A field vehicle health management system and fault prediction method

Abstract

The present invention discloses a field vehicle health management system and a fault prediction method, which belongs to the field of automobile management technology. The system includes a control data acquisition module, a motion data monitoring module, a storage database, a vehicle model management module, a health intelligent analysis module, a safety judgment module and an interactive display module; the control data acquisition module is used to collect control data during the user's vehicle driving process; the motion data monitoring module is used to collect motion data during the vehicle driving process; the storage database is used to store historical data; the vehicle model management module is used to establish a vehicle dynamic prediction model; the health intelligent analysis module is used to calculate the degree of deviation of the vehicle motion trajectory, determine the total value of the vehicle safety risk, and predict the probability of vehicle failure under different total values ​​of vehicle safety risk; the safety judgment module is used to judge whether the vehicle has a failure risk; the interactive display module is used for digital display.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile management, and in particular to a field vehicle health management system and a fault prediction method. Background Art

[0002] Field vehicle health management and fault prediction are important technologies to ensure vehicle performance and reliability. They involve real-time monitoring of vehicle status, fault diagnosis, prediction, and maintenance decision support. By integrating health monitoring and intelligent diagnostic functions, potential safety hazards can be discovered and handled in a timely manner, reducing unexpected failures and maintenance costs, thereby reducing the vehicle's full life cycle costs and improving vehicle safety and reliability.

[0003] Due to the long-distance nature of off-road vehicles, drivers often need to constantly control the vehicle during driving. Long-term and frequent vehicle control will cause irreversible damage to the vehicle. For example, in the vehicle's driving control system, frequent control of the steering wheel and pedals will reduce the service life of the driving control. For some minor control changes, it is difficult for the driver to intuitively feel them during driving, resulting in the continuous occurrence of safety accidents. The existing technology is insufficient in vehicle health management, resulting in the driver's inability to make a reasonable assessment of the vehicle's current health status, leading to further system failures and increasing the complexity of vehicle maintenance. Summary of the invention

[0004] The purpose of the present invention is to provide a field vehicle health management system and a fault prediction method to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for predicting faults of field vehicles, the method comprising the following steps:

[0006] S10, obtaining user permissions, collecting control data during the user's vehicle driving process; the control data represents the user's rotation control angle of the steering wheel and the pedaling pressure of the accelerator pedal and the brake pedal at the corresponding timestamp; collecting motion data during the vehicle driving process; the motion data represents the actual motion trajectory of the vehicle at the corresponding timestamp, image information around the vehicle and the total weight of the vehicle; wherein the collected control data and motion data are stored as historical data in the database;

[0007] S20, establishing a vehicle dynamic prediction model, analyzing the collected control data and motion data, and predicting the motion speed and direction of the vehicle under the corresponding control data;

[0008] S30, calculating the degree of deviation of the vehicle's motion trajectory according to the predicted vehicle's motion speed and direction and the vehicle's actual motion trajectory, determining the total value of the vehicle's safety risk, updating the database, and storing the degree of deviation of the vehicle's motion trajectory and the total value of the vehicle's safety risk as historical data in the database; based on the historical data, establishing a vehicle failure prediction model to predict the probability of vehicle failure under different total values ​​of vehicle safety risk;

[0009] S40: Compare the predicted probability of vehicle failure under different total vehicle safety risk values ​​with the vehicle safety failure probability threshold to determine whether the vehicle has a failure risk.

[0010] A field vehicle health management system, the system includes a control data acquisition module, a motion data monitoring module, a storage database, a vehicle model management module, a health intelligent analysis module, a safety judgment module and an interactive display module;

[0011] The control data acquisition module is used to obtain user permissions, collect control data during the user's vehicle driving process, and send the collected control data to the storage database; the control data represents the user's rotation control angle of the steering wheel and the pedal pressure of the accelerator pedal and the brake pedal at the corresponding timestamp;

[0012] The motion data monitoring module is used to collect motion data during vehicle driving and send the collected motion data to a database; the motion data represents the actual motion trajectory of the vehicle at the corresponding timestamp, image information around the vehicle and the total weight of the vehicle;

[0013] The storage database is used to store control data, motion data, vehicle motion trajectory deviation degree and vehicle safety risk total value as historical data;

