CN113119890A - Vehicle abnormity prediction method and device and electronic equipment - Google Patents
Vehicle abnormity prediction method and device and electronic equipment Download PDFInfo
- Publication number
- CN113119890A CN113119890A CN202110463152.5A CN202110463152A CN113119890A CN 113119890 A CN113119890 A CN 113119890A CN 202110463152 A CN202110463152 A CN 202110463152A CN 113119890 A CN113119890 A CN 113119890A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- abnormal
- topological graph
- full
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R16/00—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for
- B60R16/02—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements
- B60R16/023—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements for transmission of signals between vehicle parts or subsystems
- B60R16/0231—Circuits relating to the driving or the functioning of the vehicle
- B60R16/0232—Circuits relating to the driving or the functioning of the vehicle for measuring vehicle parameters and indicating critical, abnormal or dangerous conditions
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R16/00—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for
- B60R16/02—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements
- B60R16/023—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements for transmission of signals between vehicle parts or subsystems
- B60R16/0231—Circuits relating to the driving or the functioning of the vehicle
- B60R16/0232—Circuits relating to the driving or the functioning of the vehicle for measuring vehicle parameters and indicating critical, abnormal or dangerous conditions
- B60R16/0234—Circuits relating to the driving or the functioning of the vehicle for measuring vehicle parameters and indicating critical, abnormal or dangerous conditions related to maintenance or repairing of vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Mechanical Engineering (AREA)
- Automation & Control Theory (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a method and a device for predicting vehicle abnormity and electronic equipment, wherein the method comprises the following steps: acquiring working condition data of a vehicle to be predicted; carrying out depth map matching processing on the working condition data by adopting a depth map matching frame to obtain the distance between the vehicle to be predicted and each vehicle under the known abnormal condition in the depth map matching frame; and determining the abnormal condition of the vehicle to be predicted according to the distance. According to the vehicle abnormity prediction method, the depth map matching frame is adopted to predict the vehicle abnormity of the vehicle to be predicted, the accuracy is good, and the technical problem that the existing vehicle abnormity prediction method is poor in accuracy is solved.
Description
Technical Field
The invention relates to the technical field of big data, in particular to a method and a device for predicting vehicle abnormity and electronic equipment.
Background
With the popularization of vehicles, the problem of vehicle abnormality has attracted attention. The prediction of the vehicle abnormity is a crucial link in the vehicle driving safety, and a driver can carry out timely maintenance according to the result of the vehicle abnormity prediction, so that the vehicle abnormity prediction is prevented from happening in the bud.
At present, when vehicle abnormity is predicted, working condition data and abnormity early warning data of a vehicle with certain abnormity are generally analyzed, so that the conclusion that the abnormity occurs when certain parameters of the vehicle reach a specific threshold value is obtained. And when other unknown vehicles are subjected to vehicle abnormity prediction subsequently, directly matching the working condition data of the unknown vehicles with the obtained conclusion, and predicting to obtain whether the unknown vehicles are abnormal or not.
The method for predicting the vehicle abnormality is implemented by setting a threshold value, which is representative only for the abnormality of an individual vehicle, and is less accurate when the abnormality of all vehicles is predicted using the set threshold value.
In conclusion, the conventional vehicle abnormality prediction method has the technical problem of poor accuracy.
Disclosure of Invention
In view of the above, the present invention provides a method, an apparatus and an electronic device for predicting vehicle anomalies, so as to alleviate the technical problem of poor accuracy of the existing vehicle anomaly prediction method.
In a first aspect, an embodiment of the present invention provides a method for predicting a vehicle abnormality, including:
acquiring working condition data of a vehicle to be predicted;
carrying out depth map matching processing on the working condition data by adopting a depth map matching frame to obtain the distance between the vehicle to be predicted and each vehicle under the known abnormal condition in the depth map matching frame;
and determining the abnormal condition of the vehicle to be predicted according to the distance.
Furthermore, the vehicles with the known abnormal conditions cover various abnormal conditions.
Further, determining the abnormal condition of the vehicle to be predicted according to the distance includes:
determining a target distance in the distances, wherein the target distance is the distance with the minimum distance of the first N distances in the distances, and N is a positive integer;
and if the target distance is smaller than a preset threshold value, taking the abnormal condition of the vehicle with the known abnormal condition of the target corresponding to the target distance as the abnormal condition of the vehicle to be predicted.
Further, the method further comprises:
constructing an abnormal vehicle topological graph and a full-scale vehicle topological graph by adopting a mapping module;
determining a point-edge relation matrix of the abnormal vehicle topological graph according to the abnormal vehicle topological graph;
performing feature extraction on the abnormal vehicle topological graph and the full-quantity vehicle topological graph by adopting a deep neural network to obtain a first-order feature of the abnormal vehicle topological graph, a second-order feature of the abnormal vehicle topological graph, a first-order feature of the full-quantity vehicle topological graph and a vehicle vector representation of the full-quantity vehicle topological graph;
constructing an energy matrix according to the point-edge relation matrix, the first-order features of the abnormal vehicle topological graph, the second-order features of the abnormal vehicle topological graph, the first-order features of the full-quantity vehicle topological graph and the vehicle vector representation of the full-quantity vehicle topological graph;
solving the energy matrix to obtain a feature matrix, wherein elements in the feature matrix represent distances between each point in the abnormal vehicle topological graph and each point in the full-scale vehicle topological graph;
and calculating loss based on the characteristic matrix, and updating and iterating the parameters of the mapping module and the parameters of the deep neural network according to the calculated loss until convergence to obtain the depth map matching frame.
