Vehicle defect alarm method and system based on alarm rule

Abstract

The invention discloses a vehicle defect alarm method and system based on an alarm rule. The method comprises the following steps: acquiring a plurality of sensing data of a target vehicle; determining a plurality of local defects of the target vehicle according to the sensing data based on a defect detection algorithm; determining vehicle defects of the target vehicle according to all the local defects based on a preset defect association rule; determining an alarm strategy corresponding to the target vehicle according to the alarm levels corresponding to different vehicle defects and the vehicle defects; the alarm strategy is used for limiting alarm time, information and sending user side equipment. Therefore, more efficient and more accurate vehicle defect alarm can be realized based on the defect type and level, a more accurate data basis is provided for vehicle defect maintenance, and the safety degree of the vehicle is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a vehicle defect alarm method and system based on alarm rules. Background Art

[0002] With the development of automation technology and the increase in the number of vehicles in the country, more and more automation technologies are being applied in the processing or testing of vehicles to improve the accuracy of vehicle defect detection. Among them, since vehicles include a large number of complex and critical components and their defect types are also diverse, how to improve the accuracy of vehicle detection alarms is one of the important technical issues. Among the existing vehicle defect alarm processing technologies, most still rely on preset simple alarm rules or manual judgment processing to achieve more accurate alarms, and do not pay attention to the defect type and defect level to achieve more accurate alarms. Therefore, the accuracy and efficiency of their vehicle defect alarms are lacking, and product safety cannot meet the requirements. It can be seen that the existing technology has defects that need to be solved urgently. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a vehicle defect alarm method and system based on alarm rules, which can achieve more efficient and accurate vehicle defect alarms based on defect type and level, provide a more accurate data basis for vehicle defect repair, and improve the safety of the vehicle.

[0004] In order to solve the above technical problems, the first aspect of the present invention discloses a vehicle defect alarm method based on alarm rules, the method comprising:

[0005] Acquire multiple sensor data of the target vehicle;

[0006] Determining, based on the sensor data, a plurality of local defects of the target vehicle based on a defect detection algorithm;

[0007] Based on a preset defect association rule, determining the vehicle defect of the target vehicle according to all the local defects;

[0008] According to the alarm levels corresponding to different vehicle defects and the vehicle defects, the alarm strategy corresponding to the target vehicle is determined; the alarm strategy is used to limit the alarm time, information and user terminal device to which it is sent.

[0009] As an optional embodiment, in the first aspect of the present invention, the sensor data is sensor data of different data types in different vehicle parts; the data type is image data, sound data, infrared ranging data, temperature data, humidity data or light reflection modeling data.

[0010] As an optional embodiment, in the first aspect of the present invention, for each piece of sensor data, a defect detection neural network corresponding to the sensor data is determined based on the data type of the sensor data and the corresponding vehicle part; the defect detection neural network is trained using a training data set including a plurality of training sensor data of corresponding data types and vehicle parts and corresponding defect type annotations;

[0011] The sensing data is input into the defect detection neural network to obtain the local defects corresponding to the sensing data; the local defects include defect types and defect locations.

[0012] As an optional embodiment, in the first aspect of the present invention, the defect detection neural network includes a feature extraction network, a classification network, and a post-classification verification output network; the post-classification verification output network is used to determine the corresponding result correction probabilities between multiple predicted local defects in the prediction results output by the classification network and the data types and vehicle parts corresponding to the input sensor data, and multiply the predicted probability corresponding to each local defect by the corresponding result correction probability to obtain a corrected probability, wherein each of the defect detection neural networks is trained through the following steps:

[0013] Using a plurality of training data sets including training sensor data of different data types as a unified data set, and training a unified feature extraction network and a unified classification network based on the unified data set until convergence, thereby obtaining a trained unified feature extraction network and a trained unified classification network;

[0014] For each combination of data type and vehicle part, the trained unified feature extraction network and unified classification network are distilled to obtain the feature extraction network and classification network in the defect detection neural network corresponding to the combination;

[0015] Determine, from the unified data set, a training data set corresponding to the combination and multiple other associated training data sets corresponding to the same vehicle part of the training data set corresponding to the combination;

[0016] Determining, based on the labeling results corresponding to the same vehicle part in the multiple associated training data sets, multiple labeling correction results corresponding to the training data sets to obtain a corrected training data set;

[0017] According to the training data set corresponding to the combination, the feature extraction network and classification network obtained by distillation are trained until convergence, and the trained feature extraction network and classification network corresponding to the combination are obtained;

[0018] According to the corrected training data set, the preset post-classification verification output network corresponding to the combination is trained to obtain the trained post-classification verification output network corresponding to the combination.

[0019] As an optional embodiment, in the first aspect of the present invention, determining the vehicle defects of the target vehicle based on all the local defects based on a preset defect association rule includes:

[0020] For each preset vehicle defect type, obtain multiple detection defect results and corresponding detection times corresponding to the vehicle defect type in the historical database;

[0021] Calculating a weighted sum of similarities between each of the detected defect results and all of the local defects to obtain a similarity parameter corresponding to the type of vehicle defect;

[0022] Filtering out at least one centralized defect set from the plurality of defect detection results; the centralized defect set including a plurality of defect detection results whose time differences between the detection times are less than a time difference threshold;

[0023] Calculating an average of the detection times of all the detected defect results in each of the concentrated defect sets to obtain a collection time of each of the concentrated defect sets;

[0024] Calculating the average of the time difference between the collection time of each of the concentrated defect sets and the current time to obtain a time parameter corresponding to the vehicle defect type;

[0025] Calculating the product of the similarity parameter and the time parameter to obtain a priority parameter for the vehicle defect type;

[0026] At least one vehicle defect type having the priority parameter greater than a parameter threshold is screened out from all the vehicle defect types to obtain the vehicle defect of the target vehicle.