[0014] The vehicle model management module is used to establish a vehicle dynamic prediction model, analyze the control data and motion data collected in the database, predict the vehicle's motion speed and direction under the corresponding control data, and send the predicted vehicle motion speed and direction to the health intelligence analysis module;

[0015] The health intelligence analysis module is used to calculate the degree of deviation of the vehicle's motion trajectory and determine the total value of the vehicle's safety risk based on the predicted vehicle's motion speed and direction and the vehicle's actual motion trajectory; establish a vehicle failure prediction model based on historical data in the database, predict the probability of vehicle failure under different vehicle safety risk total values, send the obtained vehicle motion trajectory deviation degree and safety risk total value to the interactive display module and the storage database, send the obtained probability of vehicle failure to the safety judgment module, and send the obtained vehicle motion trajectory deviation degree, safety risk total value and vehicle failure probability to the interactive display module;

[0016] The safety judgment module is used to compare the probability of vehicle failure under different vehicle safety risk total values ​​with the vehicle safety failure probability threshold to determine whether the vehicle has a failure risk;

[0017] The interactive display module is used to provide an on-board interactive platform, and the calculated vehicle motion trajectory deviation degree, total safety risk value and probability of vehicle failure are digitally displayed through the on-board interactive platform.

[0018] Compared with the prior art, the beneficial effects achieved by the present invention are: by establishing a vehicle dynamic prediction model, the collected control data and motion data are analyzed, the vehicle's movement speed and direction are predicted, so as to determine the deviation of vehicle control, provide data support for subsequent data analysis, and improve data analysis and mining analysis capabilities; by establishing a vehicle fault prediction model, the probability of vehicle failure under different vehicle safety risk total values ​​is predicted, so that the driver can more intuitively see the current vehicle safety status, thereby reducing the complexity of vehicle maintenance. At the same time, the probability of vehicle failure is predicted, which is convenient for evaluating the vehicle's health. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the steps of a method for predicting a field vehicle fault according to the present invention;

[0020] Figure 2 It is a structural schematic diagram of a field vehicle health management system of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in 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. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] See also Figure 1-2 , the present invention provides a technical solution:

[0023] See also Figure 1 In the first embodiment, a method for predicting a vehicle failure in a field is provided, the method comprising the following steps:

[0024] S10. Obtain user permissions and collect control data during the user's vehicle driving process; the control data represents the user's rotation control angle of the steering wheel and the pedaling pressure of the accelerator pedal and the brake pedal at the corresponding timestamp; collect motion data during vehicle driving; the motion data represents the actual motion trajectory of the vehicle at the corresponding timestamp, image information around the vehicle and the total weight of the vehicle; wherein the collected control data and motion data are stored in the database as historical data.

[0025] It should be noted that the historical data includes control data, motion data, the degree of deviation of the vehicle motion trajectory and the total value of the vehicle safety risk.

[0026] Furthermore, an angle sensor is installed on the steering wheel; the angle sensor is used to collect the user's rotation control angle of the steering wheel; pedal force sensors are installed on the accelerator pedal and the brake pedal; the pedal force sensors are used to collect the user's pedaling pressure on the accelerator pedal and the brake pedal; the vehicle is installed with a panoramic camera and an on-board dynamic weighing device; the panoramic camera is used to collect image information around the vehicle; wherein, the image information around the vehicle is analyzed, the road surface texture features are extracted, and the friction coefficient of the road surface is determined based on the road surface texture features; the on-board dynamic weighing device is used to determine the total weight of the vehicle by weighing the vehicle.

[0027] It should be noted that the analyzed vehicle driving scene is a scene in which the road friction coefficient can be directly determined through image information, and there are no obstructions around that affect the image information content; the road texture features include the texture depth, texture direction, and texture roughness of the road; a mathematical model between the road texture features and the friction coefficient is established, and a CNN neural network model is used to train the CNN neural network model using road image data with known friction coefficients, and the prediction accuracy of the model is verified through test data. The trained CNN neural network model is used to analyze the image information around the vehicle, and the road friction coefficient is determined based on the extracted road texture features, which provides a basis for subsequent data analysis and improves the accuracy of data analysis.