Further, the map building module is adopted to build an abnormal vehicle topological graph and a full-scale vehicle topological graph, and the method comprises the following steps:
constructing the abnormal vehicle topological graph according to a preset abnormal vehicle topological graph construction principle, and constructing the full-quantity vehicle topological graph according to a preset full-quantity vehicle topological graph construction principle;
the abnormal vehicle topological graph comprises working condition data of a preset number of time slices with preset lengths before an abnormal vehicle is abnormal, wherein each point represents the working condition data of one time slice of one abnormal vehicle;
the full-quantity vehicle topological graph comprises the working condition data of the last time slice of the abnormal vehicle, the working condition data of the last time slice of the normal vehicle and the working condition data of the last time slice of the unknown vehicle, wherein each point represents the working condition data of the last time slice of one vehicle.
Further, calculating a loss based on the feature matrix, comprising:
calculating the difference between each target element in the feature matrix and a corresponding real value thereof, wherein the target elements are determined by each abnormal vehicle in the abnormal vehicle topological graph and a vehicle with a known vehicle abnormal condition in the full-scale vehicle topological graph, the real value is 0 when the abnormal vehicle and the vehicle with the known vehicle abnormal condition are the same vehicle, the real value is 0 when the abnormal vehicle and the vehicle with the known vehicle abnormal condition have the same vehicle abnormal condition, and the real value is 1 when the abnormal vehicle and the vehicle with the known vehicle abnormal condition have different vehicle abnormal conditions;
and summing the obtained plurality of differences to obtain the loss.
Further, the method further comprises:
obtaining a target characteristic matrix obtained in the last updating iteration;
and determining the abnormal conditions of the unknown vehicles in the topological graph of the full-scale vehicle according to the target characteristic matrix.
Further, determining the abnormal condition of the unknown vehicle in the full-scale vehicle topological graph according to the target feature matrix, including:
extracting all element values related to the unknown vehicle in the target feature matrix;
determining a target element value among the element values, wherein the target element value is the top M smallest element values among the element values, and M is a positive integer;
determining a target abnormal vehicle in the abnormal vehicle topological graph corresponding to the target element value;
and taking the abnormal situation of the target abnormal vehicle as the abnormal situation of the unknown vehicle.
In a second aspect, an embodiment of the present invention further provides a vehicle abnormality prediction apparatus, including:
the system comprises an acquisition unit, a prediction unit and a prediction unit, wherein the acquisition unit is used for acquiring working condition data of a vehicle to be predicted;
the depth map matching processing unit is used for performing depth map matching processing on the working condition data by adopting a depth map matching frame to obtain the distance between the vehicle to be predicted and each vehicle under the known abnormal condition in the depth map matching frame;
and the determining unit is used for determining the abnormal situation of the vehicle to be predicted according to the distance.
In a third aspect, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the method according to any one of the above first aspects when executing the computer program.
In a fourth aspect, embodiments of the present invention also provide a computer-readable storage medium storing machine executable instructions, which when invoked and executed by a processor, cause the processor to perform the method of any of the first aspect.
In an embodiment of the present invention, there is provided a method of predicting a vehicle abnormality, including: firstly, acquiring working condition data of a vehicle to be predicted; then, carrying out depth map matching processing on the working condition data by adopting a depth map matching frame to obtain the distance between the vehicle to be predicted and each vehicle under the known abnormal condition in the depth map matching frame; and finally, determining the abnormal condition of the vehicle to be predicted according to the distance. According to the vehicle abnormity prediction method, the depth map matching frame is adopted to predict the vehicle abnormity of the vehicle to be predicted, the accuracy is good, and the technical problem that the existing vehicle abnormity prediction method is poor in accuracy is solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a method for predicting vehicle abnormality according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating an embodiment of the present invention for determining an abnormal situation of a vehicle to be predicted according to a distance;
FIG. 3 is a flowchart of a training method of a depth map matching framework according to an embodiment of the present invention;
FIG. 4 is a flowchart of a method for calculating loss based on a feature matrix according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a vehicle abnormality prediction apparatus according to an embodiment of the present invention;
fig. 6 is a schematic diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the following embodiments, and it should be understood that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
At present, when vehicle abnormality is predicted, the vehicle abnormality is predicted by a threshold, for example: when a certain parameter of the vehicle is larger than a set threshold value, the vehicle is determined to have certain abnormity. The vehicle abnormity prediction realized by the threshold value mode has poor accuracy.