[0027] As an optional embodiment, in the first aspect of the present invention, the vehicle defect is a flaw defect, a driving-impact defect, a short-term defect, a long-term defect, or a hidden defect.

[0028] As an optional embodiment, in the first aspect of the present invention, determining the alarm strategy corresponding to the target vehicle based on the alarm levels corresponding to different vehicle defects and the vehicle defect includes:

[0029] For each of the vehicle defects, determining a defect level parameter corresponding to the vehicle defect according to a preset defect level rule;

[0030] Determining, based on the historical detection records corresponding to the vehicle type of the target vehicle, the historical alarm terminal records corresponding to the vehicle defect;

[0031] Calculate the total number of different types of alarm terminals in the historical alarm terminal records to obtain terminal complexity parameters corresponding to the vehicle defect;

[0032] Calculating the product of the defect level parameter and the terminal complexity parameter to obtain the alarm priority corresponding to the vehicle defect;

[0033] According to the historical alarm terminal records and the alarm priority corresponding to each vehicle defect, an alarm strategy corresponding to the target vehicle is determined based on a dynamic programming algorithm.

[0034] As an optional embodiment, in the first aspect of the present invention, determining the alarm strategy corresponding to the target vehicle based on the historical alarm terminal record and the alarm priority corresponding to each vehicle defect based on a dynamic programming algorithm includes:

[0035] The objective function is to maximize the number of vehicle defects determined to be alarmed in the alarm strategy and minimize the total alarm time corresponding to the alarm strategy; the total alarm time is obtained by predicting the sum of the time taken to send each alarm instruction in the alarm strategy to the corresponding user terminal device at the corresponding alarm time; the time is obtained by inputting the alarm instruction and the corresponding user terminal device into a trained time prediction model for prediction; the time prediction model is trained using a training data set including a plurality of training alarm instructions, corresponding user terminal devices, and sending time annotations;

[0036] Setting restrictions includes:

[0037] The higher the alarm priority in the alarm strategy, the earlier the corresponding alarm time of the vehicle defect;

[0038] The vehicle defect corresponding to the alarm priority in the alarm strategy is lower than the preset first priority threshold and no alarm is issued;

[0039] The average value of the alarm priorities of all the vehicle defects that generate alarms in the alarm strategy is greater than a second priority threshold; the first priority threshold is greater than the second priority threshold;

[0040] The user terminal device corresponding to each vehicle defect in the alarm strategy is the device that appears most frequently in the corresponding historical alarm terminal record;

[0041] Based on a dynamic programming algorithm, the alarm strategies for all the vehicle defects are iteratively calculated according to the objective function and the constraint conditions until they are optimal, thereby obtaining an alarm strategy corresponding to the target vehicle.

[0042] A second aspect of an embodiment of the present invention discloses a vehicle defect alarm system based on alarm rules, the system comprising:

[0043] An acquisition module, used to acquire multiple sensor data of the target vehicle;

[0044] a detection module, configured to determine a plurality of local defects of the target vehicle based on the sensor data based on a defect detection algorithm;

[0045] A verification module, configured to determine the vehicle defect of the target vehicle based on all the local defects based on a preset defect association rule;

[0046] The alarm module is used to determine the alarm strategy corresponding to the target vehicle according to the alarm levels corresponding to different vehicle defects and the vehicle defects; the alarm strategy is used to limit the alarm time, information and user terminal device to which it is sent.

[0047] As an optional embodiment, in the second aspect of the present invention, the sensor data is sensor data of different data types in different vehicle parts; the data type is image data, sound data, infrared ranging data, temperature data, humidity data or light reflection modeling data.

[0048] As an optional embodiment, in the second aspect of the present invention, for each piece of sensor data, a defect detection neural network corresponding to the sensor data is determined based on the data type of the sensor data and the corresponding vehicle part; the defect detection neural network is trained using a training data set including a plurality of training sensor data of corresponding data types and vehicle parts and corresponding defect type annotations;

[0049] The sensing data is input into the defect detection neural network to obtain the local defects corresponding to the sensing data; the local defects include defect types and defect locations.

[0050] As an optional embodiment, in the second aspect of the present invention, the defect detection neural network includes a feature extraction network, a classification network, and a post-classification verification output network; the post-classification verification output network is used to determine the corresponding result correction probabilities between multiple predicted local defects in the prediction results output by the classification network and the data types and vehicle parts corresponding to the input sensor data, and multiply the predicted probability corresponding to each local defect by the corresponding result correction probability to obtain a corrected probability, wherein each of the defect detection neural networks is trained through the following steps:

[0051] Using a plurality of training data sets including training sensor data of different data types as a unified data set, and training a unified feature extraction network and a unified classification network based on the unified data set until convergence, thereby obtaining a trained unified feature extraction network and a trained unified classification network;

[0052] For each combination of data type and vehicle part, the trained unified feature extraction network and unified classification network are distilled to obtain the feature extraction network and classification network in the defect detection neural network corresponding to the combination;

[0053] Determine, from the unified data set, a training data set corresponding to the combination and multiple other associated training data sets corresponding to the same vehicle part of the training data set corresponding to the combination;

[0054] Determining, based on the labeling results corresponding to the same vehicle part in the multiple associated training data sets, multiple labeling correction results corresponding to the training data sets to obtain a corrected training data set;

[0055] According to the training data set corresponding to the combination, the feature extraction network and classification network obtained by distillation are trained until convergence, and the trained feature extraction network and classification network corresponding to the combination are obtained;

[0056] According to the corrected training data set, the preset post-classification verification output network corresponding to the combination is trained to obtain the trained post-classification verification output network corresponding to the combination.