[0028] S20: Establish a vehicle dynamic prediction model, analyze the collected control data and motion data, and predict the movement speed and direction of the vehicle under the corresponding control data.

[0029] Specifically, the method steps are:

[0030] A vehicle dynamic prediction model was established and a simulation experiment was conducted to simulate the vehicle acceleration under different steering wheel rotation control angles, accelerator and brake pedal pressures, friction coefficients of the road surface, and gross vehicle weight when the vehicle mileage was 0.

[0031] Wherein, θ represents the user's rotation control angle of the steering wheel; F1 represents the user's pedal pressure on the accelerator pedal; F2 represents the user's pedal pressure on the brake pedal; μ represents the friction coefficient of the vehicle's driving road; w represents the total weight of the vehicle; represents the magnitude and direction of the vehicle's horizontal acceleration; f represents θ, F1, F2, μ, w and The functional relationship between

[0032] The collected control data and motion data are analyzed, and the speed and direction of the vehicle in the horizontal direction at timestamp t are obtained according to the actual motion trajectory of the vehicle. And determine the user's rotation control angle θ of the steering wheel at timestamp t t 、The user's pedal pressure on the accelerator pedal and the brake pedal F 1,t and F 2,t , gross vehicle weight w t , the friction coefficient of the road surface on which the vehicle is traveling μ t ;

[0033] According to the determined θ t 、F 1,t 、F 2,t 、w t and μ t , predict the magnitude and direction of the vehicle's horizontal speed at time t+1 when the vehicle's mileage is 0

[0034] The time step from t to t+1 is the system response time from the user's control data of the vehicle to the change of the actual motion state of the vehicle.

[0035] It should be noted that, in the present embodiment, the system response time step of the user's control angle of the steering wheel and the pedal pressure of the accelerator pedal and the brake pedal to the change in the actual motion state of the vehicle is the same; the calculated friction coefficient of the vehicle road surface from time stamp t to t+1 is obtained by averaging the friction coefficient under the actual motion trajectory; when the predicted vehicle mileage is 0, the movement speed and direction of the vehicle under the corresponding control data are used to determine the degree of deviation in vehicle control, thereby providing data support for subsequent data analysis and improving data analysis and mining capabilities.

[0036] S30. Calculate the degree of deviation of the vehicle's motion trajectory based on the predicted vehicle speed and direction and the vehicle's actual motion trajectory, determine the total value of the vehicle's safety risk, update the database, and store the degree of deviation of the vehicle's motion trajectory and the total value of the vehicle's safety risk as historical data in the database; based on the historical data, establish a vehicle failure prediction model to predict the probability of vehicle failure under different total values ​​of vehicle safety risk.

[0037] Specifically, the method steps are:

[0038] The predicted vehicle speed and direction at different timestamps and the actual vehicle trajectory are analyzed to establish a two-dimensional coordinate system in the horizontal direction. The vehicle position at time t+1 is taken as the origin of the coordinate system to determine the magnitude and direction of the vehicle's horizontal velocity at time t+1. and predicted Calculate the deviation degree E of the vehicle motion trajectory at time t+1 in the coordinate system t+1 :

[0039]

[0040] Among them, α t+1 Represents the velocity angle of the vehicle in the two-dimensional coordinate system at time t+1; Represents the predicted velocity angle of the vehicle in the two-dimensional coordinate system at time t+1;

[0041] According to the calculated E t+1 , determine the degree of deviation of the vehicle trajectory at different timestamps, and obtain the total value of vehicle safety risk

[0042]

[0043] Among them, E s Indicates the degree of deviation of the vehicle's motion trajectory at the sth moment; s = 1, 2, ..., t z ;t z Represents the timestamp corresponding to the vehicle's mileage;

[0044] The database is updated, the calculated vehicle motion trajectory deviation degree and the total value of vehicle safety risk are stored in the database as historical data, and historical data of vehicle safety accidents are screened out from the database;

[0045] It should be noted that the vehicle motion trajectory corresponding to the vehicle at the timestamp is determined, and the vehicle mileage is determined based on the vehicle's actual motion trajectory; Indicates the cumulative deviation of the vehicle trajectory calculated from the vehicle mileage from 0 to the current vehicle mileage; s and t z The definitions of are all time;

