Vehicle anti-collision safety system

Abstract

The invention discloses an anti-collision safety system for a vehicle. The anti-collision safety system comprises a storage module for storing a plurality of gathering place coordinate ranges; the audio acquisition module acquires environment audio data in an external environment in real time according to the audio acquisition instruction; the first analysis module performs similarity analysis on the environment audio data and the standard audio data to generate an infrared detection instruction; the thermal infrared detection module detects a thermal infrared contour image around the vehicle in real time according to the infrared detection instruction and analyzes the thermal infrared contour image to obtain a matching instruction; the matching module performs matching according to the matching instruction to obtain wild animal motion parameters; the detection module detects relative motion parameters; the prediction module inputs the animal coordinates, the vehicle coordinates, the wild animal motion parameters and the relative motion parameters into a collision risk prediction model to obtain a predicted collision risk value; and the vehicle control module dynamically adjusts the power output power, the braking force and the steering angle of the vehicle according to the predicted collision risk value. According to the invention, the safety of avoiding wild animals outside the visual field range is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle anti-collision, and in particular to a vehicle anti-collision safety system. Background Art

[0002] In mountainous areas where wild animals gather, wild animals often suddenly cross the road of the vehicle. If the driver fails to avoid them in a short time, they may collide or roll over, causing serious hidden dangers to the life safety of the driver and the wild animals. The vehicle anti-collision system in the prior art can only identify wild animals within the shooting range of the vehicle camera. It cannot quickly avoid animals that are not within the detection range in the current state, which leads to traffic accidents. Therefore, the vehicle anti-collision system in the prior art is not safe for avoiding wild animals outside the field of view. Summary of the invention

[0003] In view of the deficiencies in the prior art, an object of the present invention is to provide a vehicle collision avoidance safety system for improving the safety of avoiding wild animals outside the field of vision.

[0004] To achieve the above object, the present invention provides the following technical solution: a vehicle anti-collision safety system, comprising:

[0005] A storage module, used for storing the coordinate ranges of a plurality of wild animal gathering places and a matching database, wherein the coordinate ranges of the gathering places are associated with standard images and standard audio data of a plurality of wild animals;

[0006] A positioning module, connected to the storage module, for locating the vehicle coordinates in real time during the vehicle's travel, and generating an audio collection instruction when the vehicle coordinates are within the coordinate range of the gathering place;

[0007] An audio acquisition module, used to acquire ambient audio data in an external environment in real time according to the audio acquisition instruction;

[0008] A first analysis module, connected to the audio acquisition module and the storage module, for performing similarity analysis on the ambient audio data within a preset time period and the standard audio data to obtain an audio deviation, and generating an infrared detection instruction when the audio deviation is less than a preset first deviation threshold;

[0009] a thermal infrared detection module, connected to the first analysis module and the storage module, for detecting a thermal infrared contour image around the vehicle in real time according to the infrared detection instruction, performing a similarity analysis between the thermal infrared contour image and the standard image to obtain an image comprehensive deviation, and generating a matching instruction when the image comprehensive deviation is less than a preset second deviation threshold;

[0010] A matching module, connected to the thermal infrared detection module and the storage module, respectively, for matching the standard image in the matching database to obtain corresponding wild animal motion parameters according to the matching instruction;

[0011] A motion detection module is used to detect the relative motion parameters between wild animals and vehicles at the current moment;

[0012] A prediction module, connected to the thermal infrared detection module, the positioning module, the matching module and the motion detection module, respectively, for performing position analysis on the thermal infrared profile image according to the matching instruction to obtain corresponding animal coordinates, and then inputting the animal coordinates, the vehicle coordinates, the wild animal motion parameters and the relative motion parameters into a preset collision risk prediction model to predict the predicted collision risk value between the wild animal and the vehicle after the next moment;

[0013] A vehicle control module is connected to the prediction module and is used to dynamically adjust the vehicle's power output, braking force and steering angle according to the predicted collision risk value until the predicted collision risk value is lower than a preset risk threshold.

