Intelligent network control automobile control system for preventing collision of road safety vulnerable groups

Abstract

The invention discloses an intelligent network control automobile control system for preventing collision of road safety vulnerable groups, relates to the technical field, and is used for solving the problems that the emergency braking response is delayed or misjudged in actual operation and the road safety guarantee capability is reduced. A vibration feature vector is generated in combination with vehicle operation state parameters, and weighted fusion analysis is performed by adopting a living body displacement model based on a support vector machine and combining vibration spectrum features and vehicle dynamic parameters. According to matching of the vibration feature vector and a preset model, potential organism displacement behaviors are judged, the working state of a vehicle braking system is controlled, and follow-up operation is optimized in combination with the vehicle running state. According to the system, the recognition capability and classification accuracy of weak vibration signals are enhanced, the misjudgment rate and the missed judgment rate are reduced, and the timeliness and accuracy of collision early warning are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent networked vehicles and road safety technologies, and more specifically, to an intelligent networked vehicle control system for preventing collisions among vulnerable groups on the road. Background Art

[0002] With the rapid development of intelligent connected vehicle technology, the protection of vulnerable road users (such as children and the elderly) in traffic environments is gaining increasing attention. Effectively preventing collisions caused by children suddenly running out of a gap between parked vehicles, particularly in complex urban environments, has become a key area of ​​research. However, existing intelligent connected vehicle systems still have limitations in blind spot liveness detection and emergency brake failure response, impacting system reliability and safety.

[0003] The existing technology has the following deficiencies:

[0004] Currently, existing intelligent connected vehicles rely on sensors such as cameras, millimeter-wave radar, or lidar for environmental perception. However, these methods significantly reduce target recognition accuracy in blind spots (such as the narrow area between two parked vehicles), low-light conditions (such as at night or in tunnels), and inclement weather (such as rain, snow, and fog). This makes it difficult to effectively identify and warn vulnerable groups, such as children, who are small and have irregular movements.

[0005] In addition, the existing system lacks a refined vibration recognition and response mechanism for sudden contact physical disturbances (such as a sudden slight impact on the lower part of the vehicle body), which leads to delayed or misjudgment of emergency braking response in actual operation, reducing road safety assurance capabilities, increasing the probability of collision accidents and the risk of driver misoperation. Therefore, an intelligent network-controlled vehicle control system for anti-collision of vulnerable groups on road safety is proposed. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an intelligent network-controlled automobile control system for anti-collision of vulnerable groups in road safety. By embedding a distributed piezoelectric sensor array in the bumper and combining it with vehicle micro-vibration spectrum change analysis technology, it can accurately identify the displacement of biological bodies behind a stationary vehicle and realize a fast-response emergency braking function, thereby solving the technical problems of blind spot liveness detection and emergency braking failure response mentioned in the background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solutions: an intelligent network-controlled automobile control system for collision prevention for vulnerable groups in road safety, comprising a signal acquisition module, a data analysis module, a decision execution module, and a protection feedback module;

[0008] The signal acquisition module is used to collect vibration signals from the vehicle bumper area in real time and transmit the collected signals to the data analysis module;

[0009] The data analysis module extracts the spectrum features of the collected vibration signal, generates a vibration feature vector based on the vehicle operation status parameters, and transmits the vibration feature vector to the decision execution module;

[0010] The decision execution module matches the vibration eigenvector with the preset biological displacement model to determine whether there is potential biological displacement behavior, and controls the working state of the vehicle braking system based on the matching result;

[0011] The protection feedback module is used to record the action information of the decision execution module and optimize and adjust subsequent operations based on the current operating status of the vehicle.

[0012] In a preferred embodiment, the signal acquisition module includes a plurality of distributed piezoelectric sensor units, each of which is composed of a piezoelectric ceramic sheet and a signal amplification circuit. The piezoelectric ceramic sheet is fixedly mounted on an elastic support structure inside the vehicle bumper, and the signal amplification circuit is connected to the central processing unit via a flexible wire.

[0013] When the piezoelectric ceramic piece is subjected to external vibration, it generates an electrical signal. The signal amplification circuit performs preliminary amplification processing on the electrical signal and then outputs it to the data analysis module.