[0057] As an optional embodiment, in the second aspect of the present invention, the verification module determines the specific manner in which the vehicle defect of the target vehicle is determined based on all the local defects based on a preset defect association rule, including:

[0058] For each preset vehicle defect type, obtain multiple detection defect results and corresponding detection times corresponding to the vehicle defect type in the historical database;

[0059] Calculating a weighted sum of similarities between each of the detected defect results and all of the local defects to obtain a similarity parameter corresponding to the type of vehicle defect;

[0060] Filtering out at least one centralized defect set from the plurality of defect detection results; the centralized defect set including a plurality of defect detection results whose time differences between the detection times are less than a time difference threshold;

[0061] Calculating an average of the detection times of all the detected defect results in each of the concentrated defect sets to obtain a collection time of each of the concentrated defect sets;

[0062] Calculating the average of the time difference between the collection time of each of the concentrated defect sets and the current time to obtain a time parameter corresponding to the vehicle defect type;

[0063] Calculating the product of the similarity parameter and the time parameter to obtain a priority parameter for the vehicle defect type;

[0064] At least one vehicle defect type having the priority parameter greater than a parameter threshold is screened out from all the vehicle defect types to obtain the vehicle defect of the target vehicle.

[0065] As an optional embodiment, in the second aspect of the present invention, the vehicle defect is a flaw defect, a driving-impact defect, a short-term defect, a long-term defect, or a hidden defect.

[0066] As an optional embodiment, in the second aspect of the present invention, the alarm module determines the specific manner of the alarm strategy corresponding to the target vehicle based on the alarm levels corresponding to different vehicle defects and the vehicle defects, including:

[0067] For each of the vehicle defects, determining a defect level parameter corresponding to the vehicle defect according to a preset defect level rule;

[0068] Determining, based on the historical detection records corresponding to the vehicle type of the target vehicle, the historical alarm terminal records corresponding to the vehicle defect;

[0069] Calculate the total number of different types of alarm terminals in the historical alarm terminal records to obtain terminal complexity parameters corresponding to the vehicle defect;

[0070] Calculating the product of the defect level parameter and the terminal complexity parameter to obtain the alarm priority corresponding to the vehicle defect;

[0071] According to the historical alarm terminal records and the alarm priority corresponding to each vehicle defect, an alarm strategy corresponding to the target vehicle is determined based on a dynamic programming algorithm.

[0072] As an optional embodiment, in the second aspect of the present invention, the alarm module determines the specific manner of the alarm strategy corresponding to the target vehicle based on the historical alarm terminal record and the alarm priority corresponding to each vehicle defect based on a dynamic programming algorithm, including:

[0073] The objective function is to maximize the number of vehicle defects determined to be alarmed in the alarm strategy and minimize the total alarm time corresponding to the alarm strategy; the total alarm time is obtained by predicting the sum of the time taken to send each alarm instruction in the alarm strategy to the corresponding user terminal device at the corresponding alarm time; the time is obtained by inputting the alarm instruction and the corresponding user terminal device into a trained time prediction model for prediction; the time prediction model is trained using a training data set including a plurality of training alarm instructions, corresponding user terminal devices, and sending time annotations;

[0074] Setting restrictions includes:

[0075] The higher the alarm priority in the alarm strategy, the earlier the corresponding alarm time of the vehicle defect;

[0076] The vehicle defect corresponding to the alarm priority in the alarm strategy is lower than the preset first priority threshold and no alarm is issued;

[0077] The average value of the alarm priorities of all the vehicle defects that generate alarms in the alarm strategy is greater than a second priority threshold; the first priority threshold is greater than the second priority threshold;

[0078] The user terminal device corresponding to each vehicle defect in the alarm strategy is the device that appears most frequently in the corresponding historical alarm terminal record;

[0079] Based on a dynamic programming algorithm, the alarm strategies for all the vehicle defects are iteratively calculated according to the objective function and the constraint conditions until they are optimal, thereby obtaining an alarm strategy corresponding to the target vehicle.

[0080] A third aspect of the present invention discloses another vehicle defect alarm system based on alarm rules, the system comprising:

[0081] a memory storing executable program code;

[0082] a processor coupled to the memory;

[0083] The processor calls the executable program code stored in the memory to execute part or all of the steps in the vehicle defect alarm method based on alarm rules disclosed in the first aspect of the present invention.

[0084] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute some or all of the steps in the vehicle defect alarm method based on alarm rules disclosed in the first aspect of the present invention.

[0085] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0086] The present invention can determine multiple local defects of a vehicle based on sensor data based on a defect detection algorithm, and then determine more accurate vehicle defects based on defect association rules, so as to finally determine the corresponding alarm strategy for the vehicle based on the alarm levels corresponding to different vehicle defects, thereby being able to achieve more efficient and accurate vehicle defect alarms based on defect types and levels, providing a more accurate data basis for vehicle defect repairs, and improving the safety of vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0088] Figure 1 This is a flow chart of a vehicle defect alarm method based on alarm rules disclosed in an embodiment of the present invention.

[0089] Figure 2 It is a structural diagram of a vehicle defect alarm system based on alarm rules disclosed in an embodiment of the present invention.

[0090] Figure 3 It is a structural diagram of another vehicle defect alarm system based on alarm rules disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0091] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0092] The terms "first," "second," and so on, in the description and claims of the present invention and the accompanying drawings are used to distinguish between different objects, not to describe a specific order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product, or device comprising a series of steps or elements is not limited to the listed steps or elements but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or device.

[0093] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0094] The present invention discloses a vehicle defect alarm method and system based on alarm rules. These methods can determine multiple local defects of a vehicle based on sensor data using a defect detection algorithm, then more accurately identify vehicle defects based on defect association rules. Ultimately, they determine the corresponding alarm strategy for each vehicle based on the alarm level corresponding to each defect. This method enables more efficient and accurate vehicle defect alarms based on defect type and level, providing a more accurate data foundation for vehicle defect repair and improving vehicle safety. These are described in detail below.