Therefore, the embodiment provides the vehicle abnormity prediction method, the vehicle abnormity prediction is carried out on the vehicle to be predicted by adopting the depth map matching frame, and the accuracy is good.
Embodiments of the present invention are further described below with reference to the accompanying drawings.
The first embodiment is as follows:
in accordance with an embodiment of the present invention, there is provided an embodiment of a method for predicting vehicle anomalies, it is noted that the steps illustrated in the flowchart of the figures may be performed in a computer system, such as a set of computer-executable instructions, and that while a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in an order different than presented herein.
Fig. 1 is a flowchart of a method for predicting a vehicle abnormality according to an embodiment of the present invention, as shown in fig. 1, including the steps of:
step S102, obtaining working condition data of a vehicle to be predicted;
in the embodiment of the invention, the vehicle to be predicted can be any electric vehicle, and the working condition data can be acquired by a data acquisition system on the vehicle to be predicted in real time in the running process of the vehicle to be predicted.
The operating condition data may include: the present invention provides a method for monitoring a vehicle, which includes the steps of sampling time, mileage, EVCC (electric vehicle communication controller) feedback charging state, probe maximum temperature, frame number information, current vehicle speed, cell minimum voltage, probe minimum temperature, voltage, cell maximum voltage, current, whether a vehicle is parked, small battery voltage, insulation resistance, dimension, battery pack SOC (state of charge), acceleration, outdoor temperature, vehicle used life, longitude, total power, voltage difference between battery cells, temperature difference between battery cells, and the like.
Step S104, carrying out depth map matching processing on the working condition data by adopting a depth map matching frame to obtain the distance between the vehicle to be predicted and each vehicle under the known abnormal condition in the depth map matching frame;
the depth map matching frame is a frame obtained by training (which may also be referred to as update iteration) and is used for determining the distance between a vehicle corresponding to the working condition data input to the depth map matching frame and each vehicle in the depth map matching frame under the known abnormal condition.
The known abnormal condition vehicles may include one or more abnormal conditions, and when there are more abnormal conditions, the known abnormal condition vehicles may include: a normal vehicle, a vehicle in which an a abnormality occurs, a vehicle in which a B abnormality occurs, a vehicle in which a C abnormality occurs, and the like.
And step S106, determining the abnormal situation of the vehicle to be predicted according to the distance.
Specifically, the abnormal condition may be one or more of normal, occurrence of a-abnormality, occurrence of B-abnormality, occurrence of C-abnormality, and the like. The process will be described in detail below, and will not be described herein.
In an embodiment of the present invention, there is provided a method of predicting a vehicle abnormality, including: firstly, acquiring working condition data of a vehicle to be predicted; then, carrying out depth map matching processing on the working condition data by adopting a depth map matching frame to obtain the distance between the vehicle to be predicted and each vehicle under the known abnormal condition in the depth map matching frame; and finally, determining the abnormal condition of the vehicle to be predicted according to the distance. According to the vehicle abnormity prediction method, the depth map matching frame is adopted to predict the vehicle abnormity of the vehicle to be predicted, the accuracy is good, and the technical problem that the existing vehicle abnormity prediction method is poor in accuracy is solved.
The foregoing briefly introduces a method for predicting a vehicle abnormality according to the present invention, and the details thereof will be described in detail.
In an optional embodiment of the present invention, there are a plurality of abnormal situations covered by each vehicle with known abnormal situations, so that the depth map matching framework can predict a plurality of abnormal situations of the vehicle to be predicted at the same time, and the practicability is good.
In an alternative embodiment of the present invention, referring to fig. 2, the step S106 of determining the abnormal condition of the vehicle to be predicted according to the distance specifically includes the following steps:
step S201, determining a target distance in the distances, wherein the target distance is the distance with the minimum distance of the front N distances in the distances, and N is a positive integer;
specifically, after the distances are obtained, the distances are arranged in an ascending order, the first N distances are taken as target distances, and the size of N can be set according to actual needs.
Step S202, if the target distance is smaller than the preset threshold value, the abnormal situation of the vehicle with the known abnormal situation of the target corresponding to the target distance is taken as the abnormal situation of the vehicle to be predicted.
The number of the vehicles with known target abnormal conditions corresponding to the target distance can be multiple, and at this time, the multiple corresponding abnormal conditions are taken as the abnormal conditions of the vehicle to be predicted.
The above description details the process of vehicle anomaly prediction using the depth map matching framework, and the following describes the training process of the depth map matching framework in detail.
In an alternative embodiment of the present invention, referring to fig. 3, the training method of the depth map matching framework specifically includes the following steps:
s301, constructing an abnormal vehicle topological graph and a full-quantity vehicle topological graph by adopting a mapping module;
specifically, an abnormal vehicle topological graph is constructed according to a preset abnormal vehicle topological graph construction principle, and a full-quantity vehicle topological graph is constructed according to a preset full-quantity vehicle topological graph construction principle; the abnormal vehicle topological graph comprises working condition data of a preset number of time slices with preset lengths before an abnormal vehicle is abnormal, wherein each point represents the working condition data of one time slice of one abnormal vehicle; the full-scale vehicle topological graph comprises the working condition data of the last time slice of the abnormal vehicle, the working condition data of the last time slice of the normal vehicle and the working condition data of the last time slice of the unknown vehicle, wherein each point represents the working condition data of the last time slice of one vehicle.
