehicle collision signal rapid processing and transmission system based on the edge calculation, an electric vehicle collision detection algorithm based on the edge calculation is provided for detecting collision; and a signal rapid communication algorithm based on multi-path transmission is provided to rapidly transmit the collision signal.

Description

Technical Field

[0001] The present invention relates to the technical fields of edge computing, electric vehicle collision detection, and multi-path transmission, and specifically relates to a fast processing and transmission system for electric vehicle collision signals based on edge computing. Background Technique

[0002] Edge computing technology is a technology that moves data processing tasks from the cloud to edge devices, aiming to solve the real-time processing bottlenecks caused by data transmission latency and bandwidth limitations in traditional cloud computing. Edge computing can perform real-time processing of data near the data source, significantly reducing latency and improving response speed. In an electric vehicle collision signal processing system, edge computing can quickly receive data from vehicle sensors, analyze collision information in real time, and make immediate decisions. By pushing the computing tasks to the edge, edge computing can provide collision detection results in a shorter time and ensure the safety of the vehicle.

[0003] Electric vehicle collision detection technology is a technology that collects data based on vehicle internal sensors and uses algorithms to analyze and identify collision events, aiming to solve the problems that traditional safety systems fail to detect collisions in time and cannot accurately evaluate the collision intensity. Through precise collision identification and real-time processing, this technology can effectively prevent injuries after accidents and improve vehicle safety. By closely integrating electric vehicle collision detection technology with edge computing and real-time processing of collision signals through edge devices, collision events can be quickly identified, ensuring that the system can quickly respond after an accident.

[0004] Multi-path transmission technology is a technology that uses multiple communication paths for data transmission, aiming to solve the problems of insufficient bandwidth, unstable transmission, and excessive latency existing in a single transmission path. By distributing data streams among different paths, multi-path transmission can improve the reliability and speed of data transmission. In a fast processing and transmission system for electric vehicle collision signals based on edge computing, multi-path transmission technology is used to ensure that collision signals can be quickly and stably transmitted to remote platforms and emergency response systems, avoiding information transmission delays caused by faults and congestion in a certain network path. By optimizing data transmission paths and intelligently selecting the best path, the system can ensure that accident information is sent to relevant departments in a timely and accurate manner after a collision, providing crucial support for accident handling.

[0005] However, an existing rapid processing and transmission system for electric vehicle collision signals based on edge computing has the following problems: it relies on cloud processing, where data needs to be transmitted to the cloud and wait for analysis, resulting in high latency and the inability to achieve real-time response after a collision, thus affecting the timeliness and effectiveness of accident emergency handling. Secondly, traditional collision detection technologies are limited by sensor accuracy and computing power and cannot accurately judge collision events in complex and harsh environments. In addition, traditional systems rely solely on a single path for data transmission. If problems such as network interruption and weak signals occur, it will lead to the loss of collision information and delayed transmission, affecting emergency response. Summary of the Invention

[0006] The purpose of the present invention is to provide a rapid processing and transmission system for electric vehicle collision signals based on edge computing, so as to solve the problems in the existing rapid processing and transmission system for electric vehicle collision signals based on edge computing mentioned in the above background technology, namely, relying on cloud processing, where data needs to be transmitted to the cloud and wait for analysis, resulting in high latency and the inability to achieve real-time response after a collision, thus affecting the timeliness and effectiveness of accident emergency handling. Secondly, traditional collision detection technologies are limited by sensor accuracy and computing power and cannot accurately judge collision events in complex and harsh environments. In addition, traditional systems rely solely on a single path for data transmission. If problems such as network interruption and weak signals occur, it will lead to the loss of collision information and delayed transmission, affecting emergency response.

[0007] To achieve the above object, the present invention provides the following technical solutions: A rapid processing and transmission system for electric vehicle collision signals based on edge computing, including a collision signal acquisition module, a signal preprocessing module, an edge computing module, a communication and transmission module, a system operation and maintenance module, and a user interface and management module, characterized in that: The collision signal acquisition module is used to collect the sensor data of the electric vehicle in real time and send it to the edge computing device to ensure that collision events can be detected in a timely manner; The signal preprocessing module is used to filter, denoise and enhance the collected original signals to ensure the clarity and accuracy of the data for subsequent processing; The edge computing module includes a real-time monitoring and feedback unit and a collision recognition and decision-making unit. The real-time monitoring and feedback unit is used to monitor the sensor signals and system status in real time and transmit the video data to the collision recognition and decision-making unit in real time to ensure timely feedback of collision information and initiate response measures. The collision recognition and decision-making unit proposes an electric vehicle collision detection algorithm based on edge computing to analyze the processed signals, quickly identify the collision type and make corresponding decisions to ensure efficient triggering of emergency responses; The communication and transmission module includes a data reception and parsing unit and a signal rapid transmission unit. The data reception and parsing unit is used to receive and parse the processing results from the edge computing module and perform parsing to ensure that the collision information is accurately prepared for transmission. The signal rapid transmission unit proposes a signal rapid communication algorithm based on multi-path transmission to simultaneously transmit collision information through multiple communication paths to ensure low-latency and high-reliability rapid data transmission in different network environments; The system operation and maintenance module is used to monitor the health status of each module, ensure the normal operation of the system and perform fault diagnosis and repair in a timely manner to ensure the stability and reliability of the system; The user interface and management module is used to provide an interactive interface for users to ensure that users can view collision information, receive alarms and manage system settings.

[0008] Preferably, the collision signal acquisition module accesses various sensors of the vehicle, including acceleration sensors, collision sensors, gyroscope sensors, temperature sensors, and positioning sensor information, and sends the data to the edge computing device in real time to ensure that the vehicle status and external collision information can be accurately collected during a collision, providing basic data for subsequent signal processing and collision analysis.

[0009] Preferably, the signal preprocessing module processes the collected original signals by applying digital filtering technology and signal enhancement technology, including denoising, filtering, enhancing key features, and data normalization processing, to ensure that the input signal quality meets the system requirements and provide accurate data for subsequent processing modules to analyze and make decisions.

[0010] Preferably, the edge computing module includes a real-time monitoring and feedback unit. The real-time monitoring and feedback unit monitors vehicle sensors and system status through edge devices, obtains video data in real time and transmits it to the collision recognition and decision-making unit, and feeds back the system status to the vehicle's safety control system to ensure that the vehicle can respond in a timely manner when a collision occurs.

[0011] Preferably, the edge computing module includes a collision recognition and decision-making unit. The collision recognition and decision-making unit proposes an electric vehicle collision detection algorithm based on edge computing, makes decisions according to the type and severity of the collision, including activating the alarm system and calling the ambulance system. The decision-making process does not rely on the cloud, ensuring low-latency real-time response and ensuring that collisions are identified efficiently and accurately.

[0012] Preferably, the electric vehicle collision detection algorithm based on edge computing is as follows: First, receive the video data of the real-time monitoring and feedback unit, and use the deep learning object detection algorithm SSD-Inception to perform object detection on each frame of the video, generating a bounding matrix containing object category and location information. The SSD-Inception algorithm extracts features from the video frame through a convolutional neural network CNN and uses the features to locate and classify objects. The specific formula is expressed as:

[0013] B = Detect(I)

[0014] where I represents the input video frame, B represents the bounding matrix output by the SSD-Inception algorithm, Detect represents the deep learning object detection function. Then, construct a tracking strategy to associate the bounding matrices in each frame, generate the motion trajectory of the target, and achieve continuous tracking of the target. The tracking strategy function SORT uses a Kalman filter to predict the motion state of the target, generates a prediction box, matches the prediction box with the actual bounding matrix detected in the current frame, determines the target identity and updates the state by calculating the similarity matrix between targets, and achieves continuous and accurate tracking of multiple targets. The specific formula is expressed as:

[0015]

[0016] T t = SORT(B t , T t-1 )

[0017] where represents the state prediction of target i at the t-th frame, t represents the frame number index of the video, i represents the detection target number index, F represents the state transition matrix, x t-1，i represents the state of target i at the (t - 1)-th frame, u t，i represents the control vector, w t，iis represented as process noise, and Associations is represented as the matching relationship between the predicted bounding boxes and the detected bounding boxes. is represented as the bounding matrix predicted for the t-th frame, IoU is represented as the intersection over union calculated between two bounding boxes, Hungarian is represented as the maximum matching function, x t，i is represented as the predicted updated state of target i for the t-th frame, y t，i is represented as the observation vector, KUpdate is represented as the update function of the Kalman filter, T t is represented as the trajectory of the generated target for the t-th frame, SORT is represented as the tracking function, T t-1 is represented as the trajectory of the generated target for the (t - 1)-th frame. Secondly, by calculating the relative motion between the target object and the electric vehicle, the time to collision TTC is estimated. The size change rate of the bounding matrix is used to estimate TTC. The change rate of the height of the bounding matrix is calculated by linear regression to estimate TTC. The specific formula is expressed as:

[0018]

[0019] where, r high is the change rate of the height of the bounding matrix, h t and h t+n are the heights of the bounding boxes for the t-th frame and the (t + n)-th frame respectively, Δt is the time interval between frames, n is represented as the frame number index of the video. By calculating TTC in real time on the edge computing device, the collision risk is quickly judged. By analyzing the motion pattern of the target object in the horizontal direction, it is further judged whether there is a collision risk. Then, the horizontal position change of the center point of the bounding box is used to identify the horizontal motion pattern. The change rate of the horizontal position of the center point of the bounding box is calculated by linear regression to judge the motion direction of the target object. The specific formula is expressed as:

[0020]

[0021] where, x center is represented as the horizontal position of the center point of the bounding matrix, x i and w i are the abscissa and width of the upper left corner of the bounding matrix respectively, r x is represented as the change rate of the horizontal position, x center，t+n and x center，t are the horizontal positions of the center points of the bounding boxes for the (t + n)-th frame and the t-th frame respectively. Secondly, by comprehensively considering TTC and the horizontal motion pattern, it is judged whether there is a collision risk, and the collision risk information is output. The specific formula is expressed as:

[0022] TTC < δ, α < r x < α′

[0023] Among them, δ represents the TTC threshold, α and α′ are respectively the lower and upper threshold values of the horizontal motion mode. According to the thresholds of TTC and the horizontal motion mode, it is judged whether there is a collision risk. If TTC is less than the preset threshold δ and the horizontal motion mode meets the preset conditions, it is judged that there is a collision risk. Finally, the collision risk information is output to the subsequent module, ensuring that the collision risk information can be transmitted and analyzed in a timely manner. The specific formula is expressed as:

[0024] R = {TTC, r x , class, location}

[0025] Among them, R represents the collision risk information, class represents the target category information, and location represents the target position information. The electric vehicle collision detection algorithm based on edge computing realizes the accurate assessment and rapid response to the collision risk of electric vehicles through the combination of multiple advanced technologies such as deep learning object detection, real-time motion tracking, and time to collision estimation. Through the edge computing-based design, it not only improves the processing efficiency of the system, reduces the dependence on cloud computing, but also ensures that when an electric vehicle collision occurs, it can quickly respond and enhance the safety of electric vehicles.

[0026] Preferably, the communication and transmission module includes a data reception and parsing unit. The data reception and parsing unit ensures that all transmitted data can be received in a timely and accurate manner by receiving the real-time processed data output from the edge computing module, and decodes and parses through a dynamic parsing protocol to ensure that the data format and content conform to the transmission standard, facilitating subsequent processing and transmission.

[0027] Preferably, the communication and transmission module includes a signal fast transmission unit. The signal fast transmission unit proposes a signal fast communication algorithm based on multi-path transmission. Through multiple communication paths, including Wi-Fi, 5G, and LTE, it provides a redundant transmission method to ensure that data can still be transmitted smoothly when the network conditions are unstable and some paths fail, ensuring that the collision signal can be transmitted to relevant service platforms, including the traffic management system and the accident handling center, at low latency and high reliability, providing timely feedback and supporting accident emergency response.