[0014] The elastic support structure is made of silicone.

[0015] In a preferred embodiment, the data analysis module uses a short-time Fourier transform algorithm to extract spectral features of the collected vibration signal. The specific steps are as follows:

[0016] The collected vibration signal is divided into several time windows, each of which is 200 milliseconds long and has a 50% overlap between adjacent time windows;

[0017] Apply the short-time Fourier transform formula to the vibration signal in each time window to calculate the spectrum distribution. The formula is:

[0018] Where x(n) is the time series data of the vibration signal, w(nt) is the window function, f is the frequency variable, t is the time variable, and X(t,f) is the complex spectrum result;

[0019] The main frequency component and its corresponding amplitude are extracted from the spectrum results as spectrum features, and the spectrum features are combined with the vehicle operating state parameters to generate a vibration feature vector.

[0020] In a preferred embodiment, the vehicle operating status parameters include vehicle speed, acceleration, and steering wheel angle, which are obtained through the vehicle CAN bus. The data analysis module weightedly fuses the spectral features with the vehicle operating status parameters according to preset weights to obtain the final vibration feature vector.

[0021] In a preferred embodiment, the biological displacement model preset in the decision execution module is constructed based on the support vector machine algorithm, and the sample data used in the model training process includes vibration feature vectors in different scenarios and their corresponding behavior labels;

[0022] The support vector machine algorithm maps the vibration feature vector to a high-dimensional space through a kernel function and searches for the optimal classification hyperplane in the high-dimensional space. The expression of the classification hyperplane is f(x)=w T φ(x)+b;

[0023] Where x is the vibration feature vector, φ(x) is the weight vector, b is the bias term, and f(x) is the classification result;

[0024] When the classification result indicates that there is potential biological displacement behavior, the decision execution module sends a braking command to the vehicle braking system.

[0025] In a preferred embodiment, when the decision execution module sends a braking command to the vehicle braking system, the braking intensity is dynamically adjusted in combination with the current vehicle speed. The specific adjustment method is as follows:

[0026] If the vehicle speed is less than 20 kilometers per hour, the braking intensity is set to 30% of the maximum braking force;

[0027] If the vehicle speed is between 20 km / h and 40 km / h, the braking intensity is set to 50% of the maximum braking force;

[0028] If the vehicle speed is higher than 40 km / h, the braking intensity is set to 70% of the maximum braking force;

[0029] The braking command is transmitted to the electronic control unit through the vehicle's CAN bus, and the electronic control unit controls the pressure regulating valve of the hydraulic brake system according to the received braking intensity command.

[0030] In a preferred embodiment, the protection feedback module records the action information of the decision execution module, including the triggering time, duration and actual braking intensity of the braking instruction, and stores the recorded information in a non-volatile memory;

[0031] The protection feedback module optimizes and adjusts subsequent operations based on the vehicle's current operating status. The specific methods are as follows:

[0032] If the braking command is triggered three times in succession and no actual collision event is detected, the sensitivity threshold of the organism displacement model is reduced;

[0033] If the braking command is not triggered for five consecutive times but a collision event occurs, the sensitivity threshold of the organism displacement model is increased;

[0034] The adjustment range of the sensitivity threshold is limited to ±20% of the initial value.

[0035] In a preferred embodiment, the distributed piezoelectric sensor units in the signal acquisition module are evenly distributed along the length of the vehicle bumper, and the spacing between adjacent piezoelectric sensor units is 10 cm to 15 cm;

[0036] The signal amplification circuit of each piezoelectric sensor unit adopts a differential amplifier design, and the gain of the differential amplifier is set to 10 times.

[0037] In a preferred embodiment, after generating the vibration feature vector, the data analysis module uses a principal component analysis algorithm to perform dimensionality reduction processing on the vibration feature vector. The specific steps are as follows:

[0038] Calculate the covariance matrix of the vibration eigenvector and solve the eigenvalues ​​and eigenvectors of the covariance matrix;

[0039] Select the eigenvectors corresponding to the first k largest eigenvalues ​​as the principal components, where the value of k satisfies the cumulative contribution rate of not less than 95%;

[0040] The vibration eigenvector is projected into the principal component space to obtain the vibration eigenvector after dimensionality reduction.