[0095] Example 1

[0096] See also Figure 1 , Figure 1 This is a flow chart of a vehicle defect alarm method based on alarm rules disclosed in an embodiment of the present invention. Figure 1 The vehicle defect alarm method based on alarm rules described above can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 1 As shown, the vehicle defect alarm method based on alarm rules may include the following operations:

[0097] 101. Acquire multiple sensor data of the target vehicle.

[0098] 102. Based on the defect detection algorithm, multiple local defects of the target vehicle are determined according to the sensor data.

[0099] 103. Based on the preset defect association rules, determine the vehicle defects of the target vehicle according to all local defects.

[0100] 104. Determine the alarm strategy corresponding to the target vehicle based on the alarm levels corresponding to different vehicle defects and the vehicle defects.

[0101] Optionally, the alarm policy is used to limit the time, information, and user-end device to which the alarm is sent.

[0102] It can be seen that the above-mentioned embodiments of the invention can determine multiple local defects of the vehicle based on the sensor data based on the defect detection algorithm, and then determine more accurate vehicle defects based on the defect association rules, so as to finally determine the corresponding alarm strategy of the vehicle based on the alarm levels corresponding to different vehicle defects, thereby being able to achieve more efficient and accurate vehicle defect alarms based on the defect type and level, providing a more accurate data basis for vehicle defect repair, and improving the safety of the vehicle.

[0103] As an optional embodiment, in the above steps, the sensor data is sensor data of different data types in different vehicle parts; the data types are image data, sound data, infrared ranging data, temperature data, humidity data or light reflection modeling data.

[0104] It can be seen that through the above optional embodiments, the type of sensor data is limited, which can fully characterize the characteristics of the vehicle, facilitate subsequent defect detection and alarm of the vehicle, and assist in achieving more efficient and accurate vehicle defect alarms based on defect types and levels, providing a more accurate data basis for vehicle defect repairs and improving the safety of the vehicle.

[0105] As an optional embodiment, in the above steps, for each piece of sensor data, a defect detection neural network corresponding to the sensor data is determined based on the data type of the sensor data and the corresponding vehicle part; the defect detection neural network is trained using a training data set including a plurality of training sensor data of corresponding data types and vehicle parts and corresponding defect type annotations;

[0106] The sensor data is input into a defect detection neural network to obtain local defects corresponding to the sensor data; local defects include defect types and defect locations.

[0107] It can be seen that through the above optional embodiments, the corresponding defect detection neural network can be determined based on the data type of the sensing data and the corresponding vehicle part, so as to achieve accurate detection of local defects, facilitate subsequent defect correction and alarm of the vehicle, and assist in achieving more efficient and accurate vehicle defect alarms based on defect type and level, provide a more accurate data basis for vehicle defect repair, and improve the safety of the vehicle.

[0108] As an optional embodiment, in the above steps, the defect detection neural network includes a feature extraction network, a classification network, and a post-classification verification output network; the post-classification verification output network is used to determine the corresponding result correction probabilities between multiple predicted local defects in the prediction results output by the classification network and the data types corresponding to the input sensor data and the vehicle parts, and multiply the predicted probability corresponding to each local defect by the corresponding result correction probability to obtain a corrected probability, wherein each defect detection neural network is trained through the following steps:

[0109] A plurality of training data sets including training sensor data of different data types are used as a unified data set, and a unified feature extraction network and a unified classification network are trained based on the unified data set until convergence, thereby obtaining a trained unified feature extraction network and a trained unified classification network;

[0110] For each combination of data type and vehicle part, the trained unified feature extraction network and unified classification network are distilled to obtain the feature extraction network and classification network in the defect detection neural network corresponding to the combination;

[0111] Determine, from the unified data set, a training data set corresponding to the combination and multiple other associated training data sets corresponding to the same vehicle part of the training data set corresponding to the combination;

[0112] Determining multiple annotation correction results corresponding to the training datasets based on the annotation results corresponding to the same vehicle part in the multiple associated training datasets to obtain a corrected training dataset;

[0113] According to the training data set corresponding to the combination, the feature extraction network and classification network obtained by distillation are trained until convergence, and the trained feature extraction network and classification network corresponding to the combination are obtained;

[0114] According to the corrected training data set, the preset post-classification verification output network corresponding to the combination is trained to obtain the trained post-classification verification output network corresponding to the combination.

[0115] It can be seen that through the above optional embodiments, the feature network and the classification network can be uniformly trained through a unified training data set of multiple different parts and data types, and then distilled and trained separately based on different combinations to obtain a model with better prediction effect, and then the classification and verification output network is trained based on the annotation correction results between multiple training data sets, so that the final output prediction result is more accurate and reliable, which is convenient for the subsequent defect correction and alarm of the vehicle, and assists in realizing more efficient and accurate vehicle defect alarm based on defect type and level, providing a more accurate data basis for vehicle defect repair, and improving the safety of the vehicle.

[0116] As an optional embodiment, in the above steps, based on a preset defect association rule, determining the vehicle defects of the target vehicle according to all local defects includes:

[0117] For each preset vehicle defect type, obtain multiple detection defect results and corresponding detection times corresponding to the vehicle defect type in the historical database;

[0118] Calculate the weighted sum of the similarities between each defect detection result and all local defects to obtain the similarity parameter corresponding to the vehicle defect type;

[0119] Filtering out at least one centralized defect set from the plurality of defect detection results; optionally, the centralized defect set includes a plurality of defect detection results whose time difference between detection times is less than a time difference threshold;

[0120] Calculate the average detection time of all defect detection results in each defect set to obtain the collection time of each defect set;

[0121] Calculate the average of the time difference between the collection time of each defect set and the current time to obtain the time parameter corresponding to the vehicle defect type;

[0122] Calculate the product of the similarity parameter and the time parameter to obtain the priority parameter of the vehicle defect type;

[0123] At least one vehicle defect type having a priority parameter greater than a parameter threshold is screened out from all vehicle defect types to obtain the vehicle defect of the target vehicle.