[0028] Preferably, the signal fast communication algorithm based on multi-path transmission is specifically as follows: First, dynamically screen the optimal transmission path set among edge computing nodes to ensure that the electric vehicle collision signal is transmitted across the network with the lowest latency. Through the multi-path parallel transmission mechanism, the risk of single-path congestion is avoided, and the robustness of signal transmission is improved. Assuming there are M candidate communication paths, the comprehensive priority Γm of communication path m is specifically expressed by the following technical formula:

[0029]

[0030] Among them, Γm represents the comprehensive priority score of path m, m represents the index of the number of communication paths, λ1 represents the weight coefficient of bandwidth, representing the importance of bandwidth in the path priority score, λ2 represents the weight coefficient of delay, representing the importance of delay in the path priority score, and λ3 represents the weight coefficient of stability, representing the importance of path stability in the priority score. The weight coefficients satisfy λ1 + λ2 + λ3 = 1 to ensure the normalization of the priority score, B m represents the available bandwidth of communication path m, B max represents the maximum bandwidth value of all paths in the network, Δ m represents the time required for data to be transmitted from the source node to the target node through path m, Δ max represents the maximum delay threshold allowed by the system, σ m represents the transmission stability of path m, with a value range of 0 to 1. Then, according to the real-time load of the path, the signal traffic is dynamically allocated to avoid local path overload, optimize resource utilization, and maximize the overall transmission efficiency by balancing the multi-path load to meet the fast communication requirements of collision signals. The specific formula is expressed as:

[0031]

[0032] Among them, ω m represents the traffic allocation weight, L m represents the current load of the m-th communication path, represents the maximum carrying load of the m-th communication path, k represents the index of the number of communication paths, Γ k represents the comprehensive priority score of path k, represents the maximum carrying load of the k-th communication path, L k represents the current load of the k-th communication path. By combining the load status of the edge nodes, the traffic allocation is dynamically adjusted to ensure the fast transmission of electric vehicle collision signals under complex road conditions. Then, the electric vehicle collision signals are fragmented and transmitted synchronously through multiple paths, and the multi-path bandwidth superposition effect is used to shorten the transmission time. The single-path bandwidth limit is broken through by the fragmented parallel transmission strategy, thereby improving the transmission rate. Assuming that the total size of the collision signal is S and the number of fragments is N, the specific formula for the transmission time is:

[0033]

[0034] Among them, T n represents the transmission time of fragment n, n represents the index of the number of fragments, s n represents the size of the collision signal of fragment n, s n = S / N, and the overall transmission time T total is determined by the slowest fragment. The specific formula is expressed as:

[0035] T total = max{T1, T2, ..., T N}

[0036] where max represents the maximization function, and T1, T2, ..., T N respectively represent the independent shard transmission times. Secondly, by real-time monitoring of the path performance, when the delay and packet loss rate of a certain path exceed the threshold, it automatically switches to the backup path to ensure the transmission continuity and avoid signal loss and delay caused by path failures. The specific formula is expressed as:

[0037]

[0038] where t switch represents the trigger path switching time, B min represents the system's minimum required bandwidth, Δ th represents the timing delay threshold, ρ m represents the packet loss rate of the m-th communication path, and ρ th represents the packet loss rate threshold. Through the adaptive switching mechanism, it ensures the highly reliable transmission of the electric vehicle collision signal in the edge network. Finally, at the target edge node, the received shards are subjected to integrity verification and recombination to ensure the signal is error-free, and the end-to-end delay is reduced through distributed verification to ensure the signal integrity and real-time performance. The specific formula is expressed as:

[0039]

[0040] where Hash represents the preset hash function, represents the bitwise exclusive OR operation. Based on the collaborative computing ability of the edge nodes, the signal recombination is quickly completed to meet the real-time requirements of electric vehicle collision processing.

[0041] Preferably, the system operation and maintenance module ensures the stable operation of the system by regularly monitoring the operating states of the system hardware, sensors, and computing modules, and through the real-time fault detection and warning mechanism, ensures that the system can quickly perform self-repair and notify the maintenance personnel when abnormalities occur, guaranteeing the reliability and continuous availability of the system.

[0042] Preferably, the user interface and management module ensure that users can obtain system feedback in real time and remotely monitor and adjust the system through a graphical interface and an easy-to-use interaction design, ensuring that users can receive warnings and accident details in a timely manner after an accident occurs.

[0043] Compared with the prior art, the beneficial effects of the present invention are:

[0044] 1. Collision recognition and decision-making unit An electric vehicle collision detection algorithm based on edge computing is proposed. The algorithm receives video data through the real-time monitoring and feedback unit and uses the deep learning target detection algorithm SSD-Inception to process each frame of video. SSD-Inception uses the convolutional neural network CNN to extract features of targets in video frames. By generating a boundary matrix containing object categories and location information, it effectively identifies various types of targets in the scene and accurately locates their positions, providing a data basis for subsequent target tracking and collision analysis. By using a deep learning model, SSD-Inception can quickly identify objects around the vehicle in a complex driving environment, including pedestrians and obstacles. and other vehicles, and provides the system with high-precision target detection and classification results. Next, the system builds a target tracking strategy, combines the Kalman filter and the SORT algorithm, continuously tracks the boundary matrix in each frame, and predicts the target's motion trajectory. The algorithm calculates the similarity matrix between targets to achieve continuous and accurate tracking of multiple targets, ensuring that even if the target changes in complex scenes, the system can always track the target's motion state. The constructed edge computing-based target tracking strategy enables the system to process video frames and target motion data in real time on local devices without relying on the remote cloud, reducing transmission delays and improving response speed. Secondly, the system classifies the relative motion between the target object and the electric vehicle. By analyzing and estimating the collision time, the system can estimate the height change of the target object through the size change rate of the boundary matrix, and then deduce the possible collision time. By calculating the change rate of the height of the boundary matrix through linear regression, the time window of the collision can be accurately evaluated. When the estimated collision time is less than the preset threshold, the system indicates that the risk of collision is high and an emergency response needs to be initiated immediately. Compared with the traditional rule-based collision detection method, the use of estimated collision time for collision risk assessment can achieve higher-precision predictions, ensuring that electric vehicles can respond quickly before a collision occurs and minimize accident damage. In addition, the algorithm further determines whether there is a collision risk by analyzing the target object's horizontal movement pattern. By calculating the horizontal position change rate of the center point of the boundary matrix, the algorithm can identify the movement direction of the target object and accurately determine whether the target is on the collision path of the electric vehicle. If the estimated collision time is less than the preset threshold and the horizontal movement mode of the target meets the preset conditions, it is determined that there is a collision risk. The risk assessment mechanism can monitor the movement of the target object in real time during the driving of the electric vehicle and quickly determine whether emergency avoidance measures are needed. Finally, the estimated collision time and horizontal movement mode are calculated and judged in real time through the edge computing device. The system can instantly output collision risk information, including target category, location and collision time data. This information is transmitted to subsequent modules to ensure that the collision risk can be transmitted in a timely and accurate manner.And take corresponding emergency response measures. In summary, the electric vehicle collision detection algorithm based on edge computing realizes the accurate assessment and rapid response to the collision risk of electric vehicles through the combination of multiple advanced technologies such as deep learning object detection, real-time motion tracking, and collision time estimation. Through the edge computing-based design, it not only improves the processing efficiency of the system, reduces the dependence on cloud computing, but also ensures that it can respond quickly and enhance the safety of electric vehicles when a collision occurs.,