[0014] Furthermore, the wild animal movement parameters include average running speed, average running acceleration and average weight, and the relative movement parameters include relative distance, relative speed and relative acceleration.

[0015] Furthermore, the first analysis module includes:

[0016] An audio preprocessing unit, used for performing noise reduction and normalization processing on the environmental audio data in sequence to obtain preprocessed audio data;

[0017] a first extraction unit, connected to the audio preprocessing unit, configured to extract a plurality of environmental audio features from the preprocessed audio data, and to extract a plurality of standard audio features from the standard audio data;

[0018] A first calculation unit is connected to the first extraction unit and is used to calculate the audio sub-deviation between each of the environmental audio features and the corresponding standard audio features according to a preset first similarity measurement algorithm, and then obtain the audio deviation degree based on a comprehensive calculation of each of the audio sub-deviation degrees.

[0019] Furthermore, the environmental audio features include environmental timbre, environmental pitch and environmental rhythm, and the standard audio features include animal timbre, animal pitch and animal rhythm, then the audio sub-deviation degree includes timbre deviation degree, pitch deviation degree and rhythm deviation degree, and the first calculation unit inputs the timbre deviation degree, the pitch deviation degree and the rhythm deviation degree into a preset audio deviation calculation formula to calculate the audio deviation degree.

[0020] Furthermore, the audio deviation calculation formula is configured as:

[0021]

[0022] Among them, A dd Used to indicate the audio deviation, T dd It is used to indicate the timbre deviation, P dd Used to indicate the pitch deviation, R dd It is used to represent the rhythm deviation degree, and ζ1, ζ2, and ζ3 are used to represent the preset first deviation coefficient, second deviation coefficient, and third deviation coefficient, respectively.

[0023] Furthermore, the first analysis module further includes a weather acquisition unit, which is used to acquire in real time environmental weather data of the external environment corresponding to the vehicle coordinates, and obtain corresponding interference audio according to the environmental weather data analysis, and then obtain corresponding coefficient adjustment values ​​according to each of the processing;

[0024] The first calculation unit is connected to the weather acquisition unit, and is used for dynamically adjusting the first deviation coefficient, the second deviation coefficient and the third deviation coefficient according to each of the coefficient adjustment values.

[0025] Furthermore, the thermal infrared detection module comprises:

[0026] An image preprocessing unit, used for sequentially performing grayscale conversion, image enhancement, image filtering and binarization processing on the thermal infrared profile image to obtain a preprocessed image;

[0027] A second extraction unit, connected to the image preprocessing unit, for extracting a plurality of environmental image features from the preprocessed image and a plurality of animal image features from the standard image;

[0028] A second calculation unit is connected to the second extraction unit and is used to calculate the image sub-deviation between each of the environmental image features and the corresponding animal image features according to a preset second similarity measurement algorithm, and then obtain the image comprehensive deviation based on the comprehensive calculation of each of the image sub-deviations.

[0029] Furthermore, the environmental image features include environmental image edges, environmental image textures and environmental image shapes, and the animal image features include animal image edges, animal image textures and animal image shapes, then the image sub-deviations include edge deviations, texture deviations and shape deviations, and the second calculation unit inputs the edge deviations, the texture deviations and the shape deviations into a preset image deviation calculation formula to calculate the comprehensive image deviations.

[0030] Furthermore, the image deviation calculation formula is configured as:

[0031]

[0032] Among them, L cd It is used to represent the comprehensive deviation of the image, E dd Used to express the edge deviation, T dd It is used to represent the texture deviation, S dd It is used to represent the shape deviation, and χ1, χ2, and χ3 are respectively used to represent the preset fourth deviation coefficient, fifth deviation coefficient, and sixth deviation coefficient.