The abnormal vehicle is a vehicle known to have a certain abnormality.
Step S302, determining a point-edge relation matrix of the abnormal vehicle topological graph according to the abnormal vehicle topological graph;
after the abnormal vehicle topological graph is obtained, the point-edge relation matrix of the abnormal vehicle topological graph can be determined. Specifically, when the abnormal vehicle topological graph has N points in total, the point-edge relation matrix is an N × N matrix, where the size of each element is 0 or 1. For example, N points are 1, 2, 3, …, N, respectively, it is default that there is a connecting edge between the same point, then there is a connecting edge between the 1 st point and the 1 st point, then the element at the 1 st row and 1 st column position in the point-edge relation matrix is 1; if a connecting edge also exists between the 1 point and the 2 nd point, the element of the 1 st row and the 2 nd column position in the point edge relation matrix is 1; if no connecting edge exists between the 1 point and the 3 rd point, the element of the 1 st row and the 3 rd column position in the point-edge relation matrix is 0, so that the point-edge relation matrix of the abnormal vehicle topological graph can be obtained.
Step S303, extracting the features of the abnormal vehicle topological graph and the full-quantity vehicle topological graph by adopting a deep neural network to obtain a first-order feature of the abnormal vehicle topological graph, a second-order feature of the abnormal vehicle topological graph, a first-order feature of the full-quantity vehicle topological graph and a vehicle vector representation of the full-quantity vehicle topological graph;
the first-order characteristic of the abnormal vehicle topological graph is a characteristic vector output by the first-layer model after the abnormal vehicle topological graph is input into the deep neural network, and the second-order characteristic of the abnormal vehicle topological graph is a characteristic vector output by the second-layer model after the abnormal vehicle topological graph is input into the deep neural network;
the first-order characteristic of the full-quantity vehicle topological graph is a characteristic vector output by the first layer of the model after the full-quantity vehicle topological graph is input into the deep neural network, and the vehicle vector of the full-quantity vehicle topological graph is represented as a vector output by the last hidden layer before the softmax layer in the deep neural network after the abnormal vehicle topological graph is input into the deep neural network.
The deep neural network comprises: a convolutional layer with graph feature extraction capability and a time-series layer with time-series feature extraction capability.
Step S304, constructing an energy matrix according to the point-edge relation matrix, the first-order characteristic of the abnormal vehicle topological graph, the second-order characteristic of the abnormal vehicle topological graph, the first-order characteristic of the full-quantity vehicle topological graph and the vehicle vector representation of the full-quantity vehicle topological graph;
specifically, when constructing the energy matrix, it is necessary to ensure that matrix multiplication is established, and then the calculation formula of the energy matrix is customized according to the business requirements. For example, the first-order feature of the full-quantity vehicle topological graph and the vehicle vector representation of the full-quantity vehicle topological graph are summed, then the sum is summed with the first-order feature of the abnormal vehicle topological graph, the obtained result is cross-multiplied by a point-side relation matrix, and finally the second-order feature of the abnormal vehicle topological graph is obtained. However, the embodiment of the present invention does not specifically limit the calculation manner of the energy matrix.
Step S305, solving the energy matrix to obtain a characteristic matrix, wherein elements in the characteristic matrix represent the distance between each point in the abnormal vehicle topological graph and each point in the full-capacity vehicle topological graph;
specifically, each point in the abnormal vehicle topological graph corresponds to each abnormal vehicle, and each point in the full-scale vehicle topological graph corresponds to each vehicle, so that the distance between each abnormal vehicle in the abnormal vehicle topological graph and each vehicle in the full-scale vehicle topological graph is obtained.
The feature matrix is a normalized feature matrix.
And S306, calculating loss based on the characteristic matrix, and updating and iterating the parameters of the mapping module and the parameters of the deep neural network according to the calculated loss until convergence to obtain a depth map matching frame.
Specifically, referring to fig. 4, the process of calculating the loss based on the feature matrix specifically includes the following steps:
step S401, calculating the difference between each target element in the feature matrix and the corresponding real value thereof, wherein the target elements are elements determined by each abnormal vehicle in the abnormal vehicle topological graph and the vehicle with the known vehicle abnormal condition in the full vehicle topological graph, when the abnormal vehicle and the vehicle with the known vehicle abnormal condition are the same vehicle, the real value is 0, when the abnormal vehicle and the vehicle with the known vehicle abnormal condition have the same vehicle abnormal condition, the real value is 0, and when the abnormal vehicle and the vehicle with the known vehicle abnormal condition have different vehicle abnormal conditions, the real value is 1;
specifically, when there are N points in the abnormal vehicle topological graph and M points in the total vehicle topological graph, the finally obtained feature matrix W is a feature matrix of N × M, where the target element W (i, j) represents an element determined by the abnormal vehicle i in the abnormal vehicle topological graph and the vehicle j in the total vehicle topological graph with a known vehicle abnormal condition (which may be no abnormality or some abnormality).