[0124] It can be seen that through the above optional embodiments, the vehicle defect to which the vehicle currently belongs can be accurately determined based on the similarity between the historical data of the preset vehicle defect type and the current local defect and the concentration and proximity of the detection time, which facilitates the subsequent accurate alarm of the vehicle defect, and assists in achieving more efficient and accurate vehicle defect alarms based on the defect type and level, providing a more accurate data basis for vehicle defect repair and improving the safety of the vehicle.

[0125] As an optional embodiment, in the above steps, the vehicle defect is a flaw defect, a driving-impact defect, a short-term defect, a long-term defect, or a hidden defect.

[0126] It can be seen that through the above optional embodiments, the types of vehicle defects are defined, which facilitates the subsequent accurate alarm of vehicle defects, and assists in achieving more efficient and accurate vehicle defect alarms based on defect types and levels, providing a more accurate data basis for vehicle defect repairs and improving the safety of vehicles.

[0127] As an optional embodiment, in the above steps, determining the alarm strategy corresponding to the target vehicle according to the alarm levels corresponding to different vehicle defects and the vehicle defects includes:

[0128] For each vehicle defect, determine the defect level parameter corresponding to the vehicle defect according to the preset defect level rules;

[0129] Determine the historical alarm terminal record corresponding to the vehicle defect based on the historical inspection record corresponding to the vehicle type of the target vehicle;

[0130] Calculate the total number of different types of alarm terminals in the historical alarm terminal records to obtain the terminal complexity parameters corresponding to the vehicle defect;

[0131] Calculate the product of the defect level parameter and the terminal complexity parameter to obtain the alarm priority corresponding to the vehicle defect;

[0132] According to the historical alarm terminal records and alarm priority corresponding to each vehicle defect, the alarm strategy corresponding to the target vehicle is determined based on the dynamic programming algorithm.

[0133] It can be seen that through the above-mentioned optional embodiments, the corresponding historical alarm terminal records and alarm priorities can be determined by analyzing the level of each vehicle defect and the historical alarm records, so as to determine a more reasonable and accurate alarm strategy based on the dynamic programming algorithm, and realize more efficient and accurate vehicle defect alarms based on the defect type and level, providing a more accurate data basis for vehicle defect repair and improving the safety of the vehicle.

[0134] As an optional embodiment, in the above steps, determining the alarm strategy corresponding to the target vehicle based on the dynamic programming algorithm according to the historical alarm terminal records and alarm priority corresponding to each vehicle defect includes:

[0135] The objective function is set to maximize the number of vehicle defects determined to be alarmed in the alarm strategy and minimize the total alarm time corresponding to the alarm strategy; optionally, the total alarm time is obtained by predicting the sum of the time taken to send each alarm instruction in the alarm strategy to the corresponding user terminal device at the corresponding alarm time; the time is obtained by inputting the alarm instruction and the corresponding user terminal device into a trained time prediction model for prediction; the time prediction model is trained by a training data set including multiple training alarm instructions, corresponding user terminal devices, and sending time annotations;

[0136] Setting restrictions includes:

[0137] The higher the alarm priority of the vehicle defect in the alarm strategy, the earlier the corresponding alarm time;

[0138] Vehicle defects whose corresponding alarm priority in the alarm strategy is lower than the preset first priority threshold will not be alarmed;

[0139] The average value of the alarm priorities of all vehicle defects that are alarmed in the alarm strategy is greater than the second priority threshold; optionally, the first priority threshold is greater than the second priority threshold;

[0140] The user terminal device corresponding to each vehicle defect in the alarm strategy is the device that appears most frequently in the corresponding historical alarm terminal records;

[0141] Based on the dynamic programming algorithm, the alarm strategies for all vehicle defects are iteratively calculated according to the objective function and constraints until they are optimal, and the alarm strategy corresponding to the target vehicle is obtained.

[0142] It can be seen that through the above-mentioned optional embodiments, calculations can be performed based on dynamic programming algorithms and preset reasonable objective functions and constraints to determine a more reasonable and accurate alarm strategy, thereby achieving more efficient and accurate vehicle defect alarms based on defect types and levels, providing a more accurate data basis for vehicle defect repairs, and improving the safety of vehicles.

[0143] Example 2

[0144] See also Figure 2 , Figure 2 This is a schematic diagram of the structure of a vehicle defect alarm system based on alarm rules disclosed in an embodiment of the present invention. Figure 2 The vehicle defect warning system based on warning rules described above can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 2 As shown, the vehicle defect alarm system based on alarm rules may include:

[0145] The acquisition module 201 is used to acquire multiple sensor data of the target vehicle.

[0146] The detection module 202 is configured to determine a plurality of local defects of the target vehicle based on the sensor data based on a defect detection algorithm.

[0147] The verification module 203 is configured to determine the vehicle defects of the target vehicle based on all local defects based on a preset defect association rule.

[0148] The alarm module 204 is used to determine an alarm strategy corresponding to the target vehicle according to the alarm levels corresponding to different vehicle defects and the vehicle defects.

[0149] Optionally, the alarm policy is used to limit the time, information, and user-end device to which the alarm is sent.

[0150] It can be seen that the above-mentioned embodiments of the invention can determine multiple local defects of the vehicle based on the sensor data based on the defect detection algorithm, and then determine more accurate vehicle defects based on the defect association rules, so as to finally determine the corresponding alarm strategy of the vehicle based on the alarm levels corresponding to different vehicle defects, thereby being able to achieve more efficient and accurate vehicle defect alarms based on the defect type and level, providing a more accurate data basis for vehicle defect repair, and improving the safety of the vehicle.