[0033] Furthermore, the vehicle control module includes:

[0034] A control storage unit, used for storing a plurality of collision risk ranges, each of which is associated with a corresponding upper output power limit, a lower power limit, and a lower steering angle limit;

[0035] a matching unit, connected to the control storage unit, for determining the collision risk range in which the predicted collision risk value is located, and then matching within the collision risk range to obtain the upper limit output power, the lower limit power, and the lower limit steering angle under the current state of the vehicle;

[0036] a third calculation unit, connected to the matching unit, for calculating the power difference between the upper limit output power and the power output power of the vehicle in a current state, calculating the braking difference between the braking force and the lower limit power, and calculating the angle difference between the steering angle and the lower limit steering angle;

[0037] An adjustment unit, connected to the third calculation unit, is used to dynamically adjust the power output power, the braking force and the steering angle of the vehicle according to the power difference, the braking difference and the angle difference, until the predicted collision risk value is lower than the risk threshold.

[0038] Beneficial effects of the present invention:

[0039] The present invention locates the coordinates of the vehicle in real time during the outdoor driving of the vehicle, and collects environmental audio data and thermal infrared profile images when the vehicle coordinates are within the coordinate range of the gathering place, and performs progressive detection and analysis on the types of wild animals outside the field of vision by comparing the audio deviation between the environmental audio data and the standard audio data, and comparing the comprehensive image deviation between the thermal infrared profile image and the standard image, thereby ensuring the detection accuracy while improving the detection speed; and then detects the relative motion parameters of the wild animal and the vehicle at the current moment, so as to input the animal coordinates, vehicle coordinates, wild animal motion parameters, and relative motion parameters into a collision risk prediction model, and predict the predicted collision risk value between the wild animal and the vehicle after the next moment, and finally dynamically adjusts the power output power, braking force and steering angle of the vehicle according to the predicted collision risk value, so as to realize the automatic avoidance of the vehicle to the wild animal outside the field of vision, thereby improving the avoidance safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a structural schematic diagram of the vehicle anti-collision safety system of the present invention;

[0041] Figure 2 is a schematic diagram of the structure of the first analysis module in the present invention;

[0042] Figure 3 It is a schematic diagram of the structure of the thermal infrared detection module in the present invention;

[0043] Figure 4 It is a structural schematic diagram of the vehicle control module in the present invention.

[0044] Figure numerals: 1. storage module; 2. positioning module; 3. audio acquisition module; 4. first analysis module; 41. audio preprocessing unit; 42. first extraction unit; 43. first calculation unit; 44. weather acquisition unit; 5. thermal infrared detection module; 51. image preprocessing unit; 52. second extraction unit; 53. second calculation unit; 6. matching module; 7. motion detection module; 8. prediction module; 9. vehicle control module; 91. control storage unit; 92. matching unit; 93. third calculation unit; 94. adjustment unit. DETAILED DESCRIPTION

[0045] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. The same parts are represented by the same reference numerals. It should be noted that the words "front", "rear", "left", "right", "upper" and "lower" used in the following description refer to directions in the accompanying drawings, and the words "bottom surface" and "top surface", "inner" and "outer" refer to directions toward or away from the geometric center of a specific component, respectively.

[0046] Example 1, reference Figure 1, which is the first embodiment of the present invention, provides a vehicle collision avoidance safety system, which can improve the safety of avoiding wild animals outside the field of vision, including:

[0047] Storage module 1, used for storing the coordinate ranges of several gathering places of wild animals and a matching database, wherein the coordinate ranges of the gathering places are associated with standard images and standard audio data of several wild animals;

[0048] The positioning module 2 is connected to the storage module 1 and is used to locate the vehicle coordinates in real time during the vehicle driving process, and to generate an audio collection instruction when the vehicle coordinates are within the coordinate range of the gathering place;

[0049] The audio acquisition module 3 is used to acquire the ambient audio data in the external environment in real time according to the audio acquisition instruction;

[0050] The first analysis module 4 is connected to the audio acquisition module 3 and the storage module 1, and is used to perform similarity analysis on the ambient audio data and the standard audio data within a preset time period to obtain an audio deviation, and generate an infrared detection instruction when the audio deviation is less than a preset first deviation threshold;

[0051] The thermal infrared detection module 5 is connected to the first analysis module 4 and the storage module 1, and is used to detect the thermal infrared contour image around the vehicle in real time according to the infrared detection instruction, and perform similarity analysis between the thermal infrared contour image and the standard image to obtain the image comprehensive deviation, and generate a matching instruction when the image comprehensive deviation is less than a preset second deviation threshold;