In addition, the above real value is 0 or 1, when the abnormal vehicle in the abnormal vehicle topological graph is the same vehicle as the vehicle with the known vehicle abnormal condition in the full-scale vehicle topological graph, the real value is 0, when the abnormal vehicle in the abnormal vehicle topological graph and the vehicle with the known vehicle abnormal condition in the full-scale vehicle topological graph have the same vehicle abnormal condition, the real value is also 0, when the abnormal vehicle in the abnormal vehicle topological graph and the vehicle with the known vehicle abnormal condition in the full-scale vehicle topological graph have different vehicle abnormal conditions (including two conditions, one is that the abnormal vehicle in the abnormal vehicle topological graph has a certain abnormal condition and the vehicle with the known vehicle abnormal condition in the full-scale vehicle topological graph is an abnormal-free vehicle, and the other is that the abnormal vehicle in the abnormal vehicle topological graph has an A abnormal condition and the vehicle with the known vehicle abnormal condition in the full-scale vehicle topological graph has a B abnormal condition), the true value is 1.
As in the above example, if the abnormal vehicle i in the abnormal vehicle topology map is the same vehicle or has the same vehicle abnormal condition as the vehicle j in the full-scale vehicle topology map with the known vehicle abnormal condition, the corresponding true value is 0, otherwise, the true value is 1.
Step S402, summing the obtained plurality of differences to obtain a loss.
Furthermore, the parameters of the mapping module and the parameters of the deep neural network are updated and iterated according to the calculated losses, and it should be noted that after the parameters of the mapping module are updated, when the mapping module constructs the vehicle topological graph, the above-mentioned construction principle is not changed, and only the positions of the points in the constructed vehicle topological graph are changed.
In an optional embodiment of the invention, after completing the iterative updating of the depth map matching framework, the method further comprises: obtaining a target characteristic matrix obtained in the last updating iteration; extracting all element values related to the unknown vehicle from the target feature matrix; determining a target element value in the element values, wherein the target element value is the first M smallest element values in the element values, and M is a positive integer; determining a target abnormal vehicle in the abnormal vehicle topological graph corresponding to the target element value; and taking the abnormal situation of the target abnormal vehicle as the abnormal situation of the unknown vehicle.
Example two:
the embodiment of the present invention further provides a vehicle abnormality prediction apparatus, which is mainly used for executing the vehicle abnormality prediction method provided in the first embodiment of the present invention, and the following describes the vehicle abnormality prediction apparatus provided in the first embodiment of the present invention in detail.
Fig. 5 is a schematic diagram of a vehicle abnormality prediction apparatus according to an embodiment of the present invention, as shown in fig. 5, the apparatus mainly including: an acquisition unit 10, a depth map matching processing unit 20 and a determination unit 30, wherein:
the system comprises an acquisition unit, a prediction unit and a prediction unit, wherein the acquisition unit is used for acquiring working condition data of a vehicle to be predicted;
the depth map matching processing unit is used for performing depth map matching processing on the working condition data by adopting the depth map matching frame to obtain the distance between the vehicle to be predicted and each vehicle under the known abnormal condition in the depth map matching frame;
and the determining unit is used for determining the abnormal situation of the vehicle to be predicted according to the distance.
In an embodiment of the present invention, there is provided a vehicle abnormality prediction apparatus including: firstly, acquiring working condition data of a vehicle to be predicted; then, carrying out depth map matching processing on the working condition data by adopting a depth map matching frame to obtain the distance between the vehicle to be predicted and each vehicle under the known abnormal condition in the depth map matching frame; and finally, determining the abnormal condition of the vehicle to be predicted according to the distance. According to the vehicle abnormity prediction device, the depth map matching frame is adopted to predict the vehicle abnormity of the vehicle to be predicted, the accuracy is good, and the technical problem that the existing vehicle abnormity prediction method is poor in accuracy is solved.
Optionally, there are a plurality of abnormal situations covered by each known abnormal situation vehicle.
Optionally, the determining unit is further configured to: determining a target distance in the distances, wherein the target distance is the distance with the minimum distance of the front N distances in the distances, and N is a positive integer; and if the target distance is smaller than the preset threshold value, taking the abnormal condition of the vehicle with the known abnormal condition of the target corresponding to the target distance as the abnormal condition of the vehicle to be predicted.