[0046] Establish a vehicle failure prediction model, analyze the historical data of vehicle safety accidents, and determine the total vehicle safety risk value at the corresponding timestamp when the vehicle safety accident occurred in different historical data according to The Weibull distribution is used to predict the probability of vehicle failure under different total vehicle safety risk values:

[0047]

[0048] in,

[0049] Where n represents the data value of vehicle safety accidents based on historical data analysis; represents the total vehicle safety risk value corresponding to the vehicle safety accident occurring in the i-th historical data analyzed; i=1, 2, ..., n; represents the average value of the total vehicle safety risk in the n historical data analyzed; represents the shape parameter; M represents the scale parameter; P represents the probability of a vehicle safety accident; e represents the base of the natural logarithm.

[0050] It should be noted that the safety accident refers to a traffic safety accident caused by a steering wheel or pedal failure of the vehicle; when the vehicle undergoes a systematic safety inspection and maintenance, the calculated total vehicle safety risk value will be reset to zero.

[0051] It should be noted that and They are Multiply the time step 1 and Multiply the time step by 1, so and In the two-dimensional coordinate system, they represent straight lines with magnitude and direction, which can also be understood as the trajectory of the vehicle. The area of ​​the triangle formed by the origin and the vehicle's motion trajectory deviation is obtained, which reflects the deviation value of the vehicle's steering wheel, pedals and other control systems. The deviation value is used as the vehicle's safety risk value to predict the vehicle's failure risk, thereby improving the accuracy of data analysis and the ability to prevent safety risks of field vehicles.

[0052] It should be noted that as the vehicle control system continues to wear out, the control deviation of the vehicle steering wheel and pedals will continue to increase, so and The condition of being greater than 0 will always be met.

[0053] It should be noted that based on the historical data of vehicle safety accidents, the risk of vehicle control system failure is evaluated through Weibull distribution, so as to calculate the probability of vehicle failure under different total vehicle safety risk values, help drivers to evaluate the health of the vehicle, and reduce the complexity of vehicle maintenance; the shape parameter It determines the shape of the Weibull distribution, reflecting the changing trend of vehicle safety accidents with the total value of vehicle safety risks; the scale parameter M determines the shape of the Weibull distribution, which can be understood as the average total value of safety risks before a vehicle safety accident occurs; compared with the prior art, the Weibull distribution can more intuitively predict the probability of vehicle safety accidents;

[0054] It should be noted that It is about the independent variable The functional relationship between P and the dependent variable is established. The vehicle mileage is determined according to the actual vehicle motion trajectory, and the timestamps corresponding to the vehicle mileage are determined respectively. The degree of deviation of the vehicle motion trajectory is determined according to the timestamps. The obtained degree of deviation of the vehicle motion trajectory is accumulated, and finally the corresponding total value of vehicle safety risk is determined and substituted into In the functional relationship, the probability of vehicle failure is determined, which facilitates the evaluation and management of vehicle failures.

[0055] S40: Compare the predicted probability of vehicle failure under different total vehicle safety risk values ​​with the vehicle safety failure probability threshold to determine whether the vehicle has a failure risk.

[0056] Further, the method for determining whether the vehicle has a failure risk is as follows: determining the vehicle safety failure probability threshold K; comparing K with to compare; when When the vehicle is judged to have no risk of failure; when When the vehicle is judged to have a risk of failure;

[0057] Among them, an in-vehicle interactive platform is provided to digitally display the calculated vehicle motion trajectory deviation degree, total safety risk value and probability of vehicle failure.

[0058] It should be noted that the vehicle safety failure probability threshold K is used to distinguish the degree of risk of vehicle failure, and can be determined by performing data statistical analysis by automotive professionals.

[0059] In this embodiment:

[0060] Comprehensive health management and fault prediction of steering, acceleration and braking control systems of off-road vehicles; the time step from t to t+1 is 0.15s;

[0061] Obtain user permissions, collect control data during the user's vehicle driving and motion data during the vehicle driving, and store the collected data in the database;

[0062] A vehicle dynamic prediction model was established and a computer simulation experiment was conducted to simulate the vehicle acceleration under different steering wheel rotation control angles, accelerator and brake pedal pressures, friction coefficients of the road surface and total vehicle weight when the vehicle mileage was 0.