[0151] As an optional embodiment, the sensor data is sensor data of different data types at different vehicle locations; the data types are image data, sound data, infrared ranging data, temperature data, humidity data, or light reflection modeling data.

[0152] It can be seen that through the above optional embodiments, the type of sensor data is limited, which can fully characterize the characteristics of the vehicle, facilitate subsequent defect detection and alarm of the vehicle, and assist in achieving more efficient and accurate vehicle defect alarms based on defect types and levels, providing a more accurate data basis for vehicle defect repairs and improving the safety of the vehicle.

[0153] As an optional embodiment, for each piece of sensor data, a defect detection neural network corresponding to the sensor data is determined based on the data type of the sensor data and the corresponding vehicle part; the defect detection neural network is trained using a training data set including a plurality of training sensor data of corresponding data types and vehicle parts and corresponding defect type annotations;

[0154] The sensor data is input into a defect detection neural network to obtain local defects corresponding to the sensor data; local defects include defect types and defect locations.

[0155] It can be seen that through the above optional embodiments, the corresponding defect detection neural network can be determined based on the data type of the sensing data and the corresponding vehicle part, so as to achieve accurate detection of local defects, facilitate subsequent defect correction and alarm of the vehicle, and assist in achieving more efficient and accurate vehicle defect alarms based on defect type and level, provide a more accurate data basis for vehicle defect repair, and improve the safety of the vehicle.

[0156] As an optional embodiment, the defect detection neural network includes a feature extraction network, a classification network, and a post-classification verification output network; the post-classification verification output network is used to determine the corresponding result correction probabilities between multiple predicted local defects in the prediction results output by the classification network and the data types and vehicle parts corresponding to the input sensor data, and multiply the predicted probability corresponding to each local defect by the corresponding result correction probability to obtain a corrected probability, wherein each defect detection neural network is trained through the following steps:

[0157] A plurality of training data sets including training sensor data of different data types are used as a unified data set, and a unified feature extraction network and a unified classification network are trained based on the unified data set until convergence, thereby obtaining a trained unified feature extraction network and a trained unified classification network;

[0158] For each combination of data type and vehicle part, the trained unified feature extraction network and unified classification network are distilled to obtain the feature extraction network and classification network in the defect detection neural network corresponding to the combination;

[0159] Determine, from the unified data set, a training data set corresponding to the combination and multiple other associated training data sets corresponding to the same vehicle part of the training data set corresponding to the combination;

[0160] Determining multiple annotation correction results corresponding to the training datasets based on the annotation results corresponding to the same vehicle part in the multiple associated training datasets to obtain a corrected training dataset;

[0161] According to the training data set corresponding to the combination, the feature extraction network and classification network obtained by distillation are trained until convergence, and the trained feature extraction network and classification network corresponding to the combination are obtained;

[0162] According to the corrected training data set, the preset post-classification verification output network corresponding to the combination is trained to obtain the trained post-classification verification output network corresponding to the combination.

[0163] It can be seen that through the above optional embodiments, the feature network and the classification network can be uniformly trained through a unified training data set of multiple different parts and data types, and then distilled and trained separately based on different combinations to obtain a model with better prediction effect, and then the classification and verification output network is trained based on the annotation correction results between multiple training data sets, so that the final output prediction result is more accurate and reliable, which is convenient for the subsequent defect correction and alarm of the vehicle, and assists in realizing more efficient and accurate vehicle defect alarm based on defect type and level, providing a more accurate data basis for vehicle defect repair, and improving the safety of the vehicle.

[0164] As an optional embodiment, the verification module determines the specific manner in which the vehicle defect of the target vehicle is detected based on all local defects based on a preset defect association rule, including:

[0165] For each preset vehicle defect type, obtain multiple detection defect results and corresponding detection times corresponding to the vehicle defect type in the historical database;

[0166] Calculate the weighted sum of the similarities between each defect detection result and all local defects to obtain the similarity parameter corresponding to the vehicle defect type;

[0167] Filtering out at least one centralized defect set from the plurality of defect detection results; optionally, the centralized defect set includes a plurality of defect detection results whose time difference between detection times is less than a time difference threshold;

[0168] Calculate the average detection time of all defect detection results in each defect set to obtain the collection time of each defect set;

[0169] Calculate the average of the time difference between the collection time of each defect set and the current time to obtain the time parameter corresponding to the vehicle defect type;

[0170] Calculate the product of the similarity parameter and the time parameter to obtain the priority parameter of the vehicle defect type;

[0171] At least one vehicle defect type having a priority parameter greater than a parameter threshold is screened out from all vehicle defect types to obtain the vehicle defect of the target vehicle.

[0172] It can be seen that through the above optional embodiments, the vehicle defect to which the vehicle currently belongs can be accurately determined based on the similarity between the historical data of the preset vehicle defect type and the current local defect and the concentration and proximity of the detection time, which facilitates the subsequent accurate alarm of the vehicle defect, and assists in achieving more efficient and accurate vehicle defect alarms based on the defect type and level, providing a more accurate data basis for vehicle defect repair and improving the safety of the vehicle.

[0173] As an optional embodiment, the vehicle defect is a flaw defect, a driving-impact defect, a short-term defect, a long-term defect, or a hidden danger defect.

[0174] It can be seen that through the above optional embodiments, the types of vehicle defects are defined, which facilitates the subsequent accurate alarm of vehicle defects, and assists in achieving more efficient and accurate vehicle defect alarms based on defect types and levels, providing a more accurate data basis for vehicle defect repairs and improving the safety of vehicles.

[0175] As an optional embodiment, the alarm module determines the specific manner of the alarm strategy corresponding to the target vehicle according to the alarm levels corresponding to different vehicle defects and the vehicle defects, including:

[0176] For each vehicle defect, determine the defect level parameter corresponding to the vehicle defect according to the preset defect level rules;

[0177] Determine the historical alarm terminal record corresponding to the vehicle defect based on the historical inspection record corresponding to the vehicle type of the target vehicle;

[0178] Calculate the total number of different types of alarm terminals in the historical alarm terminal records to obtain the terminal complexity parameters corresponding to the vehicle defect;

[0179] Calculate the product of the defect level parameter and the terminal complexity parameter to obtain the alarm priority corresponding to the vehicle defect;

[0180] According to the historical alarm terminal records and alarm priority corresponding to each vehicle defect, the alarm strategy corresponding to the target vehicle is determined based on the dynamic programming algorithm.

[0181] It can be seen that through the above-mentioned optional embodiments, the corresponding historical alarm terminal records and alarm priorities can be determined by analyzing the level of each vehicle defect and the historical alarm records, so as to determine a more reasonable and accurate alarm strategy based on the dynamic programming algorithm, and realize more efficient and accurate vehicle defect alarms based on the defect type and level, providing a more accurate data basis for vehicle defect repair and improving the safety of the vehicle.

[0182] As an optional embodiment, the alarm module determines the specific method of the alarm strategy corresponding to the target vehicle based on the historical alarm terminal records and alarm priority corresponding to each vehicle defect based on a dynamic programming algorithm, including:

[0183] The objective function is set to maximize the number of vehicle defects determined to be alarmed in the alarm strategy and minimize the total alarm time corresponding to the alarm strategy; optionally, the total alarm time is obtained by predicting the sum of the time taken to send each alarm instruction in the alarm strategy to the corresponding user terminal device at the corresponding alarm time; the time is obtained by inputting the alarm instruction and the corresponding user terminal device into a trained time prediction model for prediction; the time prediction model is trained by a training data set including multiple training alarm instructions, corresponding user terminal devices, and sending time annotations;

[0184] Setting restrictions includes:

[0185] The higher the alarm priority of the vehicle defect in the alarm strategy, the earlier the corresponding alarm time;

[0186] Vehicle defects whose corresponding alarm priority in the alarm strategy is lower than the preset first priority threshold will not be alarmed;

[0187] The average value of the alarm priorities of all vehicle defects that are alarmed in the alarm strategy is greater than the second priority threshold; optionally, the first priority threshold is greater than the second priority threshold;

[0188] The user terminal device corresponding to each vehicle defect in the alarm strategy is the device that appears most frequently in the corresponding historical alarm terminal records;

[0189] Based on the dynamic programming algorithm, the alarm strategies for all vehicle defects are iteratively calculated according to the objective function and constraints until they are optimal, and the alarm strategy corresponding to the target vehicle is obtained.

[0190] It can be seen that through the above-mentioned optional embodiments, calculations can be performed based on dynamic programming algorithms and preset reasonable objective functions and constraints to determine a more reasonable and accurate alarm strategy, thereby achieving more efficient and accurate vehicle defect alarms based on defect types and levels, providing a more accurate data basis for vehicle defect repairs, and improving the safety of vehicles.

[0191] Example 3

[0192] See also Figure 3 , Figure 3 This is another vehicle defect alarm system based on alarm rules disclosed in an embodiment of the present invention. Figure 3The vehicle defect alarm system based on alarm rules is applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 3 As shown, the vehicle defect alarm system based on alarm rules may include:

[0193] A memory 301 storing executable program code;

[0194] a processor 302 coupled to the memory 301;

[0195] The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the vehicle defect alarm method based on alarm rules described in the first embodiment.

[0196] Example 4

[0197] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the vehicle defect alarm method based on alarm rules described in the first embodiment.

[0198] Example 5

[0199] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps of the vehicle defect alarm method based on alarm rules described in Example 1.

[0200] The foregoing description of specific embodiments of the present disclosure is intended to illustrate a method for performing a multi-tasking process. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0201] The systems, devices, modules, or units described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0202] For the convenience of description, the above devices are described as being divided into various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0203] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the embodiments of this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0204] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0205] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0206] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0207] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0208] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0209] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0210] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0211] This specification may be described in the general context of computer-executable instructions, such as program modules, executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media, including storage devices.

[0212] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.

[0213] Finally, it should be noted that the vehicle defect alarm method and system based on alarm rules disclosed in the embodiments of the present invention are only preferred embodiments of the present invention, and are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A vehicle defect alarm method based on alarm rules, characterized in that: The method comprises: Acquire multiple sensor data of the target vehicle; Determining, based on the sensor data, a plurality of local defects of the target vehicle based on a defect detection algorithm; Based on a preset defect association rule, determining the vehicle defect of the target vehicle according to all the local defects; According to the alarm levels corresponding to different vehicle defects and the vehicle defects, the alarm strategy corresponding to the target vehicle is determined; the alarm strategy is used to limit the alarm time, information and user terminal device to which it is sent.

2. The vehicle defect alarm method based on alarm rules according to claim 1, characterized in that: The sensor data is sensor data of different data types at different vehicle locations; the data types are image data, sound data, infrared ranging data, temperature data, humidity data or light reflection modeling data.

3. The vehicle defect alarm method based on alarm rules according to claim 2, characterized in that: For each of the sensing data, determining a defect detection neural network corresponding to the sensing data according to the data type of the sensing data and the corresponding vehicle part; The defect detection neural network is trained by a training data set including a plurality of training sensor data corresponding to data types and vehicle parts and corresponding defect type annotations; The sensing data is input into the defect detection neural network to obtain the local defects corresponding to the sensing data; the local defects include defect types and defect locations.

4. The vehicle defect alarm method based on alarm rules according to claim 3 is characterized in that: The defect detection neural network includes a feature extraction network, a classification network, and a post-classification verification output network; the post-classification verification output network is used to determine the corresponding result correction probabilities between a plurality of predicted local defects in the prediction results output by the classification network and the data types and vehicle parts corresponding to the input sensor data, and multiply the predicted probability corresponding to each local defect by the corresponding result correction probability to obtain a corrected probability, wherein each of the defect detection neural networks is trained through the following steps: Using a plurality of training data sets including training sensor data of different data types as a unified data set, and training a unified feature extraction network and a unified classification network based on the unified data set until convergence, thereby obtaining a trained unified feature extraction network and a trained unified classification network; For each combination of data type and vehicle part, the trained unified feature extraction network and unified classification network are distilled to obtain the feature extraction network and classification network in the defect detection neural network corresponding to the combination; Determine, from the unified data set, a training data set corresponding to the combination and multiple other associated training data sets corresponding to the same vehicle part of the training data set corresponding to the combination; Determining, based on the labeling results corresponding to the same vehicle part in the multiple associated training data sets, multiple labeling correction results corresponding to the training data sets to obtain a corrected training data set; According to the training data set corresponding to the combination, the feature extraction network and classification network obtained by distillation are trained until convergence, and the trained feature extraction network and classification network corresponding to the combination are obtained; According to the corrected training data set, the preset post-classification verification output network corresponding to the combination is trained to obtain the trained post-classification verification output network corresponding to the combination.

5. The vehicle defect alarm method based on alarm rules according to claim 1, characterized in that: The determining of the vehicle defects of the target vehicle based on all the local defects based on a preset defect association rule includes: For each preset vehicle defect type, obtain multiple detection defect results and corresponding detection times corresponding to the vehicle defect type in the historical database; Calculating a weighted sum of similarities between each of the detected defect results and all of the local defects to obtain a similarity parameter corresponding to the type of vehicle defect; Filtering out at least one centralized defect set from the plurality of defect detection results; the centralized defect set including a plurality of defect detection results whose time differences between the detection times are less than a time difference threshold; Calculating an average of the detection times of all the detected defect results in each of the concentrated defect sets to obtain a collection time of each of the concentrated defect sets; Calculating the average of the time difference between the collection time of each of the concentrated defect sets and the current time to obtain a time parameter corresponding to the vehicle defect type; Calculating the product of the similarity parameter and the time parameter to obtain a priority parameter for the vehicle defect type; At least one vehicle defect type having the priority parameter greater than a parameter threshold is screened out from all the vehicle defect types to obtain the vehicle defect of the target vehicle.

6. The vehicle defect alarm method based on alarm rules according to claim 1, characterized in that: The vehicle defects are flaw defects, driving-impact defects, short-term defects, long-term defects or hidden danger defects.

7. The vehicle defect alarm method based on alarm rules according to claim 1, characterized in that: The step of determining an alarm strategy corresponding to the target vehicle based on the alarm levels corresponding to different vehicle defects and the vehicle defects includes: For each of the vehicle defects, determining a defect level parameter corresponding to the vehicle defect according to a preset defect level rule; Determining, based on the historical detection records corresponding to the vehicle type of the target vehicle, the historical alarm terminal records corresponding to the vehicle defect; Calculate the total number of different types of alarm terminals in the historical alarm terminal records to obtain terminal complexity parameters corresponding to the vehicle defect; Calculating the product of the defect level parameter and the terminal complexity parameter to obtain the alarm priority corresponding to the vehicle defect; According to the historical alarm terminal records and the alarm priority corresponding to each vehicle defect, an alarm strategy corresponding to the target vehicle is determined based on a dynamic programming algorithm.

8. The vehicle defect alarm method based on alarm rules according to claim 7, characterized in that: The step of determining an alarm strategy corresponding to the target vehicle based on the historical alarm terminal records and the alarm priority corresponding to each vehicle defect and a dynamic programming algorithm includes: The objective function is to maximize the number of vehicle defects determined to be alarmed in the alarm strategy and minimize the total alarm time corresponding to the alarm strategy; the total alarm time is obtained by predicting the sum of the time taken to send each alarm instruction in the alarm strategy to the corresponding user terminal device at the corresponding alarm time; the time is obtained by inputting the alarm instruction and the corresponding user terminal device into a trained time prediction model for prediction; the time prediction model is trained using a training data set including a plurality of training alarm instructions, corresponding user terminal devices, and sending time annotations; Setting restrictions includes: The higher the alarm priority in the alarm strategy, the earlier the corresponding alarm time of the vehicle defect; The vehicle defect corresponding to the alarm priority in the alarm strategy is lower than the preset first priority threshold and no alarm is issued; The average value of the alarm priorities of all the vehicle defects that generate alarms in the alarm strategy is greater than a second priority threshold; the first priority threshold is greater than the second priority threshold; The user terminal device corresponding to each vehicle defect in the alarm strategy is the device that appears most frequently in the corresponding historical alarm terminal record; Based on a dynamic programming algorithm, the alarm strategies for all the vehicle defects are iteratively calculated according to the objective function and the constraint conditions until they are optimal, thereby obtaining an alarm strategy corresponding to the target vehicle.

9. A vehicle defect alarm system based on alarm rules, characterized in that: The system comprises: An acquisition module, used to acquire multiple sensor data of the target vehicle; a detection module, configured to determine a plurality of local defects of the target vehicle based on the sensor data based on a defect detection algorithm; A verification module, configured to determine the vehicle defect of the target vehicle based on all the local defects based on a preset defect association rule; The alarm module is used to determine the alarm strategy corresponding to the target vehicle according to the alarm levels corresponding to different vehicle defects and the vehicle defects; the alarm strategy is used to limit the alarm time, information and user terminal device to which it is sent.

10. A vehicle defect alarm system based on alarm rules, characterized in that: The system comprises: a memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the vehicle defect alarm method based on alarm rules as described in any one of claims 1-8.

Images