[0052] A matching module 6, connected to the thermal infrared detection module 5 and the storage module 1, respectively, for matching the standard image in the matching database according to the matching instruction to obtain the corresponding wild animal movement parameters;

[0053] A motion detection module 7, used to detect the relative motion parameters between the wild animal and the vehicle at the current moment;

[0054] The prediction module 8 is respectively connected to the thermal infrared detection module 5, the positioning module 2, the matching module 6 and the motion detection module 7, and is used to perform position analysis on the thermal infrared profile image according to the matching instruction to obtain the corresponding animal coordinates, and then input the animal coordinates, vehicle coordinates, wild animal motion parameters, and relative motion parameters into a preset collision risk prediction model to predict the predicted collision risk value between the wild animal and the vehicle after the next moment;

[0055] The vehicle control module 9 is connected to the prediction module 8 and is used to dynamically adjust the vehicle's power output, braking force and steering angle according to the predicted collision risk value until the predicted collision risk value is lower than a preset risk threshold.

[0056] Working principle of embodiment 1:

[0057] The storage module 1 may be a non-volatile memory in the vehicle's on-board computer, used to store the gathering place coordinate range of the gathering place of wild animals and the data of the matching database. The positioning module 2 may be a GPS positioning device installed on the vehicle. The audio acquisition module 3 may be a recording device installed on the outside of the vehicle, which collects environmental audio data during the driving process of the vehicle. The first analysis module 4 may be an audio data analysis program installed on the on-board computer, which is used to perform similarity analysis on the environmental audio data within 30 seconds and the standard audio data to obtain the audio deviation, and when the audio deviation is less than the preset first deviation threshold, it indicates that the audio similarity between the environmental audio data and the wild animals is high at this time, so an infrared detection instruction is generated for further detection. Among them, the thermal infrared detection module 5 may be a thermal infrared detection device with a data processing chip, which is installed on the outside of the vehicle, and is used to detect the thermal infrared contour image around the vehicle in real time after receiving the infrared detection instruction, and use the data processing chip to perform similarity analysis on the thermal infrared contour image and the standard image to obtain the image comprehensive deviation, and when the image comprehensive deviation is less than the second deviation threshold, it indicates that the image similarity between the thermal infrared contour image and the standard image is high at this time, and a matching instruction is generated. The matching module 6 may be a matching program installed on the on-board computer, and is used to match the standard image in the matching database according to the matching instruction to obtain the wild animal motion parameters, the wild animal motion parameters include average running speed, average running acceleration and average weight, and the relative motion parameters include relative distance, relative speed and relative acceleration. The motion detection module 7 may be a motion state detection device, which is internally provided with a distance sensor, a speed sensor and an acceleration sensor, which are respectively used to detect the relative distance, relative speed and relative acceleration. The prediction module 8 may be a collision risk prediction model installed on the on-board computer, and the collision risk prediction model is based on the GBDT model as the basic model and is trained based on the historical data in the training database. The historical data includes historical animal coordinates, historical vehicle coordinates, historical running speed, historical running acceleration, historical weight, historical relative distance, historical relative speed, historical relative acceleration and historical collision risk values ​​at several historical moments, wherein the historical animal coordinates, historical vehicle coordinates, historical running speed, historical running acceleration, historical weight, historical relative distance, historical relative speed and historical relative acceleration are used as input parameters of the GBDT model, and the historical collision risk value at the next moment from the historical moment is used as the output parameter of the GBDT model. A collision risk prediction model is obtained through several iterations and precision verification. The collision risk prediction model can predict the predicted collision risk value at the next moment based on the real-time input animal coordinates, vehicle coordinates, average running speed, average running acceleration, average weight, relative distance, relative speed and relative acceleration.The vehicle control module 9 can be a driving status control program in the driving computer, and can dynamically adjust the vehicle's power output, braking force and steering angle according to the predicted collision risk value, which can effectively avoid collisions between the vehicle and wild animals.

[0058] This embodiment locates the vehicle coordinates in real time during outdoor driving, and collects environmental audio data and thermal infrared profile images when the vehicle coordinates are within the coordinate range of the gathering place. By comparing the audio deviation between the environmental audio data and the standard audio data, and comparing the comprehensive image deviation between the thermal infrared profile image and the standard image, the types of wild animals outside the field of view are detected and analyzed in a progressive manner, thereby improving the detection speed and ensuring the detection accuracy. Then, the relative motion parameters of the wild animal and the vehicle at the current moment are detected, so that the animal coordinates, vehicle coordinates, wild animal motion parameters, and relative motion parameters are input into the collision risk prediction model to predict the predicted collision risk value between the wild animal and the vehicle after the next moment. Finally, the power output power, braking force and steering angle of the vehicle are dynamically adjusted according to the predicted collision risk value, so that the vehicle can automatically avoid the wild animals outside the field of view, thereby improving the safety of avoidance.

[0059] Preferably, Figure 2 As shown, the first analysis module 4 includes:

[0060] The audio preprocessing unit 41 is used to perform noise reduction and normalization processing on the environmental audio data in sequence to obtain preprocessed audio data;

[0061] A first extraction unit 42, connected to the audio preprocessing unit 41, is used to extract a number of environmental audio features from the preprocessed audio data, and to extract a number of standard audio features from the standard audio data;

[0062] The first calculation unit 43 is connected to the first extraction unit 42, and is used to calculate the audio sub-deviation between each environmental audio feature and the corresponding standard audio feature according to a preset first similarity measurement algorithm, and then obtain the audio deviation degree based on the comprehensive calculation of each audio sub-deviation degree.

[0063] Specifically, in this embodiment, the audio preprocessing unit 41 first performs noise reduction processing on the environmental audio data, which helps to remove the background noise that may be introduced during the audio recording process. The means of noise reduction processing include Wiener noise filter. The audio preprocessing unit 41 then normalizes the environmental audio elements and adjusts the amplitude of the sound signal so that the sound signal has a consistent energy level, which is more conducive to signal comparison. The first extraction unit 42 uses fast Fourier transform to process the preprocessed audio data and standard audio data, so that the preprocessed audio data and standard audio data are converted from the time domain to the frequency domain, and then combined with the filter group and discrete cosine transform to extract a number of environmental audio features and a number of standard audio features. The first calculation unit 43 uses the Euclidean distance formula to calculate the audio sub-deviation between the environmental audio features and the corresponding standard audio features.

[0064] Preferably, the environmental audio features include environmental timbre, environmental pitch and environmental rhythm, the standard audio features include animal timbre, animal pitch and animal rhythm, then the audio sub-deviation includes timbre deviation, pitch deviation and rhythm deviation, the first calculation unit 43 inputs the timbre deviation, pitch deviation and rhythm deviation into a preset audio deviation calculation formula to calculate the audio deviation.

[0065] Preferably, the audio deviation calculation formula is configured as:

[0066]

[0067] Among them, A dd Used to indicate audio deviation, T dd Used to indicate the degree of timbre deviation, P dd Used to indicate pitch deviation, R dd Used to indicate the rhythm deviation, ζ1, ζ2, ζ3 are used to indicate the preset first deviation coefficient, second deviation coefficient and third deviation coefficient respectively.

[0068] Specifically, in this embodiment, the first deviation coefficient, the second deviation coefficient, and the third deviation coefficient can be adjusted according to the environment during the audio recording process to ensure the calculation accuracy of the audio deviation degree.

[0069] Preferably, the first analysis module 4 further includes a weather acquisition unit 44, which is used to acquire the environmental weather data of the external environment corresponding to the vehicle coordinates in real time, and obtain the corresponding interference audio according to the environmental weather data analysis, and then obtain the corresponding coefficient adjustment value according to each processing;

[0070] The first calculation unit 43 is connected to the weather acquisition unit 44, and is used to dynamically adjust the first deviation coefficient, the second deviation coefficient and the third deviation coefficient according to each coefficient adjustment value.

[0071] Specifically, in this embodiment, the weather acquisition unit 44 obtains the environmental weather data of the wild animal gathering place in the weather system through the on-board computer network. The environmental weather data is associated with different interference audios, including tidal sounds, thunder, rain sounds, and wind sounds. In the noise reduction process of the audio preprocessing unit 41, it is impossible to ensure that the interference audio is completely removed, so it is necessary to generate a residual noise interference value based on the interference audio and the noise filtering performance of the Wiener noise filter. The first analysis module 4 is provided with an adjustment database, and the adjustment database stores the correlation relationship between the residual noise interference value and the coefficient adjustment value. According to the residual noise interference value, the corresponding coefficient adjustment value is matched in the adjustment database, and finally the coefficient adjustment value is multiplied with the original first deviation coefficient, second deviation coefficient, and third deviation coefficient to obtain the adjusted first deviation coefficient, second deviation coefficient, and third deviation coefficient. Since the influence of the residual noise on the audio similarity calculation is eliminated, the calculation accuracy of the audio deviation degree is higher.

[0072] Preferably, Figure 3 As shown, the thermal infrared detection module 5 includes:

[0073] An image preprocessing unit 51 is used to sequentially perform grayscale, image enhancement, image filtering and binarization processing on the thermal infrared profile image to obtain a preprocessed image;

[0074] The second extraction unit 52 is connected to the image preprocessing unit 51 and is used to extract a number of environmental image features in the preprocessed image and a number of animal image features in the standard image;

[0075] The second calculation unit 53 is connected to the second extraction unit 52, and is used to calculate the image sub-deviation between each environmental image feature and the corresponding animal image feature according to a preset second similarity measurement algorithm, and then calculate the image comprehensive deviation based on the comprehensive calculation of each image sub-deviation.

[0076] Specifically, in this embodiment, the second extraction unit 52 extracts a number of environmental image features in the preprocessed image according to a feature extraction algorithm based on a convolutional neural network, and extracts a number of animal image features in the standard image.

[0077] Preferably, the environmental image features include environmental image edges, environmental image textures and environmental image shapes, and the animal image features include animal image edges, animal image textures and animal image shapes, then the image sub-deviations include edge deviations, texture deviations and shape deviations, and the second calculation unit 53 inputs the edge deviations, texture deviations and shape deviations into a preset image deviation calculation formula to calculate the comprehensive image deviation.

[0078] Preferably, the image deviation calculation formula is configured as:

[0079]

[0080] Among them, L cd Used to indicate the comprehensive deviation of the image, E dd Used to indicate edge deviation, T dd Used to indicate texture deviation, S dd It is used to indicate the shape deviation, and χ1, χ2, and χ3 are respectively used to indicate the preset fourth deviation coefficient, fifth deviation coefficient, and sixth deviation coefficient.

[0081] Preferably, Figure 4 As shown, the vehicle control module 9 includes:

[0082] A control storage unit 91 is used to store a plurality of collision risk ranges, each of which is associated with a corresponding upper output power limit, a lower power limit and a lower steering angle limit;

[0083] A matching unit 92, connected to the control storage unit 91, is used to determine the collision risk range of the predicted collision risk value, and then match the upper limit output power, lower limit power and lower limit steering angle of the vehicle in the current state within the collision risk range;

[0084] The third calculation unit 93 is connected to the matching unit 92 and is used to calculate the power difference between the upper limit output power and the power output power of the vehicle in the current state, calculate the braking difference between the braking force and the lower limit power, and calculate the angle difference between the steering angle and the lower limit steering angle;

[0085] The adjustment unit 94 is connected to the third calculation unit 93 and is used to dynamically adjust the vehicle's power output power, braking force and steering angle according to the power difference, braking difference and angle difference, until the predicted collision risk value is lower than the risk threshold.

[0086] Specifically, in this embodiment, a plurality of collision risk ranges are pre-stored in the control storage unit 91, and each collision risk range is associated with a corresponding upper output power limit, a lower power limit, and a lower steering angle limit. The settings of the upper output power limit, the lower power limit, and the lower steering angle limit can ensure driving safety and prevent collisions between vehicles and wild animals. When the predicted collision risk value obtained by prediction is within a certain collision risk range, the upper output power limit within the collision risk range is subtracted from the power output power of the vehicle in the current state to obtain a power difference value, the braking force of the vehicle in the current state is subtracted from the lower power limit within the collision risk range to obtain a braking difference value, and the steering angle of the vehicle in the current state is subtracted from the lower steering angle limit within the collision risk range to obtain an angle difference value. The power difference value, the braking difference value, and the angle difference value are the values ​​of the power output power, the braking force, and the steering angle of the vehicle in the current state that need to be adjusted urgently. By adjusting the vehicle state, the power difference value, the braking difference value, and the angle difference value are continuously reduced until the predicted collision risk value is lower than the risk threshold value. At this time, the risk of collision between the vehicle and the wild animal is reduced to a risk-free state, which can ensure driving safety.

[0087] The above are only preferred embodiments of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A vehicle anti-collision safety system, characterized in that: include: A storage module (1) is used to store the coordinate ranges of a plurality of wild animal gathering places and a matching database, wherein the coordinate ranges of the gathering places are associated with standard images and standard audio data of a plurality of wild animals; A positioning module (2), connected to the storage module (1), is used to locate the vehicle coordinates in real time during the vehicle's travel, and to generate an audio collection instruction when the vehicle coordinates are within the coordinate range of the gathering place; An audio acquisition module (3) is used to acquire ambient audio data in an external environment in real time according to the audio acquisition instruction; A first analysis module (4) is connected to the audio acquisition module (3) and the storage module (1), and is used to perform similarity analysis on the ambient audio data within a preset time period and the standard audio data to obtain an audio deviation, and to generate an infrared detection instruction when the audio deviation is less than a preset first deviation threshold; a thermal infrared detection module 5), connected to the first analysis module (4) and the storage module (1), for detecting a thermal infrared contour image around the vehicle in real time according to the infrared detection instruction, performing a similarity analysis between the thermal infrared contour image and the standard image to obtain a comprehensive image deviation, and generating a matching instruction when the comprehensive image deviation is less than a preset second deviation threshold; A matching module (6), connected to the thermal infrared detection module 5) and the storage module (1), respectively, for matching the standard image in the matching database according to the matching instruction to obtain corresponding wild animal movement parameters; A motion detection module (7), used to detect the relative motion parameters between the wild animal and the vehicle at the current moment; A prediction module (8) is respectively connected to the thermal infrared detection module 5), the positioning module (2), the matching module (6) and the motion detection module (7), and is used to perform position analysis on the thermal infrared profile image according to the matching instruction to obtain corresponding animal coordinates, and then input the animal coordinates, the vehicle coordinates, the wild animal motion parameters, and the relative motion parameters into a preset collision risk prediction model to predict the predicted collision risk value between the wild animal and the vehicle after the next moment; A vehicle control module (9) is connected to the prediction module (8) and is used to dynamically adjust the power output, braking force and steering angle of the vehicle according to the predicted collision risk value until the predicted collision risk value is lower than a preset risk threshold.

2. The vehicle anti-collision safety system according to claim 1, characterized in that: The wild animal movement parameters include average running speed, average running acceleration and average weight, and the relative movement parameters include relative distance, relative speed and relative acceleration.

3. The vehicle anti-collision safety system according to claim 1, characterized in that: The first analysis module (4) comprises: An audio preprocessing unit (41), used for performing noise reduction and normalization processing on the environmental audio data in sequence to obtain preprocessed audio data; A first extraction unit (42), connected to the audio preprocessing unit (41), used for extracting a plurality of environmental audio features from the preprocessed audio data, and extracting a plurality of standard audio features from the standard audio data; A first calculation unit (43) is connected to the first extraction unit (42) and is used to calculate the audio sub-deviation between each of the environmental audio features and the corresponding standard audio features according to a preset first similarity measurement algorithm, and then obtain the audio deviation degree based on a comprehensive calculation of each of the audio sub-deviation degrees.

4. The vehicle anti-collision safety system according to claim 3, characterized in that: The environmental audio features include environmental timbre, environmental pitch and environmental rhythm, the standard audio features include animal timbre, animal pitch and animal rhythm, and the audio sub-deviation degree includes timbre deviation degree, pitch deviation degree and rhythm deviation degree. The first calculation unit (43) inputs the timbre deviation degree, the pitch deviation degree and the rhythm deviation degree into a preset audio deviation calculation formula to calculate the audio deviation degree.

5. The vehicle anti-collision safety system according to claim 4, characterized in that: The audio deviation calculation formula is configured as: Among them, A dd Used to express the audio deviation degree, T dd It is used to indicate the timbre deviation, P dd Used to indicate the pitch deviation, R dd It is used to represent the rhythm deviation degree, and ζ1, ζ2, and ζ3 are used to represent the preset first deviation coefficient, second deviation coefficient, and third deviation coefficient, respectively.

6. The vehicle anti-collision safety system according to claim 5, characterized in that: The first analysis module (4) further comprises a weather acquisition unit (44) for acquiring in real time environmental weather data of the external environment corresponding to the vehicle coordinates, and obtaining corresponding interference audio according to the environmental weather data analysis, and then obtaining corresponding coefficient adjustment values ​​according to each of the processing; The first calculation unit (43) is connected to the weather acquisition unit (44) and is used to dynamically adjust the first deviation coefficient, the second deviation coefficient and the third deviation coefficient according to each of the coefficient adjustment values.

7. The vehicle anti-collision safety system according to claim 1, characterized in that: The thermal infrared detection module 5) comprises: An image preprocessing unit (51) is used to sequentially perform grayscale, image enhancement, image filtering and binarization processing on the thermal infrared profile image to obtain a preprocessed image; A second extraction unit (52), connected to the image preprocessing unit (51), is used to extract a number of environmental image features in the preprocessed image, and to extract a number of animal image features in the standard image; A second calculation unit (53) is connected to the second extraction unit (52) and is used to calculate the image sub-deviation between each of the environmental image features and the corresponding animal image features according to a preset second similarity measurement algorithm, and then obtain the image comprehensive deviation based on the comprehensive calculation of each of the image sub-deviations.

8. The vehicle anti-collision safety system according to claim 7, characterized in that: The environmental image features include environmental image edges, environmental image textures and environmental image shapes, and the animal image features include animal image edges, animal image textures and animal image shapes. The image sub-deviations include edge deviations, texture deviations and shape deviations. The second calculation unit (53) inputs the edge deviations, texture deviations and shape deviations into a preset image deviation calculation formula to calculate the image comprehensive deviation.

9. The vehicle anti-collision safety system according to claim 8, characterized in that: The image deviation calculation formula is configured as: Among them, L cd It is used to represent the comprehensive deviation of the image, E dd Used to express the edge deviation, T dd It is used to represent the texture deviation, S dd It is used to represent the shape deviation, and χ1, χ2, and χ3 are respectively used to represent the preset fourth deviation coefficient, fifth deviation coefficient, and sixth deviation coefficient.

10. The vehicle anti-collision safety system according to claim 1, characterized in that: The vehicle control module (9) comprises: A control storage unit (91) is used to store a plurality of collision risk ranges, each of which is associated with a corresponding upper output power limit, a lower power limit and a lower steering angle limit; a matching unit (92), connected to the control storage unit (91), and used to determine the collision risk range in which the predicted collision risk value is located, and then to match the upper limit output power, the lower limit power, and the lower limit steering angle in the current state of the vehicle within the collision risk range; a third calculation unit (93), connected to the matching unit (92), for calculating the power difference between the upper limit output power and the power output power of the vehicle in a current state, calculating the braking difference between the braking force and the lower limit power, and calculating the angle difference between the steering angle and the lower limit steering angle; An adjustment unit (94) is connected to the third calculation unit (93) and is used to dynamically adjust the power output power, the braking force and the steering angle of the vehicle according to the power difference, the braking difference and the angle difference, until the predicted collision risk value is lower than the risk threshold.

Images