[0041] In a preferred embodiment, the decision execution module introduces a confidence assessment mechanism when determining whether there is potential biological displacement behavior. The specific steps are as follows:

[0042] Calculate the distance between the vibration feature vector and the classification hyperplane. The smaller the distance, the higher the confidence level.

[0043] When the confidence level is lower than the preset threshold, the decision execution module issues a warning sound through the vehicle alarm system;

[0044] When the confidence level is higher than the preset threshold, the decision execution module sends a braking command and triggers the vehicle alarm system at the same time.

[0045] The technical effects and advantages of the present invention are as follows:

[0046] 1. The present invention collects vibration signals from the vehicle bumper area in real time, generates vibration feature vectors based on vehicle operating status parameters, and adopts a biological displacement model based on a support vector machine. This method combines weighted fusion analysis of vibration spectrum characteristics with vehicle dynamic parameters, matches the vibration feature vectors with a preset biological displacement model, determines whether there is potential biological displacement behavior, controls the operating state of the vehicle braking system based on the matching results, and optimizes and adjusts subsequent operations based on the vehicle's current operating status. This enhances the recognition ability and classification accuracy of weak vibration signals, reduces the false positive rate and missed positive rate, and improves the timeliness and accuracy of collision warnings. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 The present invention is a schematic diagram of the module structure of an intelligent network-controlled automobile control system for anti-collision of vulnerable groups in road safety.

[0048] Figure 2 The present invention is a schematic diagram of the installation structure of a distributed piezoelectric sensor unit in an intelligent network-controlled automobile control system for anti-collision of vulnerable groups in road safety. DETAILED DESCRIPTION

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0050] Example 1

[0051] The present invention provides an intelligent network-controlled vehicle control system for preventing collisions among vulnerable groups in road safety. Figure 1 and Figure 2 Provide detailed explanation.

[0052] like Figure 1 As shown, the system includes a signal acquisition module, a data analysis module, a decision execution module and a protection feedback module. These modules realize data interaction and functional collaboration through the vehicle CAN bus.

[0053] The distributed piezoelectric sensor unit in the signal acquisition module is installed inside the vehicle bumper to sense external vibration signals and transmit the signals to the data analysis module;

[0054] The data analysis module processes the signal and generates a vibration feature vector which is then transmitted to the decision execution module;

[0055] The decision execution module determines whether there is potential biological displacement behavior based on the vibration eigenvector and controls the working state of the vehicle braking system;

[0056] The protection feedback module records the system's action information and optimizes and adjusts subsequent operations.

[0057] The specific structure of the signal acquisition module is as follows Figure 2 As shown, its core component is a piezoelectric ceramic sheet, which is fixed inside the vehicle bumper through an elastic support structure.

[0058] The elastic support structure is made of silicone material with a thickness ranging from 3mm to 5mm. It can absorb high-frequency noise while enhancing the sensitivity of signal acquisition.

[0059] Each piezoelectric ceramic piece is connected to a signal amplification circuit. The signal amplification circuit adopts a differential amplifier design with a gain set to 10 times to suppress common-mode noise and improve signal quality.

[0060] The signal amplification circuit is connected to the central processing unit through a flexible wire, and transmits the initially amplified electrical signal to the data analysis module.

[0061] The distributed piezoelectric sensor units are evenly distributed along the length of the vehicle bumper, with a spacing of 10cm to 15cm between adjacent piezoelectric sensor units, ensuring coverage of the entire bumper area.

[0062] After receiving the vibration signal transmitted by the signal acquisition module, the data analysis module uses the short-time Fourier transform algorithm to extract the spectrum features of the signal.

[0063] First, the vibration signal is divided into several time windows, each with a length of 200ms, and there is a 50% overlap between adjacent time windows.

[0064] Then, the short-time Fourier transform formula is applied to the vibration signal in each time window to calculate the spectrum distribution, which is:

[0065] Where x(n) is the time series data of the vibration signal, w(nt) is the window function, f is the frequency variable, t is the time variable, and X(t,f) is the complex spectrum result.