[0063] In this embodiment, when the vehicle mileage is 0, the user's steering wheel rotation control angle is the same as the angle of change of the vehicle's acceleration direction in the horizontal direction, and the traction force F1 in the same direction as the vehicle's forward direction is determined according to the user's pedal pressure F1 on the accelerator pedal. 牵1 , the traction force F2 opposite to the vehicle's forward direction is determined according to the user's pedal pressure F2 on the brake pedal 牵2 , analyze the collected images around the vehicle, determine the friction coefficient μ of the road surface on which the vehicle is traveling, and determine the total weight w of the vehicle based on the vehicle weighing instrument; at this time, the predicted θ, F1, F2, μ, and w are The functional relationship between them is:

[0064] Where g represents the acceleration of gravity; direction a is the user's steering wheel rotation control angle;

[0065] Analyze the collected control data and motion data to predict the horizontal speed and direction of the vehicle at time t+1 when the vehicle mileage is 0

[0066] The predicted vehicle speed and direction at different timestamps and the actual vehicle trajectory are analyzed to establish a two-dimensional coordinate system in the horizontal direction. The vehicle position at time t+1 is taken as the origin of the coordinate system to determine the magnitude and direction of the vehicle's horizontal velocity at time t+1. and predicted Calculate the deviation degree E of the vehicle motion trajectory at time t+1 in the coordinate system t+1 :

[0067]

[0068] According to the calculated E t+1 , determine the degree of deviation of the vehicle trajectory at different timestamps, and obtain the total value of vehicle safety risk

[0069]

[0070] A vehicle failure prediction model is established, and the Weibull distribution is used to predict the probability of vehicle failure under different total vehicle safety risk values:

[0071]

[0072] The calculated vehicle motion trajectory deviation degree, total safety risk value and probability of vehicle failure are digitally displayed through the vehicle interactive platform; the vehicle safety failure probability threshold K=85 is determined;

[0073] At a certain time t0, the predicted probability of vehicle failure at this time Determine if the vehicle is at risk of failure, send a vehicle failure risk signal to the user, and remind the user to go to a car repair shop for repairs.

[0074] See also Figure 2 , in this implementation 2: a field vehicle health management system is provided, the system includes a control data acquisition module, a motion data monitoring module, a storage database, a vehicle model management module, a health intelligent analysis module, a safety judgment module and an interactive display module;

[0075] The control data acquisition module is used to obtain user permissions, collect control data during the user's vehicle driving process, and send the collected control data to the storage database; the control data represents the user's rotation control angle of the steering wheel and the pedal pressure of the accelerator pedal and the brake pedal at the corresponding timestamp;

[0076] The motion data monitoring module is used to collect motion data during vehicle driving and send the collected motion data to a database; the motion data represents the actual motion trajectory of the vehicle at the corresponding timestamp, image information around the vehicle and the total weight of the vehicle;

[0077] The storage database is used to store control data, motion data, vehicle motion trajectory deviation degree and vehicle safety risk total value as historical data;

[0078] The vehicle model management module is used to establish a vehicle dynamic prediction model, analyze the control data and motion data collected in the database, predict the vehicle's motion speed and direction under the corresponding control data, and send the predicted vehicle motion speed and direction to the health intelligence analysis module;

[0079] The health intelligence analysis module is used to calculate the degree of deviation of the vehicle's motion trajectory and determine the total value of the vehicle's safety risk based on the predicted vehicle's motion speed and direction and the vehicle's actual motion trajectory; establish a vehicle failure prediction model based on historical data in the database, predict the probability of vehicle failure under different vehicle safety risk total values, send the obtained vehicle motion trajectory deviation degree and safety risk total value to the interactive display module and the storage database, send the obtained probability of vehicle failure to the safety judgment module, and send the obtained vehicle motion trajectory deviation degree, safety risk total value and vehicle failure probability to the interactive display module;

[0080] The safety judgment module is used to compare the probability of vehicle failure under different vehicle safety risk total values ​​with the vehicle safety failure probability threshold to determine whether the vehicle has a failure risk;

[0081] The interactive display module is used to provide an on-board interactive platform, and the calculated vehicle motion trajectory deviation degree, total safety risk value and probability of vehicle failure are digitally displayed through the on-board interactive platform.