Optionally, the apparatus is further configured to: constructing an abnormal vehicle topological graph and a full-scale vehicle topological graph by adopting a mapping module; determining a point-edge relation matrix of the abnormal vehicle topological graph according to the abnormal vehicle topological graph; performing feature extraction on the abnormal vehicle topological graph and the full-quantity vehicle topological graph by adopting a deep neural network to obtain a first-order feature of the abnormal vehicle topological graph, a second-order feature of the abnormal vehicle topological graph, a first-order feature of the full-quantity vehicle topological graph and a vehicle vector representation of the full-quantity vehicle topological graph; constructing an energy matrix according to the point-edge relation matrix, the first-order characteristic of the abnormal vehicle topological graph, the second-order characteristic of the abnormal vehicle topological graph, the first-order characteristic of the full-quantity vehicle topological graph and the vehicle vector representation of the full-quantity vehicle topological graph; solving the energy matrix to obtain a characteristic matrix, wherein elements in the characteristic matrix represent distances between each point in the abnormal vehicle topological graph and each point in the full-scale vehicle topological graph; and calculating loss based on the characteristic matrix, and updating and iterating the parameters of the mapping module and the parameters of the deep neural network according to the calculated loss until convergence to obtain a depth map matching frame.
Optionally, the apparatus is further configured to: constructing an abnormal vehicle topological graph according to a preset abnormal vehicle topological graph construction principle, and constructing a full-quantity vehicle topological graph according to a preset full-quantity vehicle topological graph construction principle; the abnormal vehicle topological graph comprises working condition data of a preset number of time slices with preset lengths before an abnormal vehicle is abnormal, wherein each point represents the working condition data of one time slice of one abnormal vehicle; the full-scale vehicle topological graph comprises the working condition data of the last time slice of the abnormal vehicle, the working condition data of the last time slice of the normal vehicle and the working condition data of the last time slice of the unknown vehicle, wherein each point represents the working condition data of the last time slice of one vehicle.
Optionally, the device is further configured to calculate a difference between each target element in the feature matrix and a corresponding true value thereof, where the target elements are determined by each abnormal vehicle in the abnormal vehicle topology map and a vehicle with a known vehicle abnormal condition in the full-scale vehicle topology map, the true value is 0 when the abnormal vehicle and the vehicle with the known vehicle abnormal condition are the same vehicle, the true value is 0 when the abnormal vehicle and the vehicle with the known vehicle abnormal condition have the same vehicle abnormal condition, and the true value is 1 when the abnormal vehicle and the vehicle with the known vehicle abnormal condition have different vehicle abnormal conditions; the resulting plurality of differences are summed to obtain the penalty.
Optionally, the device is further configured to obtain a target feature matrix obtained in the last update iteration; and determining the abnormal conditions of the unknown vehicles in the topological graph of the full-scale vehicle according to the target characteristic matrix.
Optionally, the device is further configured to extract all element values related to the unknown vehicle in the target feature matrix; determining a target element value in the element values, wherein the target element value is the first M smallest element values in the element values, and M is a positive integer; determining a target abnormal vehicle in the abnormal vehicle topological graph corresponding to the target element value; and taking the abnormal situation of the target abnormal vehicle as the abnormal situation of the unknown vehicle.
The device provided by the embodiment of the present invention has the same implementation principle and technical effect as the method embodiments, and for the sake of brief description, reference may be made to the corresponding contents in the method embodiments without reference to the device embodiments.
As shown in fig. 6, an electronic device 600 provided in an embodiment of the present application includes: a processor 601, a memory 602 and a bus, wherein the memory 602 stores machine-readable instructions executable by the processor 601, when the electronic device is operated, the processor 601 and the memory 602 communicate with each other through the bus, and the processor 601 executes the machine-readable instructions to execute the steps of the method for predicting the vehicle abnormality.
Specifically, the memory 602 and the processor 601 can be general-purpose memories and processors, which are not specifically limited herein, and the prediction method of the vehicle abnormality can be performed when the processor 601 runs a computer program stored in the memory 602.
The processor 601 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware or instructions in the form of software in the processor 601. The Processor 601 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the device can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components. The various methods, steps, and logic blocks disclosed in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software module may be located in ram, flash memory, rom, prom, or eprom, registers, etc. storage media as is well known in the art. The storage medium is located in the memory 602, and the processor 601 reads the information in the memory 602 and completes the steps of the method in combination with the hardware thereof.
In response to the above method for predicting vehicle anomalies, the present application also provides a computer-readable storage medium storing machine-executable instructions that, when invoked and executed by a processor, cause the processor to perform the steps of the above method for predicting vehicle anomalies.
The vehicle abnormality prediction device provided by the embodiment of the application can be specific hardware on the device, or software or firmware installed on the device. The device provided by the embodiment of the present application has the same implementation principle and technical effect as the foregoing method embodiments, and for the sake of brief description, reference may be made to the corresponding contents in the foregoing method embodiments where no part of the device embodiments is mentioned. It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the foregoing systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
For another example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments provided in the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing an electronic device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the vehicle marking method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus once an item is defined in one figure, it need not be further defined and explained in subsequent figures, and moreover, the terms "first", "second", "third", etc. are used merely to distinguish one description from another and are not to be construed as indicating or implying relative importance.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present application, and are used for illustrating the technical solutions of the present application, but not limiting the same, and the scope of the present application is not limited thereto, and although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope disclosed in the present application; such modifications, changes or substitutions do not depart from the scope of the embodiments of the present application. Are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (11)
1. A method of predicting a vehicle abnormality, characterized by comprising:
acquiring working condition data of a vehicle to be predicted;
carrying out depth map matching processing on the working condition data by adopting a depth map matching frame to obtain the distance between the vehicle to be predicted and each vehicle under the known abnormal condition in the depth map matching frame;
and determining the abnormal condition of the vehicle to be predicted according to the distance.