[0066] The main frequency component and its corresponding amplitude are extracted from the spectrum results as spectrum features, and the spectrum features are combined with the vehicle operating state parameters to generate a vibration feature vector.

[0067] Vehicle operating status parameters include vehicle speed, acceleration, and steering wheel angle, which are obtained through the vehicle's CAN bus.

[0068] The data analysis module performs weighted fusion of the spectrum features and the vehicle operating status parameters according to preset weights to obtain the final vibration feature vector.

[0069] In order to reduce the dimension of the vibration eigenvector, the data analysis module further uses the principal component analysis algorithm to perform dimensionality reduction processing on it. The specific steps include calculating the covariance matrix of the vibration eigenvector and solving the eigenvalues ​​and eigenvectors of the covariance matrix, selecting the eigenvectors corresponding to the first k largest eigenvalues ​​as the principal components, where the value of k satisfies the cumulative contribution rate of not less than 95%, and finally projecting the vibration eigenvector into the principal component space to obtain the vibration eigenvector after dimensionality reduction.

[0070] The decision execution module builds a biological displacement model based on the support vector machine algorithm. The sample data used in the model training process includes vibration feature vectors in different scenarios and their corresponding behavior labels.

[0071] The support vector machine algorithm maps the vibration feature vector to a high-dimensional space through a kernel function and searches for the optimal classification hyperplane in the high-dimensional space. The expression of the classification hyperplane is f(x)=w T φ(x)+b;

[0072] Among them, x is the vibration feature vector, φ(x) is the weight vector, b is the bias term, and f(x) is the classification result.

[0073] When the classification result indicates that there is potential biological displacement behavior, the decision execution module sends a braking command to the vehicle braking system.

[0074] To avoid misjudgment, the decision execution module introduces a confidence evaluation mechanism to calculate the distance between the vibration feature vector and the classification hyperplane. The smaller the distance, the higher the confidence.

[0075] When the confidence level is lower than the preset threshold, the decision execution module does not send a braking command but instead issues a warning sound through the vehicle alarm system;

[0076] When the confidence level is higher than the preset threshold, the decision execution module sends a braking command and triggers the vehicle alarm system at the same time.

[0077] The braking command is transmitted to the electronic control unit through the vehicle's CAN bus, and the electronic control unit controls the pressure regulating valve of the hydraulic brake system according to the received braking intensity command.

[0078] The braking intensity is dynamically adjusted according to the vehicle's current speed. When the speed is lower than 20km / h, the braking intensity is set to 30% of the maximum braking force; when the speed is between 20km / h and 40km / h, the braking intensity is set to 50% of the maximum braking force; when the speed is higher than 40km / h, the braking intensity is set to 70% of the maximum braking force.

[0079] The protection feedback module records the action information of the decision execution module, including the triggering time, duration and actual braking intensity of the braking instruction, and stores the recorded information in a non-volatile memory.

[0080] The protection feedback module optimizes and adjusts subsequent operations based on the vehicle's current operating status. If the braking command is triggered three times in a row and no actual collision event is detected, the sensitivity threshold of the organism displacement model is lowered;

[0081] If the braking command is not triggered for five consecutive times but a collision event occurs, the sensitivity threshold of the biological displacement model is increased; the adjustment range of the sensitivity threshold is limited to within ±20% of the initial value.

[0082] In addition, the protection feedback module also considers the influence of environmental factors. It obtains the temperature and humidity data of the current environment through the on-board temperature and humidity sensor. When the temperature is below 0°C or the humidity is above 80%, the sensitivity threshold of the organism displacement model is appropriately increased. The light intensity data of the current environment is obtained through the on-board light sensor. When the light intensity is below 100 lux, the sensitivity threshold of the organism displacement model is appropriately lowered.

[0083] In this embodiment, the connection relationship and functional interaction process between the signal acquisition module, data analysis module, decision execution module and protection feedback module are as follows: Figure 1 shown.

[0084] The signal acquisition module senses the external vibration signal through the distributed piezoelectric sensor unit and transmits the signal to the data analysis module.