[0082] In this embodiment:

[0083] The control data acquisition module is interconnected with the database; the motion data monitoring module is interconnected with the database; the database is interconnected with the vehicle model management module; the database is interconnected with the health intelligent analysis module; the vehicle model management module is interconnected with the health intelligent analysis module; the health intelligent analysis module is interconnected with the interactive display module; the health intelligent analysis module is interconnected with the safety judgment module.

[0084] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for predicting outdoor vehicle failures, characterized in that: The method comprises the following steps: S10, obtaining user permissions, collecting control data during the user's vehicle driving process; the control data represents the user's rotation control angle of the steering wheel and the pedaling pressure of the accelerator pedal and the brake pedal at the corresponding timestamp; collecting motion data during the vehicle driving process; the motion data represents the actual motion trajectory of the vehicle at the corresponding timestamp, image information around the vehicle and the total weight of the vehicle; wherein the collected control data and motion data are stored as historical data in the database; S20, establishing a vehicle dynamic prediction model, analyzing the collected control data and motion data, and predicting the motion speed and direction of the vehicle under the corresponding control data; The method steps of step S20 are: S201, establish a vehicle dynamic prediction model, conduct simulation experiments, simulate the vehicle mileage is 0, the user has different steering wheel rotation control angles, accelerator pedal and brake pedal pressure, vehicle road friction coefficient and vehicle gross weight, and obtain the corresponding vehicle acceleration. Wherein, θ represents the user's rotation control angle of the steering wheel; F1 represents the user's pedal pressure on the accelerator pedal; F2 represents the user's pedal pressure on the brake pedal; μ represents the friction coefficient of the vehicle's driving road; w represents the total weight of the vehicle; represents the magnitude and direction of the vehicle's horizontal acceleration; f represents θ, F1, F2, μ, w and The functional relationship between S202: Analyze the collected control data and motion data, and obtain the speed and direction of the vehicle in the horizontal direction at timestamp t according to the actual motion trajectory of the vehicle. And determine the user's rotation control angle θ of the steering wheel at timestamp t t 、The user's pedal pressure on the accelerator pedal and the brake pedal F 1,t and F 2,t , gross vehicle weight w t , the friction coefficient of the road surface on which the vehicle is traveling μ t ; S203, according to the determined θ t 、F 1,t 、F 2,t 、w t and μ t , predict the magnitude and direction of the vehicle's horizontal speed at time t+1 when the vehicle's mileage is 0 Among them, the time step from t to t+1 is the system response time from the user's control data of the vehicle to the change of the actual motion state of the vehicle; S30, calculating the degree of deviation of the vehicle's motion trajectory according to the predicted vehicle's motion speed and direction and the vehicle's actual motion trajectory, determining the total value of the vehicle's safety risk, updating the database, and storing the degree of deviation of the vehicle's motion trajectory and the total value of the vehicle's safety risk as historical data in the database; based on the historical data, establishing a vehicle failure prediction model to predict the probability of vehicle failure under different total values ​​of vehicle safety risk; The method steps of step S30 are: S301, analyzing the predicted vehicle movement speed and direction at different time stamps and the actual movement trajectory of the vehicle, establishing a two-dimensional coordinate system in the horizontal direction, The vehicle position at time t+1 is taken as the origin of the coordinate system to determine the magnitude and direction of the vehicle's horizontal velocity at time t+1. and predicted Calculate the deviation degree E of the vehicle motion trajectory at time t+1 in the coordinate system t+1 : Among them, α t+1 Represents the velocity angle of the vehicle in the two-dimensional coordinate system at time t+1; Represents the predicted velocity angle of the vehicle in the two-dimensional coordinate system at time t+1; S302, according to the calculated E t+1 , determine the degree of deviation of the vehicle trajectory at different timestamps, and obtain the total value of vehicle safety risk Among them, E s Indicates the degree of deviation of the vehicle's motion trajectory at the sth moment; s = 1, 2, ..., t z ;t z Represents the timestamp corresponding to the vehicle's mileage; The database is updated, the calculated vehicle motion trajectory deviation degree and the total value of vehicle safety risk are stored in the database as historical data, and historical data of vehicle safety accidents are screened out from the database; S303: Establish a vehicle failure prediction model, analyze historical data of vehicle safety accidents, and determine the total vehicle safety risk value at the corresponding time stamp when the vehicle safety accident occurred in different historical data. according to The Weibull distribution is used to predict the probability of vehicle failure under different total vehicle safety risk values: in, Where n represents the data value of vehicle safety accidents based on historical data analysis; represents the total vehicle safety risk value corresponding to the vehicle safety accident occurring in the i-th historical data analyzed; i=1, 2, ..., n; represents the average value of the total vehicle safety risk in the n historical data analyzed; represents the shape parameter; M represents the scale parameter; P represents the probability of a vehicle safety accident; S40, comparing the predicted probability of vehicle failure under different vehicle safety risk total values ​​with the vehicle safety failure probability threshold to determine whether the vehicle has a failure risk; An angle sensor is installed on the steering wheel; the angle sensor is used to collect the user's rotation control angle of the steering wheel; pedal force sensors are installed on the accelerator pedal and the brake pedal; the pedal force sensors are used to collect the user's pedaling pressure on the accelerator pedal and the brake pedal; a panoramic camera and an on-board dynamic weighing device are installed on the vehicle; the panoramic camera is used to collect image information around the vehicle; wherein the image information around the vehicle is analyzed, the road surface texture features are extracted, and the friction coefficient of the road surface is determined based on the road surface texture features; the on-board dynamic weighing device is used to determine the total weight of the vehicle by weighing the vehicle.