2. The method of claim 1, wherein each known abnormal condition vehicle covers a plurality of abnormal conditions.
3. The method of claim 1, wherein determining an abnormal condition of the vehicle to be predicted from the distance comprises:
determining a target distance in the distances, wherein the target distance is the distance with the minimum distance of the first N distances in the distances, and N is a positive integer;
and if the target distance is smaller than a preset threshold value, taking the abnormal condition of the vehicle with the known abnormal condition of the target corresponding to the target distance as the abnormal condition of the vehicle to be predicted.
4. The method of claim 1, further comprising:
constructing an abnormal vehicle topological graph and a full-scale vehicle topological graph by adopting a mapping module;
determining a point-edge relation matrix of the abnormal vehicle topological graph according to the abnormal vehicle topological graph;
performing feature extraction on the abnormal vehicle topological graph and the full-quantity vehicle topological graph by adopting a deep neural network to obtain a first-order feature of the abnormal vehicle topological graph, a second-order feature of the abnormal vehicle topological graph, a first-order feature of the full-quantity vehicle topological graph and a vehicle vector representation of the full-quantity vehicle topological graph;
constructing an energy matrix according to the point-edge relation matrix, the first-order features of the abnormal vehicle topological graph, the second-order features of the abnormal vehicle topological graph, the first-order features of the full-quantity vehicle topological graph and the vehicle vector representation of the full-quantity vehicle topological graph;
solving the energy matrix to obtain a feature matrix, wherein elements in the feature matrix represent distances between each point in the abnormal vehicle topological graph and each point in the full-scale vehicle topological graph;
and calculating loss based on the characteristic matrix, and updating and iterating the parameters of the mapping module and the parameters of the deep neural network according to the calculated loss until convergence to obtain the depth map matching frame.
5. The method of claim 4, wherein constructing the abnormal vehicle topology map and the full-scale vehicle topology map using the mapping module comprises:
constructing the abnormal vehicle topological graph according to a preset abnormal vehicle topological graph construction principle, and constructing the full-quantity vehicle topological graph according to a preset full-quantity vehicle topological graph construction principle;
the abnormal vehicle topological graph comprises working condition data of a preset number of time slices with preset lengths before an abnormal vehicle is abnormal, wherein each point represents the working condition data of one time slice of one abnormal vehicle;
the full-quantity vehicle topological graph comprises the working condition data of the last time slice of the abnormal vehicle, the working condition data of the last time slice of the normal vehicle and the working condition data of the last time slice of the unknown vehicle, wherein each point represents the working condition data of the last time slice of one vehicle.
6. The method of claim 4, wherein computing a loss based on the feature matrix comprises:
calculating the difference between each target element in the feature matrix and a corresponding real value thereof, wherein the target elements are determined by each abnormal vehicle in the abnormal vehicle topological graph and a vehicle with a known vehicle abnormal condition in the full-scale vehicle topological graph, the real value is 0 when the abnormal vehicle and the vehicle with the known vehicle abnormal condition are the same vehicle, the real value is 0 when the abnormal vehicle and the vehicle with the known vehicle abnormal condition have the same vehicle abnormal condition, and the real value is 1 when the abnormal vehicle and the vehicle with the known vehicle abnormal condition have different vehicle abnormal conditions;
and summing the obtained plurality of differences to obtain the loss.
7. The method of claim 4, further comprising:
obtaining a target characteristic matrix obtained in the last updating iteration;
and determining the abnormal conditions of the unknown vehicles in the topological graph of the full-scale vehicle according to the target characteristic matrix.
8. The method of claim 7, wherein determining the unknown vehicle anomalies in the full-scale vehicle topology from the target feature matrix comprises:
extracting all element values related to the unknown vehicle in the target feature matrix;
determining a target element value among the element values, wherein the target element value is the top M smallest element values among the element values, and M is a positive integer;
determining a target abnormal vehicle in the abnormal vehicle topological graph corresponding to the target element value;
and taking the abnormal situation of the target abnormal vehicle as the abnormal situation of the unknown vehicle.
9. A vehicle abnormality prediction apparatus characterized by comprising:
the system comprises an acquisition unit, a prediction unit and a prediction unit, wherein the acquisition unit is used for acquiring working condition data of a vehicle to be predicted;
the depth map matching processing unit is used for performing depth map matching processing on the working condition data by adopting a depth map matching frame to obtain the distance between the vehicle to be predicted and each vehicle under the known abnormal condition in the depth map matching frame;
and the determining unit is used for determining the abnormal situation of the vehicle to be predicted according to the distance.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the steps of the method of any of the preceding claims 1 to 8 are implemented when the computer program is executed by the processor.