[0085] The data analysis module processes the signal and generates a vibration feature vector which is then transmitted to the decision execution module.

[0086] The decision execution module determines whether there is potential biological displacement behavior based on the vibration eigenvector and controls the working state of the vehicle braking system.

[0087] The protection feedback module records the system's action information and optimizes and adjusts subsequent operations.

[0088] Data interaction between modules is achieved through the vehicle CAN bus, ensuring the real-time and reliability of system operation.

[0089] In order to better enable relevant personnel in this technical field to fully understand and implement the present invention, the specific implementation principle of the present invention is supplemented below with reference to a specific application scenario.

[0090] In an urban road environment, when a vehicle is parked on the roadside, there may be children or the elderly, who are vulnerable to road safety, moving behind the stationary vehicle.

[0091] At this time, if a child suddenly rushes out from the gap between stationary vehicles, the driver may not be able to take braking measures in time due to blind spots.

[0092] The present invention uses a distributed piezoelectric sensor array and vibration characteristic analysis technology to quickly identify and trigger the emergency braking function in the early stages of biological displacement behavior, thereby effectively avoiding the occurrence of collision accidents.

[0093] First, multiple distributed piezoelectric sensor units in the signal acquisition module are installed inside the vehicle bumper, such as Figure 2 shown.

[0094] These piezoelectric ceramic discs are fixed to the inside of the bumper by an elastic support structure and are arranged in such a way as to ensure uniform coverage of the entire bumper area.

[0095] When a child approaches the rear of a stationary vehicle and generates a small vibration, the piezoelectric ceramic plate generates an electrical signal due to the external vibration.

[0096] Since the elastic support structure is made of silicone material with a thickness ranging from 3mm to 5mm, it can effectively absorb high-frequency noise and enhance the sensitivity of signal acquisition.

[0097] The initially generated electrical signal is processed by a signal amplification circuit designed as a differential amplifier, with the gain set to 10 times to suppress common-mode noise, and then transmitted to the central processing unit via a flexible wire.

[0098] Next, the data analysis module receives the vibration signal from the signal acquisition module and extracts its spectrum features.

[0099] Specifically, the vibration signal is divided into several time windows, each time window is 200 ms long, and there is a 50% overlap between adjacent time windows.

[0100] By applying the short-time Fourier transform algorithm to the signal in each time window, its spectrum distribution is calculated and the main frequency component and its amplitude are extracted as spectrum features.

[0101] At the same time, the data analysis module obtains vehicle operating status parameters such as speed, acceleration, and steering wheel angle through the on-board CAN bus, and weightedly fuses these parameters with the spectrum characteristics according to preset weights to generate the final vibration feature vector.

[0102] In order to reduce the dimension of the vibration eigenvector, the data analysis module further uses the principal component analysis algorithm to perform dimensionality reduction processing on it, selects the top k principal components with a cumulative contribution rate of not less than 95%, and finally obtains the vibration eigenvector after dimensionality reduction.

[0103] Subsequently, the decision execution module performs matching analysis on the received vibration feature vector based on the biological displacement model constructed by the support vector machine algorithm.

[0104] The model maps the vibration feature vector to a high-dimensional space through a kernel function and searches for the optimal classification hyperplane in the high-dimensional space.

[0105] When the classification results indicate the existence of potential biological displacement behavior, the decision execution module introduces a confidence evaluation mechanism to calculate the distance between the vibration feature vector and the classification hyperplane.

[0106] If the distance is small and the confidence level is higher than a preset threshold, the decision execution module determines it as a high-confidence event and sends a braking command to the vehicle's braking system.

[0107] At this time, the braking intensity is dynamically adjusted according to the vehicle's current speed: when the speed is lower than 20km / h, the braking intensity is set to 30% of the maximum braking force; when the speed is between 20km / h and 40km / h, the braking intensity is set to 50% of the maximum braking force; when the speed is higher than 40km / h, the braking intensity is set to 70% of the maximum braking force.

[0108] The braking command is transmitted to the electronic control unit through the vehicle's CAN bus, and the electronic control unit controls the pressure regulating valve of the hydraulic braking system to achieve the emergency braking function.