[0045] 2. The signal fast transmission unit proposes a signal fast communication algorithm based on multi-path transmission. This algorithm first evaluates the comprehensive priority scores of multiple candidate communication paths and dynamically selects the most suitable set of transmission paths to optimize the quality of data transmission. The comprehensive priority score takes into account multiple factors such as the bandwidth, delay, and stability of the path, ensuring that in a complex network environment, even if a certain path encounters a bandwidth bottleneck and excessive delay, other paths can still make up for its deficiencies and guarantee the fast and reliable transmission of electric vehicle collision signals. Specifically, the edge computing node intelligently and dynamically allocates signal traffic according to the real-time network conditions and the load conditions of the transmission paths, avoiding the overall transmission efficiency being affected by a single path being overloaded. This technology ensures that the collision signals can be evenly transmitted on multiple paths, thus maximizing the utilization rate of resources and the transmission rate. By optimizing traffic allocation and multi-path load balancing, the system can effectively solve the problem of network bandwidth resource allocation, improve the overall throughput of the system while meeting the fast transmission requirements of electric vehicle collision signals. For each path, the system calculates the comprehensive priority according to its bandwidth, delay, and stability parameters, dynamically adjusts the allocation of data streams, and avoids any path being overloaded, ensuring that each path can operate under the optimal load. In addition, through the shard parallel transmission strategy of multi-path transmission, the system breaks through the limitation of the bandwidth of a single path and greatly shortens the transmission time. The collision signal is divided into multiple small fragments and synchronously transmitted through different paths. The transmission time of each shard is determined by the bandwidth and delay of the selected path, and the slowest transmission shard will determine the overall transmission time. This strategy utilizes the bandwidth superposition effect to increase the transmission rate of the collision signal. Through shard parallel transmission and path load balancing, the system can shorten the transmission time and meet the requirement that the collision signal needs to be immediately transmitted to the emergency response system after an electric vehicle collision. To ensure the continuity and stability of transmission, the algorithm also designs a mechanism for real-time monitoring of path performance. When the delay and packet loss rate of a certain transmission path exceed the set threshold, the system will automatically switch to the backup path to ensure the reliable transmission of the signal. The path switching mechanism makes decisions based on the real-time information of bandwidth, delay, and packet loss rate, thus avoiding the loss and delay of collision signals caused by single-path failures and congestion, and enhancing the high reliability of the system. The time threshold and packet loss rate threshold that trigger path switching can be adjusted according to the real-time network conditions, further improving the adaptability and robustness of the system in various complex network environments. Finally, after receiving all the shards, the target edge node performs integrity verification and recombination to ensure that all data is correct. Through distributed checksum collaborative calculation, the edge node can quickly reassemble each shard into the complete collision signal data, thus improving the reliability of the signal. Utilizing the computing power of the edge node, the system can immediately perform signal verification and recombination while receiving the signal, greatly reducing the additional delay caused by data loss and errors, and further meeting the high requirements for real-time performance of the electric vehicle collision processing system. Generally speaking,Application of a signal fast communication algorithm based on multipath transmission in an electric vehicle collision signal processing system. Through technical means such as dynamically selecting the optimal transmission path, balancing multipath load, fragmenting and parallel transmission, and automatic path switching, the signal transmission rate, reliability, and stability are improved. The system can ensure that collision signals can be quickly and efficiently transmitted to the emergency response platform in a complex network environment, providing strong protection for the safety of electric vehicles. Brief Description of the Drawings

[0046] Figure 1 It is a schematic structural diagram of the present invention; Detailed Embodiment

[0047] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0048] Please refer to Figure 1, the present invention provides a fast processing and transmission system for electric vehicle collision signals based on edge computing, including a collision signal acquisition module, a signal preprocessing module, an edge computing module, a communication and transmission module, a system operation and maintenance module, and a user interface and management module. It is characterized in that: the collision signal acquisition module is used to collect the sensor data of the electric vehicle in real time and send it to the edge computing device to ensure that collision events can be detected in a timely manner; the signal preprocessing module is used to filter, denoise, and enhance the acquired original signal to ensure the clarity and accuracy of the data for subsequent processing; the edge computing module includes a real-time monitoring and feedback unit and a collision recognition and decision-making unit. The real-time monitoring and feedback unit is used to monitor the sensor signals and system status in real time to ensure timely feedback of collision information and initiate response measures. The collision recognition and decision-making unit proposes an electric vehicle collision detection algorithm based on edge computing to analyze the processed signals, quickly identify the collision type, and make corresponding decisions to ensure efficient triggering of emergency responses; the communication and transmission module includes a data reception and parsing unit and a signal fast transmission unit. The data reception and parsing unit is used to receive and parse the processing results from the edge computing module and perform parsing to ensure that the collision information is accurately prepared for transmission. The signal fast transmission unit proposes a signal fast communication algorithm based on multi-path transmission to simultaneously transmit collision information through multiple communication paths to ensure fast data transmission with low latency and high reliability in different network environments; the system operation and maintenance module is used to monitor the health status of each module, ensure the normal operation of the system, and perform fault diagnosis and repair in a timely manner to ensure the stability and reliability of the system; the user interface and management module is used to provide an interactive interface for users to ensure that users can view collision information, receive alarms, and manage system settings.

[0049] Refer to Figure 1 , further, the collision signal acquisition module accesses various sensors of the vehicle, including acceleration sensors, collision sensors, gyroscope sensors, temperature sensors, and positioning sensor information, and sends the data to the edge computing device in real time to ensure that the vehicle status and external collision information can be accurately collected during a collision, providing basic data for subsequent signal processing and collision analysis.

[0050] Refer to Figure 1 , further, the signal preprocessing module processes the acquired original signal by applying digital filtering technology and signal enhancement technology, including denoising, filtering, enhancing key features, and data normalization processing, to ensure that the input signal quality meets the system requirements and provides accurate data for subsequent processing modules to analyze and make decisions.

[0051] Refer to Figure 1, Further, the edge computing module includes a real-time monitoring and feedback unit. The real-time monitoring and feedback unit monitors the vehicle sensors and system status through edge devices, obtains video data in real time and transmits it to the collision recognition and decision-making unit, and feeds back the system status to the vehicle's safety control system to ensure that the vehicle can respond in a timely manner when a collision occurs.

[0052] Refer to Figure 1 , Further, the edge computing module includes a collision recognition and decision-making unit. The collision recognition and decision-making unit proposes an electric vehicle collision detection algorithm based on edge computing, makes decisions according to the type and severity of the collision, including activating the alarm system and calling the ambulance system. The decision-making process does not depend on the cloud, ensuring low-latency real-time response and ensuring that collisions are identified efficiently and accurately.

[0053] Refer to Figure 1 , Further, the electric vehicle collision detection algorithm based on edge computing is specifically as follows: First, receive the video data of the real-time monitoring and feedback unit, and use the deep learning object detection algorithm SSD-Inception to perform object detection on each frame of the video, generating a bounding matrix containing object category and location information. The SSD-Inception algorithm extracts features in the video frame through a convolutional neural network CNN and uses the features to locate and classify objects. The specific formula is expressed as:

[0054] B = Detect(I)

[0055] where I represents the input video frame, B represents the bounding matrix output by the SSD-Inception algorithm, Detect represents the deep learning object detection function. Then, construct a tracking strategy to associate the bounding matrices in each frame, generate the motion trajectory of the target, and achieve continuous tracking of the target. The tracking strategy function SORT uses a Kalman filter to predict the motion state of the target, generate a prediction box, match the prediction box with the actual bounding matrix detected in the current frame, determine the target identity and update the state by calculating the similarity matrix between targets, and achieve continuous and accurate tracking of multiple targets. The specific formula is expressed as:

[0056]

[0057] T t = SORT(B t , T t-1 )

[0058] where, represents the state prediction of target i in the t-th frame, t represents the frame number index of the video, i represents the detection target number index, F represents the state transition matrix, x t-1，i represents the state of target i in the (t - 1)-th frame, ut，i Denoted as the control vector, w t，i Denoted as the process noise, Associations denotes the matching relationship between the predicted box and the detection box, Denoted as the boundary matrix predicted at the t-th frame, IoU denotes the intersection over union calculated between two bounding boxes, Hungarian denotes the maximum matching function, x t，i Denoted as the predicted updated state of target i at the t-th frame, y t，i Denoted as the observation vector, KUpdate denotes the update function of the Kalman filter, T t Denoted as the trajectory of the generated target at the t-th frame, SORT denotes the tracking function, T t-1 Denoted as the trajectory of the generated target at the (t - 1)-th frame. Secondly, by calculating the relative motion between the target object and the electric vehicle, the time to collision TTC is estimated. The size change rate of the boundary matrix is used to estimate TTC. The change rate of the height of the boundary matrix is calculated by linear regression to estimate TTC. The specific formula is expressed as:

[0059]

[0060] where, r high The change rate of the height of the boundary matrix, h t and h t+n are the heights of the bounding boxes at the t-th frame and the (t + n)-th frame respectively, Δt is the time interval between frames, n denotes the frame number index of the video. By calculating TTC in real time on the edge computing device, the collision risk is quickly judged. By analyzing the motion pattern of the target object in the horizontal direction, it is further judged whether there is a collision risk. Then, the horizontal position change of the center point of the bounding box is used to identify the horizontal motion pattern. The change rate of the horizontal position of the center point of the bounding box is calculated by linear regression to judge the motion direction of the target object. The specific formula is expressed as:

[0061]

[0062] where, x center Denoted as the horizontal position of the center point of the boundary matrix, x i and w i are the abscissa and width of the upper left corner of the boundary matrix respectively, r x Denoted as the horizontal position change rate, x center,t+n and x center，t are the horizontal positions of the center points of the bounding boxes at the (t + n)-th frame and the t-th frame respectively. Secondly, by combining TTC and the horizontal motion pattern, it is judged whether there is a collision risk, and the collision risk information is output. The specific formula is expressed as:

[0063] TTC < δ, α < r x < α′

[0064] Among them, δ represents the TTC threshold, α and α′ are the lower and upper threshold values of the horizontal motion mode respectively. According to the thresholds of TTC and the horizontal motion mode, it is judged whether there is a collision risk. If TTC is less than the preset threshold δ and the horizontal motion mode meets the preset conditions, it is judged that there is a collision risk. Finally, the collision risk information is output to the subsequent module, ensuring that the collision risk information can be transmitted and analyzed in a timely manner. The specific formula is expressed as:

[0065] R = {TTC, r x , class, location}

[0066] Among them, R represents the collision risk information, class represents the target category information, and location represents the target location information. The electric vehicle collision detection algorithm based on edge computing realizes the accurate evaluation and rapid response to the collision risk of electric vehicles through the combination of multiple advanced technologies such as deep learning object detection, real-time motion tracking, and collision time estimation. Through the edge computing-based design, it not only improves the processing efficiency of the system, reduces the dependence on cloud computing, but also ensures that it can respond quickly when an electric vehicle collision occurs and improves the safety of electric vehicles.

[0067] Refer to Figure 1 , further, the communication and transmission module includes a data reception and parsing unit. The data reception and parsing unit ensures that all transmitted data can be received in a timely and accurate manner by receiving the real-time processed data output from the edge computing module, and decodes and parses through a dynamic parsing protocol to ensure that the data format and content meet the transmission standards, facilitating subsequent processing and transmission.

[0068] Refer to Figure 1 , further, the communication and transmission module includes a signal fast transmission unit. The signal fast transmission unit proposes a signal fast communication algorithm based on multi-path transmission. Through multiple communication paths, including Wi-Fi, 5G, and LTE, it provides a redundant transmission method to ensure that data can still be transmitted smoothly when the network conditions are unstable and some paths fail, ensuring that the collision signal can be transmitted to relevant service platforms, including traffic management systems and accident handling centers, under low latency and high reliability, providing timely feedback and supporting accident emergency response.

[0069] Refer to Figure 1 , further, the signal fast communication algorithm based on multi-path transmission is specifically as follows: First, dynamically screen the optimal transmission path set among edge computing nodes to ensure that the electric vehicle collision signal is transmitted across the network with the lowest latency. Through the multi-path parallel transmission mechanism, the risk of single-path congestion is avoided, and the robustness of signal transmission is improved. Suppose there are M candidate communication paths, and the comprehensive priority Γ of communication path m mThe specific technical formula is expressed as:

[0070]

[0071] Among them, Γm represents the comprehensive priority score of path m, m represents the index of the number of communication paths, λ1 represents the weight coefficient of bandwidth, representing the importance of bandwidth in the path priority score, λ2 represents the weight coefficient of delay, representing the importance of delay in the path priority score, λ3 represents the weight coefficient of stability, representing the importance of path stability in the priority score. The weight coefficients satisfy λ1 + λ2 + λ3 = 1 to ensure the normalization of the priority score, B m represents the available bandwidth of communication path m, B max represents the maximum bandwidth value of all paths in the network, Δ m represents the time required for data to be transmitted from the source node to the target node through path m, Δ max represents the maximum delay threshold allowed by the system, σ m represents the transmission stability of path m, with a value range of 0 to 1. Then, according to the real-time load of the path, the signal traffic is dynamically allocated to avoid local path overload, optimize resource utilization, and maximize the overall transmission efficiency by balancing the multi-path load to meet the fast communication requirements of collision signals. The specific formula is expressed as:

[0072]

[0073] Among them, ω m represents the traffic allocation weight, L m represents the current load of the m-th communication path, represents the maximum bearing load of the m-th communication path, k represents the index of the number of communication paths, Γ k represents the comprehensive priority score of path k, represents the maximum bearing load of the k-th communication path, L k represents the current load of the k-th communication path. By combining the load status of the edge nodes, the traffic allocation is dynamically adjusted to ensure the fast transmission of electric vehicle collision signals under complex road conditions. Then, the electric vehicle collision signals are fragmented and transmitted synchronously through multiple paths, and the multi-path bandwidth superposition effect is used to shorten the transmission time. The single-path bandwidth limit is broken through by the fragmented parallel transmission strategy, thereby improving the transmission rate. Assuming that the total size of the collision signal is S and the number of fragments is N, the specific formula for the transmission time is:

[0074]

[0075] Among them, T n represents the transmission time of fragment n, n represents the index of the number of fragments, s nThe collision signal size represented as shard n, s n = S / N, overall transmission time T total Determined by the slowest shard, the specific formula is expressed as:

[0076] T total = max{T1, T2,..., T N}

[0077] Among them, max represents the maximization function, T1, T2,..., T N respectively represent the independent shard transmission times. Secondly, by real-time monitoring the path performance, when the delay and packet loss rate of a certain path exceed the threshold, it automatically switches to the backup path to ensure the transmission continuity and avoid signal loss and delay caused by path failures. The specific formula is expressed as:

[0078]

[0079] Among them, t switch represents the trigger path switching time, B min represents the system's minimum required bandwidth, Δ th represents the delay threshold, ρ m represents the packet loss rate of the mth communication path, ρ th represents the packet loss rate threshold. Through the adaptive switching mechanism, it ensures the highly reliable transmission of the electric vehicle collision signal in the edge network. Finally, at the target edge node, the received shards are subjected to integrity verification and recombination to ensure the signal is error-free, and the end-to-end delay is reduced through distributed verification to ensure the signal integrity and real-time performance. The specific formula is expressed as:

[0080]

[0081] Among them, Hash represents the preset hash function, represents the bitwise exclusive OR operation. Based on the collaborative computing ability of the edge node, the signal recombination is quickly completed to meet the real-time requirements of electric vehicle collision processing.

[0082] Refer to Figure 1 , furthermore, the system operation and maintenance module ensures the stable operation of the system by regularly monitoring the operating states of the system hardware, sensors, and computing modules, and through the real-time fault detection and warning mechanism, it ensures that the system can quickly perform self-repair and notify the maintenance personnel when abnormalities occur, guaranteeing the reliability and continuous availability of the system.

[0083] Refer to Figure 1, Further, through a graphical interface and an easy-to-use interaction design, the user interface and the management module ensure that the user can obtain real-time system feedback and remotely monitor and adjust the system, so as to ensure that the user can receive a warning and accident details in the first time after a vehicle accident.

[0084] During specific use, first, the collision signal acquisition module is used to collect the sensor data of the electric vehicle in real time and send it to the edge computing device to ensure that a collision event can be detected in time; then, the signal preprocessing module is used to filter, denoise and enhance the collected original signal to ensure the clarity and accuracy of the data for subsequent processing; secondly, the edge computing module includes a real-time monitoring and feedback unit and a collision recognition and decision-making unit. The real-time monitoring and feedback unit is used to monitor the sensor signals and system status in real time to ensure timely feedback of collision information and initiate response measures. The collision recognition and decision-making unit proposes an electric vehicle collision detection algorithm based on edge computing to analyze the processed signals, quickly identify the collision type and make corresponding decisions to ensure efficient triggering of emergency responses; then, the communication and transmission module includes a data reception and parsing unit and a signal fast transmission unit. The data reception and parsing unit is used to receive and parse the processing results from the edge computing module and perform parsing to ensure that the collision information is accurately prepared for transmission. The signal fast transmission unit proposes a signal fast communication algorithm based on multi-path transmission to transmit the collision information through multiple communication paths simultaneously to ensure fast data transmission with low latency and high reliability in different network environments; secondly, the system operation and maintenance module is used to monitor the health status of each module, ensure the normal operation of the system and perform fault diagnosis and repair in time to ensure the stability and reliability of the system; finally, the user interface and management module is used to provide an interaction interface for the user to ensure that the user can view the collision information, receive alarms and manage system settings.

[0085] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A rapid processing and transmission system for electric vehicle collision signals based on edge computing, comprising a collision signal acquisition module, a signal preprocessing module, an edge computing module, a communication and transmission module, a system operation and maintenance module, and a user interface and management module, characterized in that: The collision signal acquisition module is used to collect the sensor data of the electric vehicle in real time and send it to the edge computing device to ensure that collision events can be detected in a timely manner; The signal preprocessing module is used to filter, denoise, and enhance the collected original signals to ensure the clarity and accuracy of the data for subsequent processing; The edge computing module includes a real-time monitoring and feedback unit and a collision recognition and decision-making unit. The real-time monitoring and feedback unit is used to monitor the sensor signals and system status in real time, and transmit the video data to the collision recognition and decision-making unit in real time, and transmit the video data to the collision recognition and decision-making unit in real time, ensuring timely feedback of collision information and initiation of response measures. The collision recognition and decision-making unit proposes an electric vehicle collision detection algorithm based on edge computing to analyze the processed signals, quickly identify the collision type and make corresponding decisions to ensure efficient triggering of emergency responses; The communication and transmission module includes a data reception and parsing unit and a signal fast transmission unit. The data reception and parsing unit is used to receive and parse the processing results from the edge computing module and perform parsing to ensure that the collision information is accurately prepared for transmission. The signal fast transmission unit proposes a signal fast communication algorithm based on multi-path transmission to transmit collision information simultaneously through multiple communication paths to ensure fast data transmission with low latency and high reliability in different network environments; The system operation and maintenance module is used to monitor the health status of each module, ensure the normal operation of the system, and perform fault diagnosis and repair in a timely manner to ensure the stability and reliability of the system; The user interface and management module is used to provide an interactive interface for users to ensure that users can view collision information, receive alarms, and manage system settings.

2. The fast processing and transmission system for electric vehicle collision signals based on edge computing according to claim 1, wherein: The collision signal acquisition module accesses various sensors of the vehicle, including acceleration sensors, collision sensors, gyroscope sensors, temperature sensors, and positioning sensor information, and sends the data to the edge computing device in real time to ensure that the vehicle's state and external collision information can be accurately collected during a collision, providing basic data for subsequent signal processing and collision analysis.

3. The fast processing and transmission system for electric vehicle collision signals based on edge computing according to claim 1, characterized in that: The signal preprocessing module processes the collected original signals by applying digital filtering technology and signal enhancement technology, including denoising, filtering, enhancing key features, and data normalization processing, to ensure that the quality of the input signals meets the system requirements and provides accurate data for subsequent processing modules to analyze and make decisions.

4. A rapid processing and transmission system for electric vehicle collision signals based on edge computing according to claim 1, characterized in that: The edge computing module includes a real-time monitoring and feedback unit and a collision recognition and decision-making unit. The real-time monitoring and feedback unit monitors the vehicle sensors and system status through edge devices, obtains video data in real time and transmits it to the collision recognition and decision-making unit, and feeds back the system status to the vehicle's safety control system to ensure that the vehicle can respond in a timely manner during a collision. The collision recognition and decision-making unit proposes an electric vehicle collision detection algorithm based on edge computing to make decisions according to the type and severity of the collision, including activating the alarm system and calling the ambulance system. The decision-making process does not rely on the cloud, ensuring low-latency real-time response and ensuring that collisions are identified efficiently and accurately.

5. The rapid processing and transmission system for electric vehicle collision signals based on edge computing according to claim 4, characterized in that, First, receive the video data of the real-time monitoring and feedback unit, and use the deep learning object detection algorithm SSD-Inception to perform object detection on each frame of the video, generating a bounding matrix containing object category and location information. The SSD-Inception algorithm extracts features in the video frame through a convolutional neural network CNN and uses the features to locate and classify objects. The specific formula is expressed as: B = Detect(I) where I represents the input video frame, B represents the bounding matrix output by the SSD-Inception algorithm, and Detect represents the deep learning object detection function. Then, construct a tracking strategy to associate the bounding matrices in each frame, generate the motion trajectory of the target, and achieve continuous tracking of the target. The tracking strategy function SORT uses a Kalman filter to predict the motion state of the target, generates a predicted bounding box, matches the predicted bounding box with the actual bounding matrix detected in the current frame, determines the target identity and updates the state by calculating the similarity matrix between targets, and achieves continuous and accurate tracking of multiple targets. The specific formula is expressed as: T t = SORT(B t , T t-1 ) Among them, represents the state prediction of target i in the t-th frame, where t represents the frame number index of the video, i represents the target number index of the detection, F represents the state transition matrix, and x t-1，i represents the state of target i in the (t - 1)-th frame, and u t，i represents the control vector, and w t，i represents the process noise. Associations represents the matching relationship between the predicted box and the detected box, represents the predicted bounding matrix in the t-th frame, IoU represents the intersection over union calculated between two bounding boxes, Hungarian represents the maximum matching function, and x t，i represents the updated state prediction of target i in the t-th frame, and y t，i represents the observation vector, KUpdate represents the update function of the Kalman filter, and T t represents the trajectory of the generated target in the t-th frame, and SORT represents the tracking function, and T t-1 represents the trajectory of the generated target in the (t - 1)-th frame. Secondly, by calculating the relative motion between the target object and the electric vehicle, the time to collision TTC is estimated. The size change rate of the bounding matrix is used to estimate TTC. The change rate of the height of the bounding matrix is calculated by linear regression, and then TTC is estimated. The specific formula is expressed as: where r high is the change rate of the boundary matrix height, h t and h t+n are the bounding box heights of the t-th frame and the (t + n)-th frame respectively, Δt is the time interval between frames, n represents the frame number index of the video. By calculating TTC in real time on the edge computing device, the collision risk can be quickly judged. By analyzing the motion pattern of the target object in the horizontal direction, it is further judged whether there is a collision risk. Then, the horizontal position change of the center point of the bounding box is used to identify the horizontal motion pattern, and the horizontal position change rate of the center point of the bounding box is calculated by linear regression, and then the motion direction of the target object is judged. The specific formula is expressed as: Among them, x center represents the horizontal position of the center point of the boundary matrix, x i and w i are respectively the abscissa and width of the upper left corner of the boundary matrix, r x represents the horizontal position change rate, x center，t+n and x center，t are respectively the horizontal positions of the center points of the bounding boxes in the (t + n)-th frame and the t-th frame. Secondly, by comprehensively considering TTC and the horizontal motion mode, it is judged whether there is a collision risk, and the collision risk information is output. The specific formula is expressed as: TTC < δ, α < r x < α' where δ represents the TTC threshold, and α and α′ are the lower and upper threshold values of the horizontal motion mode respectively. According to the TTC and the threshold of the horizontal motion mode, judge whether there is a collision risk. If the TTC is less than the preset threshold δ and the horizontal motion mode meets the preset conditions, it is judged that there is a collision risk. Finally, output the collision risk information to the subsequent module to ensure that the collision risk information can be transmitted and analyzed in a timely manner. The specific formula is expressed as: R = {TTC, r x , class, location} where R represents the collision risk information, class represents the target category information, and location represents the target location information. The electric vehicle collision detection algorithm based on edge computing realizes the accurate assessment and rapid response to the collision risk of electric vehicles through the combination of multiple advanced technologies such as deep learning object detection, real-time motion tracking, and collision time estimation. Through the edge computing-based design, it not only improves the processing efficiency of the system, reduces the dependence on cloud computing, but also ensures that it can respond quickly and improve the safety of electric vehicles when an electric vehicle collision occurs.

6. The fast processing and transmission system for electric vehicle collision signals based on edge computing according to claim 1, characterized in that The communication and transmission module includes a data reception and parsing unit and a signal fast transmission unit. The data reception and parsing unit ensures that all transmitted data can be received in a timely and accurate manner by receiving the real-time processed data output from the edge computing module, and decodes and parses through a dynamic parsing protocol to ensure that the data format and content meet the transmission standards, facilitating subsequent processing and transmission. The signal fast transmission unit proposes a signal fast communication algorithm based on multi-path transmission, providing redundant transmission methods through multiple communication paths, including Wi-Fi, 5G, and LTE, to ensure that data can still be transmitted smoothly when the network conditions are unstable and some paths fail, and ensure that the collision signal can be transmitted to relevant service platforms, including traffic management systems and accident handling centers, under low latency and high reliability, providing timely feedback and supporting accident emergency response.

7. The fast processing and transmission system for electric vehicle collision signals based on edge computing according to claim 6, characterized in that, First, dynamically screen the optimal set of transmission paths among edge computing nodes to ensure that electric vehicle collision signals are transmitted across the network with the lowest latency. Through the multi-path parallel transmission mechanism, avoid the congestion risk of a single path and improve the robustness of signal transmission. Suppose there are M candidate communication paths, and the comprehensive priority Γm of communication path m is specifically expressed by the following technical formula: Among them, Γm represents the comprehensive priority score of path m, m represents the index of the number of communication paths, λ1 represents the weight coefficient of bandwidth, representing the importance of bandwidth in the path priority score, λ2 represents the weight coefficient of delay, representing the importance of delay in the path priority score, λ3 represents the weight coefficient of stability, representing the importance of path stability in the priority score. The weight coefficients satisfy λ1 + λ2 + λ3 = 1 to ensure the normalization of the priority score, B m represents the available bandwidth of communication path m, B max represents the maximum bandwidth value of all paths in the network, Δ m represents the time required for data to be transmitted from the source node to the target node through path m, Δ max represents the maximum delay threshold allowed by the system, σ m represents the transmission stability of path m, and its value range is from 0 to 1. Then, the signal flow is dynamically allocated according to the real-time load of the path to avoid local path overload, optimize resource utilization, maximize the overall transmission efficiency by balancing the multi-path load, and meet the fast communication requirements of the collision signal. The specific formula is expressed as: Among them, ω m is expressed as the traffic allocation weight, L m is expressed as the current load of the m-th communication path, is expressed as the maximum carrying load of the m-th communication path, k is expressed as the index of the number of communication paths, Γ k is expressed as the comprehensive priority score of path k, is expressed as the maximum carrying load of the k-th communication path, L k is expressed as the current load of the k-th communication path. By combining the load status of edge nodes, the traffic allocation is dynamically adjusted to ensure the rapid transmission of electric vehicle collision signals under complex road conditions. Then, after fragmenting the electric vehicle collision signals, they are synchronously transmitted through multiple paths, and the multi-path bandwidth superposition effect is used to shorten the transmission time. The single-path bandwidth limit is broken through by the fragment parallel transmission strategy, thereby improving the transmission rate. Assume that the total size of the collision signal is S, the number of fragments is N, and the specific formula for the transmission time is expressed as: Among them, T n represents the transmission time of slice n, where n represents the index of the number of slices, and s n represents the collision signal magnitude of slice n, and s n = S / N, and the overall transmission time T total is determined by the slowest slice, and the specific formula is expressed as: T total = max{T1, T2,..., T N} where max represents the maximization function, and T1, T2, ..., T N respectively represent independent slice transmission times. Secondly, by real-time monitoring the path performance, when the delay and packet loss rate of a certain path exceed the threshold, it automatically switches to the backup path to ensure transmission continuity and avoid signal loss and delay caused by path failures. The specific formula is expressed as: Among them, t switch represents the trigger path switching time, B min represents the minimum required system bandwidth, Δ th represents the timing delay threshold, ρ m represents the packet loss rate of the m-th communication path, ρ th represents the packet loss rate threshold. Through the adaptive switching mechanism, the high-reliability transmission of the electric vehicle collision signal in the edge network is ensured. Finally, at the target edge node, the integrity verification and recombination of the received shards are performed to ensure the signal is error-free, and the end-to-end delay is reduced through distributed verification to guarantee the signal integrity and real-time performance. The specific formula is expressed as: Among them, Hash is represented as a preset hash function, is represented as a bitwise exclusive OR operation. Based on the collaborative computing ability of edge nodes, signal recombination can be quickly completed to meet the real-time requirements of electric vehicle collision handling.

8. The fast processing and transmission system for electric vehicle collision signals based on edge computing according to claim 1, wherein: The system operation and maintenance module ensures the stable operation of the system by regularly monitoring the operating status of system hardware, sensors, and computing modules, and through the real-time fault detection and warning mechanism, ensures that the system can quickly self-repair and notify maintenance personnel when anomalies occur, guaranteeing the reliability and continuous availability of the system.

9. The fast processing and transmission system for electric vehicle collision signals based on edge computing according to claim 1, wherein: The user interface and management module, through a graphical interface and user-friendly interaction design, ensures that users can obtain real-time system feedback and remotely monitor and adjust the system, ensuring that users can receive warnings and accident details in a timely manner after an accident occurs.

Images