2. A method for predicting outdoor vehicle failures according to claim 1, characterized in that: In step S40, the method for determining whether the vehicle has a failure risk is as follows: determining a vehicle safety failure probability threshold K; comparing K with to compare; when When the vehicle is judged to have no risk of failure; when When the vehicle is judged to have a risk of failure; Among them, an in-vehicle interactive platform is provided to digitally display the calculated vehicle motion trajectory deviation degree, total safety risk value and probability of vehicle failure.

3. A field vehicle health management system, using a field vehicle fault prediction method as described in any one of claims 1-2, characterized in that: The system includes a control data acquisition module, a motion data monitoring module, a storage database, a vehicle model management module, a health intelligence analysis module, a safety judgment module and an interactive display module; The control data acquisition module is used to obtain user permissions, collect control data during the user's vehicle driving process, and send the collected control data to the storage database; the control data represents the user's rotation control angle of the steering wheel and the pedal pressure of the accelerator pedal and the brake pedal at the corresponding timestamp; The motion data monitoring module is used to collect motion data during vehicle driving and send the collected motion data to a database; the motion data represents the actual motion trajectory of the vehicle at the corresponding timestamp, image information around the vehicle and the total weight of the vehicle; The storage database is used to store control data, motion data, vehicle motion trajectory deviation degree and vehicle safety risk total value as historical data; The vehicle model management module is used to establish a vehicle dynamic prediction model, analyze the control data and motion data collected in the database, predict the vehicle's motion speed and direction under the corresponding control data, and send the predicted vehicle motion speed and direction to the health intelligence analysis module; The health intelligence analysis module is used to calculate the degree of deviation of the vehicle's motion trajectory and determine the total value of the vehicle's safety risk based on the predicted vehicle's motion speed and direction and the vehicle's actual motion trajectory; Based on the historical data in the database, a vehicle failure prediction model is established to predict the probability of vehicle failure under different vehicle safety risk total values, the obtained vehicle motion trajectory deviation degree and safety risk total value are sent to the interactive display module and the storage database, the obtained probability of vehicle failure is sent to the safety judgment module, and the obtained vehicle motion trajectory deviation degree, safety risk total value and vehicle failure probability are sent to the interactive display module; The safety judgment module is used to compare the probability of vehicle failure under different vehicle safety risk total values ​​with the vehicle safety failure probability threshold to determine whether the vehicle has a failure risk; The interactive display module is used to provide an on-board interactive platform, and the calculated vehicle motion trajectory deviation degree, total safety risk value and probability of vehicle failure are digitally displayed through the on-board interactive platform.

Images