[0109] At the same time, the protection feedback module records the action information of the decision execution module, including the trigger time, duration and actual braking intensity of the braking command, and stores this information in the non-volatile memory.

[0110] If the braking command is triggered three times in a row but no actual collision event is detected, the protection feedback module appropriately lowers the sensitivity threshold of the organism displacement model to reduce the possibility of misjudgment; if the braking command is not triggered five times in a row but a collision event occurs, the sensitivity threshold of the organism displacement model is increased to improve the system's responsiveness.

[0111] In addition, the protection feedback module also optimizes and adjusts the model based on environmental factors. For example, it obtains ambient temperature and humidity data through the vehicle's temperature and humidity sensors. When the temperature is below 0°C or the humidity is above 80%, the model's sensitivity threshold is appropriately increased.

[0112] The vehicle-mounted light sensor is used to obtain ambient light intensity data. When the light intensity is lower than 100 lux, the sensitivity threshold of the model is appropriately lowered.

[0113] In summary, the present invention uses a distributed piezoelectric sensor array to sense external vibration signals, and combines spectrum feature extraction, support vector machine algorithm, and dynamic braking control strategy to achieve accurate recognition and rapid response to the displacement behavior of organisms behind a stationary vehicle.

[0114] The modules collaborate through the vehicle's CAN bus, ensuring real-time and reliable system operation. This technical solution not only significantly improves the safety of intelligent network-controlled vehicles in complex urban road environments, but also provides more comprehensive protection for vulnerable groups on the road.

[0115] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0116] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0117] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0118] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0119] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0120] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0121] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0122] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0123] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0124] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. An intelligent network-controlled vehicle control system for collision prevention for vulnerable groups on road safety, characterized by: It includes signal acquisition module, data analysis module, decision execution module and protection feedback module; The signal acquisition module is used to collect vibration signals from the vehicle bumper area in real time and transmit the collected signals to the data analysis module; The data analysis module extracts the spectrum features of the collected vibration signal, generates a vibration feature vector based on the vehicle operation status parameters, and transmits the vibration feature vector to the decision execution module; The decision execution module matches the vibration eigenvector with the preset biological displacement model to determine whether there is potential biological displacement behavior, and controls the working state of the vehicle braking system based on the matching result; The protection feedback module is used to record the action information of the decision execution module and optimize and adjust subsequent operations based on the current operating status of the vehicle.

2. The intelligent network-controlled vehicle control system for preventing collisions for vulnerable groups on road safety according to claim 1, characterized in that: The signal acquisition module includes multiple distributed piezoelectric sensor units. Each piezoelectric sensor unit consists of a piezoelectric ceramic piece and a signal amplification circuit. The piezoelectric ceramic piece is fixedly mounted on the elastic support structure inside the vehicle bumper. The signal amplification circuit is connected to the central processing unit via a flexible wire. When the piezoelectric ceramic piece is subjected to external vibration, it generates an electrical signal. The signal amplification circuit performs preliminary amplification processing on the electrical signal and then outputs it to the data analysis module. The elastic support structure is made of silicone.

3. The intelligent network-controlled vehicle control system for preventing collisions for vulnerable groups on road safety according to claim 2, characterized in that: The data analysis module uses the short-time Fourier transform algorithm to extract the spectrum features of the collected vibration signal. The specific steps are as follows: The collected vibration signal is divided into several time windows, each of which is 200 milliseconds long and has a 50% overlap between adjacent time windows; Apply the short-time Fourier transform formula to the vibration signal in each time window to calculate the spectrum distribution. The formula is: Where x(n) is the time series data of the vibration signal, w(nt) is the window function, f is the frequency variable, t is the time variable, and X(t,f) is the complex spectrum result; The main frequency component and its corresponding amplitude are extracted from the spectrum results as spectrum features, and the spectrum features are combined with the vehicle operating state parameters to generate a vibration feature vector.

4. The intelligent network-controlled vehicle control system for collision prevention for vulnerable groups on road safety according to claim 3, characterized in that: Vehicle operating status parameters include speed, acceleration, and steering wheel angle, which are obtained through the on-board CAN bus. The data analysis module weightedly fuses the spectral characteristics with the vehicle operating status parameters according to preset weights to obtain the final vibration feature vector.

5. The intelligent network-controlled vehicle control system for anti-collision of vulnerable groups on road safety according to claim 1 is characterized by: The biological displacement model preset in the decision execution module is built based on the support vector machine algorithm. The sample data used in the model training process includes vibration feature vectors in different scenarios and their corresponding behavior labels; The support vector machine algorithm maps the vibration feature vector to a high-dimensional space through a kernel function and searches for the optimal classification hyperplane in the high-dimensional space. The expression of the classification hyperplane is f(x)=w T φ(x)+b; Where x is the vibration feature vector, φ(x) is the weight vector, b is the bias term, and f(x) is the classification result; When the classification result indicates that there is potential biological displacement behavior, the decision execution module sends a braking command to the vehicle braking system.

6. The intelligent network-controlled vehicle control system for preventing collisions for vulnerable groups on road safety according to claim 5, characterized in that: When the decision execution module sends a braking command to the vehicle's braking system, it dynamically adjusts the braking intensity based on the vehicle's current speed. The specific adjustment method is as follows: If the vehicle speed is less than 20 kilometers per hour, the braking intensity is set to 30% of the maximum braking force; If the vehicle speed is between 20 km / h and 40 km / h, the braking intensity is set to 50% of the maximum braking force; If the vehicle speed is higher than 40 km / h, the braking intensity is set to 70% of the maximum braking force; The braking command is transmitted to the electronic control unit through the vehicle's CAN bus, and the electronic control unit controls the pressure regulating valve of the hydraulic brake system according to the received braking intensity command.

7. The intelligent network-controlled vehicle control system for collision prevention for vulnerable groups on road safety according to claim 6, characterized in that: The protection feedback module records the action information of the decision execution module, including the triggering time, duration and actual braking intensity of the braking command, and stores the recorded information in a non-volatile memory; The protection feedback module optimizes and adjusts subsequent operations based on the vehicle's current operating status. The specific methods are as follows: If the braking command is triggered three times in succession and no actual collision event is detected, the sensitivity threshold of the organism displacement model is reduced; If the braking command is not triggered for five consecutive times but a collision event occurs, the sensitivity threshold of the organism displacement model is increased; The adjustment range of the sensitivity threshold is limited to ±20% of the initial value.

8. The intelligent network-controlled vehicle control system for collision prevention for vulnerable groups on road safety according to claim 2, characterized in that: The distributed piezoelectric sensor units in the signal acquisition module are evenly distributed along the length of the vehicle bumper, and the spacing between adjacent piezoelectric sensor units is 10 cm to 15 cm; The signal amplification circuit of each piezoelectric sensor unit adopts a differential amplifier design, and the gain of the differential amplifier is set to 10 times.

9. The intelligent network-controlled vehicle control system for preventing collisions for vulnerable groups on road safety according to claim 3, characterized in that: After generating the vibration feature vector, the data analysis module uses the principal component analysis algorithm to reduce the dimension of the vibration feature vector. The specific steps are as follows: Calculate the covariance matrix of the vibration eigenvector and solve the eigenvalues ​​and eigenvectors of the covariance matrix; Select the eigenvectors corresponding to the first k largest eigenvalues ​​as the principal components, where the value of k satisfies the cumulative contribution rate of not less than 95%; The vibration eigenvector is projected into the principal component space to obtain the vibration eigenvector after dimensionality reduction.

10. The intelligent network-controlled vehicle control system for collision prevention for vulnerable groups on road safety according to claim 5, characterized in that: The decision execution module introduces a confidence assessment mechanism when determining whether there is potential biological displacement behavior. The specific steps are as follows: Calculate the distance between the vibration feature vector and the classification hyperplane. The smaller the distance, the higher the confidence level. When the confidence level is lower than the preset threshold, the decision execution module issues a warning sound through the vehicle alarm system; When the confidence level is higher than the preset threshold, the decision execution module sends a braking command and triggers the vehicle alarm system at the same time.

Images