11. A computer readable storage medium having stored thereon machine executable instructions which, when invoked and executed by a processor, cause the processor to perform the method of any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110463152.5A CN113119890B (en) | 2021-04-27 | 2021-04-27 | Vehicle abnormity prediction method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110463152.5A CN113119890B (en) | 2021-04-27 | 2021-04-27 | Vehicle abnormity prediction method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113119890A true CN113119890A (en) | 2021-07-16 |
| CN113119890B CN113119890B (en) | 2023-04-18 |
Family
ID=76780888
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110463152.5A Active CN113119890B (en) | 2021-04-27 | 2021-04-27 | Vehicle abnormity prediction method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113119890B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108263312A (en) * | 2018-01-25 | 2018-07-10 | 奇瑞汽车股份有限公司 | Vehicle trouble alarm method and device |
| KR102025491B1 (en) * | 2018-04-03 | 2019-09-25 | 순천향대학교 산학협력단 | Augmented forward collision warning system and method based on prediction of vehicle braking distance |
| CN111284426A (en) * | 2018-12-06 | 2020-06-16 | 通用汽车环球科技运作有限责任公司 | Vehicle fault root cause diagnosis |
| CN111694917A (en) * | 2020-06-10 | 2020-09-22 | 北京嘀嘀无限科技发展有限公司 | Vehicle abnormal track detection and model training method and device |
| CN112529104A (en) * | 2020-12-23 | 2021-03-19 | 东软睿驰汽车技术（沈阳）有限公司 | Vehicle fault prediction model generation method, fault prediction method and device |
-
2021
- 2021-04-27 CN CN202110463152.5A patent/CN113119890B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108263312A (en) * | 2018-01-25 | 2018-07-10 | 奇瑞汽车股份有限公司 | Vehicle trouble alarm method and device |
| KR102025491B1 (en) * | 2018-04-03 | 2019-09-25 | 순천향대학교 산학협력단 | Augmented forward collision warning system and method based on prediction of vehicle braking distance |
| CN111284426A (en) * | 2018-12-06 | 2020-06-16 | 通用汽车环球科技运作有限责任公司 | Vehicle fault root cause diagnosis |
| CN111694917A (en) * | 2020-06-10 | 2020-09-22 | 北京嘀嘀无限科技发展有限公司 | Vehicle abnormal track detection and model training method and device |
| CN112529104A (en) * | 2020-12-23 | 2021-03-19 | 东软睿驰汽车技术（沈阳）有限公司 | Vehicle fault prediction model generation method, fault prediction method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113119890B (en) | 2023-04-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111572350A (en) | Electric automobile fire-starting early warning method and device | |
| EP3709039B1 (en) | Battery life learning device, battery life prediction device, method and program | |
| JP6651006B2 (en) | Automatic discovery method of gas station POI, automatic discovery device, storage medium and device | |
| EP4160784A1 (en) | Deterioration estimation device, model generation device, deterioration estimation method, model generation method, and program | |
| CN115422696B (en) | Module digital twin model building method, system, terminal and medium | |
| CN112816885A (en) | Battery abnormity detection method and device, electronic equipment and storage medium | |
| EP4005862A1 (en) | Method and system for detecting faults in a charging infrastructure system for electric vehicles | |
| CN111965476A (en) | Low-voltage diagnosis method based on graph convolution neural network | |
| Sreenuch et al. | Probabilistic Monte-Carlo method for modelling and prediction of electronics component life | |
| JP2018531458A6 (en) | Automatic discovery method, automatic discovery apparatus, storage medium and equipment for gas station POI | |
| AU2021204479A1 (en) | Battery failure or inadequate charge condition prediction method and system | |
| CN113657668A (en) | Power load prediction method and system based on LSTM network | |
| CN113219336A (en) | Battery deterioration determination system, method, and non-transitory storage medium storing program | |
| CN112434260A (en) | Road traffic state detection method and device, storage medium and terminal | |
| CN114399321A (en) | Business system stability analysis method, device and equipment | |
| CN116593897A (en) | Power battery fault diagnosis method, system, vehicle and storage medium | |
| CN115805810A (en) | Battery failure prediction method, apparatus, device, storage medium, and program product | |
| CN113326639B (en) | Method and device for determining automatic driving test scene and electronic equipment | |
| CN113119890B (en) | Vehicle abnormity prediction method and device and electronic equipment | |
| CN113868884A (en) | Power battery multi-model fault-tolerant fusion modeling method based on evidence theory | |
| CN115166521A (en) | Battery capacity evaluation method and device and electronic equipment | |
| CN113311336A (en) | Battery cell level capacity evaluation method and device and electronic equipment | |
| CN114043875A (en) | Remaining mileage estimation deviation analysis method and system based on big data | |
| CN113868236B (en) | Electric vehicle fault monitoring method, equipment and computer readable storage medium | |
| CN111721542B (en) | System and method for detecting faults or model mismatch |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |