A method for diagnosing a failure of a steer-by-wire device

Abstract

The application discloses a failure diagnosis method of a steer-by-wire device. The working states of various functional units in the steer-by-wire device are monitored by a plurality of function monitoring sensors in a sensor network. When a failure of a functional unit is monitored, the functional unit is identified in the form of failure identification data, and the working state of the steer-by-wire device is evaluated according to a steering function failure evaluation parameter obtained from the failure identification data of the functional unit, so as to give a prompt for repairing or maintaining the steer-by-wire device. The failure identification data of the functional unit is obtained according to a unique mass value of each function monitoring sensor that discovers the failure. The failure identification data can not only identify the failure state of each functional unit of the steer-by-wire device, but also identify the failure position, so that the failure identification mode is greatly simplified, and the failure position of the function unit with a redundant design can be accurately identified.

Description

Technical Field

[0001] This application relates to the field of automotive steering system technology, and specifically to a fault diagnosis method for a steer-by-wire device. Background Technology

[0002] The automotive steer-by-wire system consists of three main parts: the steering wheel assembly, the steering actuator assembly, and the main controller (ECU), as well as auxiliary systems such as a fault diagnosis system and power supply. The steering wheel assembly includes the steering wheel, steering wheel angle sensor, torque sensor, and steering wheel return torque motor. Its main function is to convert the driver's steering intention (by measuring the steering wheel angle) into a digital signal and transmit it to the main controller; simultaneously, it receives torque signals from the main controller and generates steering wheel return torque to provide the driver with corresponding road feel information. The steering actuator assembly includes front wheel angle sensors, steering actuator motors, steering motor controllers, and front wheel steering components. Its function is to receive commands from the main controller and, through the steering motor controller, control the rotation of the steering wheels to achieve the driver's steering intention. The main controller analyzes and processes the collected signals, determines the vehicle's motion state, and sends commands to the steering wheel return motor and steering motor to control their operation, ensuring ideal vehicle response under various conditions. This reduces the driver's workload by compensating for changes in steering characteristics with vehicle speed. Simultaneously, the controller can also recognize the driver's operating commands and determine whether the driver's steering operation is reasonable in the current state. When the car is in an unstable state or the driver issues an incorrect command, the steer-by-wire system will block the driver's incorrect steering operation and automatically perform stabilization control to restore the car to a stable state as quickly as possible.

[0003] The fault diagnosis system is a crucial module of the steer-by-wire system. It comprises a series of monitoring and implementation algorithms that identify and address different fault types and levels to maximize the vehicle's operational stability. As one of the most widely used modes of transportation, vehicle safety is paramount; therefore, the ability to quickly and accurately identify fault diagnosis results is of utmost importance. Summary of the Invention

[0004] The main technical problem addressed in this application is how to quickly and accurately identify and locate the fault diagnosis results of the steer-by-wire device in order to improve the safety and reliability of the linear steering system.

[0005] According to a first aspect, one embodiment provides a fault diagnosis method for a steer-by-wire device, comprising:

[0006] In response to a preset fault detection command, fault diagnosis data collected by the sensor network is acquired; the sensor network includes multiple functional monitoring sensors, each of which is arranged in the steer-by-wire device to monitor the working status of each functional unit in the steer-by-wire device; wherein, each functional unit in the steer-by-wire device includes functional units added based on safety redundancy.

[0007] The fault status of each functional unit in the steer-by-wire device is determined based on the fault diagnosis data.

[0008] Based on the identification parameters of the functional monitoring sensors of the corresponding functional unit that has malfunctioned, the fault identification data of the functional unit that has malfunctioned is obtained; the identification parameters of each functional monitoring sensor include a preset confidence mass function (mass) value, and the confidence mass function (mass) values ​​corresponding to each functional monitoring sensor are different;

[0009] A steering function fault assessment parameter is obtained based on the fault identification data of each of the functional units that malfunctioned.

[0010] The steering function fault assessment parameters are displayed; the fault assessment parameters are used to evaluate and identify the functional implementation status of each functional unit of the steer-by-wire device; the steering function fault assessment parameters are real values.

[0011] When the value of the fault assessment parameter is greater than a preset functional monitoring assessment threshold, the steering-by-wire device is prompted to undergo inspection or maintenance.

[0012] In one embodiment, the fault diagnosis method further includes:

[0013] The functional unit that has malfunctioned is identified and displayed based on the steering function fault assessment parameters.

[0014] In one embodiment, the fault diagnosis method further includes:

[0015] Based on the fault identification data of the functional unit that has malfunctioned, the identification parameters of the corresponding functional monitoring sensor are obtained, so as to identify and display the fault location of the functional unit that has malfunctioned based on the identification parameters of the functional monitoring sensor.

[0016] In one embodiment, obtaining fault identification data of the malfunctioning functional unit based on the identification parameters of the functional monitoring sensor corresponding to the malfunctioning functional unit includes:

[0017] The confidence mass function (mass) values ​​of all functional monitoring sensors identified as faulty in the functional unit are calculated according to a preset confidence mass function algorithm to obtain fault identification data of the functional unit. The confidence mass function algorithm ensures that after calculating the confidence mass function values ​​of any number of functional monitoring sensors in the functional unit, the fault identification data values ​​obtained by different functional monitoring sensors will be different. This is used to inversely deduce the identification parameters of the functional monitoring sensors involved in the calculation based on the values ​​of the fault identification data, and then identify the fault location of the functional unit that has malfunctioned based on the identification parameters of the functional monitoring sensors.

[0018] In one embodiment, obtaining a steering function fault assessment parameter based on the fault identification data of each of the functional units that has failed includes:

[0019] The fault identification data of each functional unit that has failed is calculated according to the confidence quality function algorithm to obtain the steering function fault assessment parameters.

[0020] In one embodiment, the confidence mass function algorithm includes summation and / or product operations.

[0021] In one embodiment, the confidence mass function mass value is a real number greater than 0 and less than 1.

[0022] In one embodiment, the steer-by-wire device includes a steering mechanism module, an actuator motor module, and / or an associated module. The steering mechanism module includes functional units of mechanical structure for realizing steering function. The actuator motor module includes functional units of electromechanical control for realizing steering function. The associated module includes functional units for assisting and / or monitoring the steering mechanism module and the actuator motor module. The associated module includes sensors, a power supply, a communication and diagnostic link, and / or a control processor.

[0023] According to a second aspect, one embodiment provides a computer-readable storage medium storing a program that can be executed by a processor to implement the fault diagnosis method as described in the first aspect.

[0024] According to a third aspect, one embodiment provides a computer program product including a computer program and / or instructions, which, when executed by a processor, implement the fault diagnosis method as described in the first aspect.

[0025] According to the fault diagnosis method of the above embodiments, since the value of the fault identification data corresponds to all functional monitoring sensors that are determined to have malfunctioned, the fault identification data can not only identify the fault status of each functional unit (including redundant functional units) of the steer-by-wire device, but also identify the fault location of the functional unit, thereby greatly simplifying the fault identification method (displaying only a single value) and accurately identifying the fault location of redundant functional units. Attached Figure Description

[0026] Figure 1 This is a functional structural block diagram of a steer-by-wire device in one embodiment;

[0027] Figure 2 This is a schematic diagram illustrating the communication redundancy implementation of a data communication circuit in one embodiment.

[0028] Figure 3 This is a flowchart illustrating a fault diagnosis method in one embodiment. Detailed Implementation

[0029] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of this application. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to this application are not shown or described in the specification. This is to avoid obscuring the core parts of this application with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0030] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0031] The serial numbers assigned to components in this document, such as "first" and "second," are used only to distinguish the described objects and have no sequential or technical meaning. The terms "connection" and "linkage" used in this application, unless otherwise specified, include both direct and indirect connections (linkages).

[0032] Compared to other steering systems, steer-by-wire (SBW) eliminates the mechanical connection between the steering wheel and the steering wheels, transmitting driver control commands via electrical signals. For future vehicles to achieve fully autonomous driving, SBW can achieve steering servo control silently through the steering wheel, making it the most ideal active steering solution. To meet the high safety requirements of Level 3 and above autonomous driving for lateral vehicle control, redundant design of the steering wheel actuators is necessary.

[0033] Please refer to Figure 1 The diagram below illustrates the functional structure of a steer-by-wire device in one embodiment. The steer-by-wire device 1 includes a steering mechanism module 10, an actuator motor module 20, and / or an associated module 30. The steering mechanism module 10 includes mechanical structural units for implementing steering functions. The actuator motor module 20 includes electromechanical control units for implementing steering functions. The associated module 30 includes functional units for assisting and / or monitoring the steering mechanism module 10 and the actuator motor module 20. The associated module 30 includes sensors, a power supply, a communication and diagnostic link, and a control processor, among other devices or equipment related to the steer-by-wire device. Redundancy design ensures the safety and reliability of the steer-by-wire system; however, during maintenance, faults in the redundant parts are difficult to detect, increasing maintenance complexity. For example, please refer to... Figure 2 This is a schematic diagram of the communication redundancy implementation of the data communication circuit in one embodiment. The communication signal needs to pass through line A and line B. a1 and a2 are the redundancy design of line A, and b1 and b2 are the redundancy design of line B. When one of a1, a2 and / or b1, b2 fails, the A and B paths can still operate normally. However, when performing fault detection, since the A and B paths can still work normally, it is difficult to detect the faulty one among a1, a2, b1, and b2.

[0034] In this embodiment, different confidence mass function (mass) values ​​are set for the functional monitoring sensors of each functional unit of the steer-by-wire device, and these mass values ​​are used as identification parameters for the functional monitoring sensors. During maintenance and fault assessment, fault identification data of the faulty functional unit is obtained based on the identification parameters of the functional monitoring sensors of the corresponding faulty functional unit. Since the values ​​of the fault identification data correspond to all functional monitoring sensors determined to be faulty, the fault identification data can not only identify the fault status of each functional unit of the steer-by-wire device (including redundant functional units), but also identify the fault location of the functional unit. This greatly simplifies the fault identification method (displaying only a single value) and accurately identifies the fault location of redundant functional units.

[0035] Example 1:

[0036] Please refer to Figure 3 This is a flowchart illustrating a fault diagnosis method in one embodiment. The fault diagnosis method for a steer-by-wire device includes:

[0037] Step 101: Obtain fault diagnosis data.

[0038] In response to a preset fault detection command, fault diagnosis data collected by the sensor network is acquired. The sensor network includes multiple functional monitoring sensors, each positioned within the steer-by-wire device to monitor the operating status of various functional units within the device. These functional units include those added based on safety redundancy. In one embodiment, each functional unit is equipped with multiple functional monitoring sensors, each specifically positioned at a fixed location within the functional unit. These sensors monitor the fault status of functional units, including those with redundancy. When a fault is detected at a fixed location within a functional unit, the faulty functional unit and its location are determined based on the monitoring sensors at that location.

[0039] like Figure 1 As shown, since the steer-by-wire device 1 includes a steering mechanism module 10, an actuator motor module 20, and / or an associated module 30, the types, models, and operating modes of the corresponding functional monitoring sensors differ, resulting in different raw data types and parameters. Therefore, it is necessary to standardize the raw data of various types and parameters to simplify fault diagnosis. In one embodiment, the monitoring data acquired by each functional monitoring sensor is grouped and fused under the same data dimension, and data dimensionality reduction is performed, specifically including:

[0040] The discrete time-series data of several characteristic state signals were collected by a sensor network that is arranged on different modules of the steer-by-wire device. It is assumed that the number of individual functional monitoring sensors in the sensor network is M, regardless of whether each independent sensor collects the same type of physical signal, and that they have the same system update time and belong to the uniform sampling method.

[0041] First, considering that each functional monitoring sensor may have a different sampling rate, the numbering of the nth functional monitoring sensor in the sensor network containing M functional monitoring sensors is defined by the following formula:

[0042] n = 1, 2, ..., M; where M represents the sensor with the highest sampling rate, 1 represents the sensor with the lowest sampling rate, and the sensor numbered n has a sampling rate of Speed. n .

[0043] Secondly, it is assumed that the sampling rate ratio of each functional monitoring sensor, regardless of whether it has the same or different sampling rates, is a positive integer, where the sampling rate ratio of the functional monitoring sensor numbered i is defined as follows:

[0044] n i = Speed M Speed i (n) i ∈Z);

[0045] Among them, Speed M The sensor sampling rate corresponding to the highest sampling rate, where Z is a set of positive integers.

[0046] The sensor network utilizes data collected by each monitoring sensor or the relative strength of their signals, and performs group fusion to reduce data dimensionality and facilitate the subsequent extraction of key data features. The group fusion operation includes:

[0047] First, each functional monitoring sensor in the sensor network installed on the steer-by-wire device is described using discrete state-space equations. Specific steps include:

[0048] Consider the discrete-state state space system update equation for the i-th functional monitoring sensor in the sensor network, as shown in the following equation (where i is referred to as i=1, 2, ..., M):

[0049] x k+1 (i) =A k (i) x k (i) +q k (i) ;

[0050] Where x is the discrete state quantity data point collected by the sensor in the sensor network, which can be real data or its relative signal strength data. The relative signal strength is defined here as the ratio of the magnitude of the data value to the threshold value of the data; A is the system state transition matrix, and q is the predicted Gaussian noise of the state transition link, which satisfies that each element of its vector follows a multidimensional normal distribution with a mean of 0.

[0051] The covariance matrix is:

[0052] Q=E(qq T );

[0053] Where E(·) is the desired operation, i.e., q k ~N(0,Q k );

[0054] And x, q∈R[K×1] , A∈R [K×K] (K∈R);

[0055] Here, the superscripts [K×1] and [K×K] represent the dimensions of the real vectors x and q and the real matrix A, respectively. K represents one dimension of the real vector or real matrix. For most sensor types, K=1, but for some integrated sensors, such as the torque angle sensor of a steer-by-wire device, K≥1. Furthermore, the superscript (i) in the above formula represents the number of the i-th functional monitoring sensor in the sensor network.

[0056] Consider the discrete-state state space system measurement equation for the i-th functional monitoring sensor in the sensor network as follows:

[0057] z k (i) =H k (i) x k (i) +r k (i) ;

[0058] Where x represents discrete state quantity data points collected by sensors in the sensor network, which can be real data or its signal strength, and the signal strength is defined as the ratio of the data to the data threshold value; H is the system measurement matrix, and r is the Gaussian noise of the measurement link, which satisfies that each element of its vector follows a multidimensional normal distribution with a mean of 0.

[0059] The covariance matrix is:

[0060] R=E(rr T );

[0061] Where E(·) is the expectation operation, i.e., r k ~N(0,R k );

[0062] And x, r∈R [K×1] H∈R [qi×K] (K,q i ∈R);

[0063] The superscripts here are [K×1] and [q]. i [×K] represents the dimension of the real vectors x and r, and the real matrix H, where K represents a dimension of the real vector or real matrix. For most sensor types, K=1, but for some integrated sensors, such as the torque angle sensor in a steer-by-wire device, K≥1. i Let q represent the number of state variables that the i-th functional monitoring sensor in the sensor network wants to observe. i≤K. Secondly, the superscript (i) in the above equation represents the number of the i-th functional monitoring sensor in the sensor network; for the system update equation and the system measurement equation, the subscript k of the data vector refers to the k-th discrete sampled data point at the i-th sensor sampling rate scale represented by the superscript (i), and the subscript k of the matrix refers to the k-th discrete state value of the matrix corresponding to the k-th discrete sampled data point at the i-th sensor sampling rate scale represented by the superscript (i), for example, the covariance matrix Q. k and R k Measurement matrix H k and state transition matrix A k And for different sampling rates Speed i For a single sensor (numbered i), there is always a scale alignment relationship as shown in the following equation:

[0064] x k (i) = x nik (M) ;

[0065] Where, x nik (M) Represents the highest sampling rate Speed M The nth sensor at the sampling rate scale, numbered M i k discrete sampled data points. Scale alignment means establishing a one-to-one correspondence between the subscripts and numerical relationships of the kth discrete sampled data point or noise data point at the sampling rate scale of the i-th sensor and its corresponding kth discrete state value of the matrix, and the k′th discrete sampled data point or noise data point at the sampling rate scale of the M-th sensor (i.e., the highest sampling rate) and its corresponding k′th discrete state value of the matrix.

[0066] Consider extending the aforementioned scale alignment relationship to the system update equations, following the method shown below:

[0067] Substituting the scale alignment relation into the system update equation, we get:

[0068] x k (i) = x nik (M) ;

[0069] x k+1 (i) = x nik+ni (M) =A nik+ni-1 (M) x nik+ni-1 (M ) +q nik+ni-1 (M) ;

[0070] Continue expanding the above formula until the indices of each vector and matrix become n. i If k is given, then the derivation result is as shown in the following formula:

[0071] x k+1 (i) =···=

[0072] ;

[0073] The system update equation for the i-th sensor sampling rate scale is as follows:

[0074] ;

[0075] By comparison, the relationship is shown in the following formula:

[0076] ;

[0077] ;

[0078] If we assume that for noisy data points q k The scale alignment relationship is as follows:

[0079] ;

[0080] Among them, B k (i) Satisfy B k (i) =[A1···A m In the form of ], W k (i) =[q1···q m ] T The form (q here) m (where m is a row vector and m ∈ R), by analogy, the derivation result should satisfy the relationship shown in the following formula:

[0081] ;

[0082] W k (i) =[( q nik (M) ) T (q) nik+1 (M) ) T ···( q nik+ni-1 (M) ) T ] T

[0083] =[q nik (M);q nik+1 (M) ;··· ;q nik+ni-1 (M) ];

[0084] Here, E represents the identity matrix, and the superscript T indicates the transpose operation of a vector or matrix.

[0085] Based on the above, the system update equation at the sampling rate scale of the i-th sensor is rewritten as follows:

[0086] x k+1 (i) =A k (i) x k (i) +q k (i)

[0087] = ;

[0088] Wherein, the symbol ~ indicates that it follows a distribution, N(x, y) represents a Gaussian distribution, where x is the expectation of the distribution, y is the variance of the distribution, x is a vector with a dimension greater than 1, and y is a matrix with a dimension greater than 1×1.

[0089] Based on the rewritten system update equation and system measurement equation, Kalman filtering estimation is performed on the k-th discrete sampling data point at the sampling rate scale of the i-th functional monitoring sensor to obtain the optimal estimate. That is, the Kalman filtering estimation step is performed through the equation set shown below:

[0090] ;

[0091] ;

[0092] ;

[0093] ;

[0094] Among them, H k (i) R k (i) The function of each sensor in the sensor network above the different modules of the steering-by-wire system is determined by the characteristics of its own monitoring sensor. Additionally, it should be noted that after the Kalman filtering stage, the output of the iterative loop should consist of two parts:

[0095] Speed ​​of the sampling rate of the i-th sensor i The optimal estimate of the k-th discrete sampled data point at the specified scale. And the corresponding updated covariance matrix P, which describes the randomness error. k|k (i) ; where i satisfies i∈{1,2,…,M}, and the subscript k|k indicates that the estimated value is the optimal estimated state quantity under the current state (i.e. the kth discrete sampled data point under the sampling rate scale) obtained by correcting the coarse optimal estimated state quantity under the current state (i.e. the kth discrete sampled data point under the sampling rate scale) according to the least mean square estimation method.

[0096] The following method combines the scale-aligned state update equation of each functional monitoring sensor with the system measurement equation of each functional monitoring sensor and then applies Kalman filtering to group and fuse the estimated information of individual sensors to obtain the fused estimate corresponding to the sensor with the highest sampling rate. For several signals within each fusion group, the corresponding fusion method is performed according to the following steps, including:

[0097] First, the M functional monitoring sensors in the sensor network are grouped according to the type of signals they collect, resulting in G groups. Each group of signals is designated as Group1, Group2, ..., Group... g To represent a group of individual sensors, either single or a number of them;

[0098] Here, the subscript g satisfies g∈{1,2,…,G}, where:

[0099] M=∑ G g=1 card(Group g Here, card(·) is the operation for counting the elements in a combination, and it is specified that for any g∈Z (g=1,2,…,G), n g = card(Group g )∈{1,2,…,M g}, and also:

[0100] ;

[0101] Secondly, the group fusion algorithm is divided into two branches:

[0102] If for any single non-maximum sampling rate sensor number i∈{1,2,…,M} in the grouped signal, g -2, M g -1}, when kmodn i ≠0 (here n) i =Speed Mg / Speed i n iWhen ∈Z), output several signals within that group according to the following set of equations, aligning them to the Mth order. g Speed, the highest sampling rate of each sensor Mg The optimal estimate of the k-th discrete sampled data point after group fusion at the specified scale. and its corresponding update covariance matrix describing the randomness error. :

[0103] ;

[0104] in, , The output after passing through the Kalman filter is the Mth... g Sensor sampling rate Speed Mg The optimal estimate of the k-th discrete sampled data point at the scale and its corresponding updated covariance matrix describing the randomness error. If for a single non-maximum sampling rate sensor i within a grouped signal, there exists one or more i... J ≜i=i1,i2,…,i num (i1, i2, ..., i should be present here) num ≥1, J, num∈Z, and J, num≤M g -1, where the subscripts J and num refer to the sequence number of a single non-maximum sampling rate sensor i that satisfies the condition and their total number), when kmodn iJ =0 (here J∈{1,2,…,num},n) iJ = Speed Mg / Speed iJ n iJ When ∈Z), output several signals within that group according to the method described in the following set of equations, aligning them to the Mth order. g Speed, the highest sampling rate of each sensor Mg The optimal estimate of the k-th discrete sampled data point after group fusion at the specified scale. and its corresponding update covariance matrix describing the randomness error. :

[0105] ;

[0106] ;

[0107] ;

[0108] ;

[0109] Wherein, φ represents the weight of the updated covariance of the best estimate after grouping and fusion of signals collected by several individual sensors under the signal group of the sensor network. That is, it represents the weight of the updated covariance of the signal collected by a certain individual sensor under the signal group of the sensor network after Kalman filtering relative to the updated covariance of the best estimate after grouping and fusion. The subscript represents the sequence number of the discrete data points that can be sampled simultaneously with the sensor at the highest sampling rate.

[0110] After the grouping and fusion algorithm, the new discrete data point matrix of state variables obtained by the sensor network containing M functional monitoring sensors after grouping and fusion and data dimensionality reduction should be obtained. That is, the discrete state variable data point matrix of the sensor network of the steer-by-wire device satisfies the transformation relationship shown in the following formula:

[0111] [x (1) x (2) ··· x (M) → [C1 C2··· C G ] ;

[0112] Where, x (m) (m∈Z) represents the column vector composed of discrete data points collected by the m-th functional monitoring sensor in the sensor network, which can be understood as the m-th state vector describing the health status of the steer-by-wire device. C g (g∈{1,2,…,G}) represents the column vector composed of the g-th discrete data points of the steer-by-wire system, used to determine the health status of the steer-by-wire device after data processing. This can be understood as the g-th state vector used by the identification framework to determine the health status of the steer-by-wire device. Furthermore, for vector C... g The following relationship is satisfied:

[0113] ;

[0114] in, This refers to the individual sensor with the highest sampling rate within the g-th group of signals after group fusion, at its sampling rate Speed. Mg The fused signal result vector at the specified scale satisfies the following relationship:

[0115] ;

[0116] Where k∈Z, and its value depends on the sampling duration of the steering-by-wire device, as well as the characteristics of the individual sensor with the highest sampling rate within that signal group, i.e., the sampling rate Speed. MgThe magnitude of the value needs to be clarified, in order to guarantee the value of any vector C. g With consistent data dimensions, selecting a suitable benchmark sensor with the highest sampling rate during group fusion can achieve group fusion and data dimensionality reduction under the same data dimension.

[0117] Step 102: Obtain the fault status.

[0118] The fault status of each functional unit in the steer-by-wire device is determined based on the fault diagnosis data.

[0119] Step 103: Obtain fault identification data.

[0120] Based on the identification parameters of the functional monitoring sensors of the corresponding faulty functional unit, fault identification data of the faulty functional unit is obtained. The identification parameters of each functional monitoring sensor include a preset confidence mass function (mass) value, and the mass value corresponding to each functional monitoring sensor is different.

[0121] Step 104: Obtain steering function fault assessment parameters.

[0122] Based on the fault identification data of each functional unit that malfunctions, a steering function fault assessment parameter is obtained. The steering function fault assessment parameter is displayed. The fault assessment parameter is used to evaluate and identify the functional status of each functional unit of the steer-by-wire device. The steering function fault assessment parameter is a real value.

[0123] Step 105, Inspection and Maintenance Tips.

[0124] When the value of the fault assessment parameter is greater than a preset functional monitoring assessment threshold, the steer-by-wire device is prompted to be inspected or maintained.

[0125] Step 106: Display the faulty functional unit.

[0126] During vehicle operation, the user (driver) is only concerned with whether the vehicle is functioning normally. Therefore, the main control interface is relatively simple, only indicating whether the main functional units of the vehicle are working properly. In this embodiment, only one value (steering function fault assessment parameter) is needed to determine whether the steer-by-wire device has malfunctioned. Furthermore, during inspection and maintenance, it is necessary to know which specific functional unit in the steer-by-wire device has malfunctioned. In one embodiment, the malfunctioning functional unit is identified and displayed based on the steering function fault assessment parameter. To achieve this, it is only necessary to establish a correspondence between the steering function fault assessment parameter and all malfunctioning functional units.

[0127] Step 107: Display the location of the fault.

[0128] Based on the fault identification data of the malfunctioning functional unit, the identification parameters of the corresponding functional monitoring sensors for that functional unit are obtained. Based on these identification parameters, the fault location of the malfunctioning functional unit is identified and displayed. To identify the fault location of a functional unit based on the identification parameters of the functional monitoring sensors, it is only necessary to establish a correspondence between the identification parameters and the mass confidence function values ​​(mass values) of all functional monitoring sensors that detect fault locations within the functional unit. Specifically, this includes:

[0129] The confidence mass function (mass) values ​​of all functional monitoring sensors identified as faulty within a functional unit are calculated according to a preset confidence mass function algorithm to obtain fault identification data for the functional unit. The confidence mass function algorithm ensures that after calculating the mass values ​​of any number of functional monitoring sensors within the functional unit, the fault identification data values ​​obtained will differ depending on the participating functional monitoring sensors. This allows for the derivation of the identification parameters of the participating functional monitoring sensors based on the fault identification data values, and further identification of the fault location of the functional unit based on these parameters. Similarly, the same confidence mass function algorithm is used to calculate the fault identification data for each faulty functional unit to obtain steering function fault assessment parameters, thereby establishing a correspondence between the steering function fault assessment parameters and all faulty functional units. In one embodiment, the confidence mass function algorithm includes summation and / or product operations. In one embodiment, the confidence mass function value is a real number greater than 0 and less than 1.

[0130] To facilitate understanding of the implementation of the fault diagnosis method for the steer-by-wire device disclosed in this application, specific embodiments are described below, including:

[0131] (1) Define the fault category.

[0132] Establish a fault classification sequence, specifically including:

[0133] For the steering gear mechanical module, actuator motor module, and associated modules, the possible N1, N2, and N are... j (j≥2) types of faults, with the following rules and settings:

[0134] For the steering gear mechanical module, the N1 possible fault categories are logically independent and mutually exclusive, and the set of fault category subsequences is as follows:

[0135] Fault_set1 = {{Fault 1，1}, {Fault 2，1}, ..., {Fault N1-1，1}, {Fault N1，1}};

[0136] Fault_set2 = {{Fault 1，2}, {Fault 2，2}, ..., {Fault N2-1，2}, {Fault N2，2}};

[0137] Fault_set j ={{Fault 1，j}, {Fault 2，j}, ..., {Fault Nj-1，j}, {Fault Nj，j}};

[0138] Wherein, Fault_set represents the fault classification sequence, with index 1 indicating that the subsequence corresponds to the steering mechanism module of the steer-by-wire device, index 2 indicating that the subsequence corresponds to the actuator motor module of the steer-by-wire device, index j indicating that the subsequence corresponds to the associated module of the steer-by-wire device;

[0139] {Fault} represents the possible fault categories. The first item of its subscript represents the sequence number of the possible fault category. The second item, 1, means that there are N1 possible fault categories, all belonging to the steering mechanism module of the steer-by-wire system. The second item, 2, means that there are N2 possible fault categories, all belonging to the steering mechanism module of the steer-by-wire system. The second item, j, means that there are N possible fault categories. j All fault categories belong to the j-th module of the steer-by-wire device, i.e., the associated module of the steer-by-wire device.

[0140] For steer-by-wire systems, regardless of whether fault diagnosis of any other modules of the steer-by-wire system is included in the method, the set of all possible fault category sequences for the steer-by-wire system should be as follows:

[0141] Fault_set= Fault_set1∪Fault_set2∪···∪Fault_set j ;

[0142] Where j≥2, Fault_set represents the set of all possible fault categories of the steer-by-wire device, hereinafter referred to as the discrimination frame, and has:

[0143] Here, card(·) is the operation to count the number of elements in a set, and N is the set of positive integers.

[0144] For the steering gear mechanical module, the set of fault classification subsequences corresponding to the N1 possible fault categories has the following logical relationship:

[0145] {Fault p，1}∩{Fault q，1}=φ;

[0146] Where φ is an empty set, p, q∈{1, 2, ..., N1-1, N1}, and p≠q.

[0147] For the motor module, the set of fault classification subsequences corresponding to the N2 possible fault categories has the following logical relationship:

[0148] {Fault p，2}∩{Fault q，2}=φ;

[0149] Where φ is an empty set, p, q∈{1, 2, ..., N2-1, N2}, and p≠q.

[0150] For the j-th any other related module of the steer-by-wire device, the possible total N j The set of fault classification subsequences corresponding to each fault category has the following logical relationship:

[0151] {Fault p，j}∩{Fault q，j}=φ;

[0152] Where φ is an empty set, p, q∈{1, 2, ..., N} j -1,N j}, and p≠q.

[0153] For each fault category element in the identification framework:

[0154] {Fault n，I}(1≤I≤j,j,I∈N,n∈1,2,···,N I -1,N I});

[0155] Assign a confidence mass function value (mass) to each element, assuming the confidence mass function value falls within the interval [0, 1]. This process is called "basic probability allocation", hereinafter referred to as allocation, and can be expressed in the form shown in the following formula:

[0156] ;

[0157] Where 1≤I≤j,I∈N,j≥2,n∈{1,2,···,N1-1,N1},N is the set of positive integers, and the symbol → represents a one-to-one mapping relationship.

[0158] To identify each multi-element non-empty subset in the Fault_set framework:

[0159] ;

[0160] A confidence mass function value is also assigned to each multi-element non-empty subset. Assuming the confidence mass function value is in the interval [0,1], the assignment can be expressed in the form shown below:

[0161] ;

[0162] Where, i∈{1,2,···,2 card(Fault_set) - card(Fault_set)-1}; the symbol → represents a one-to-one mapping relationship, and card(fault_set)-1} i )>1, where card(·) is the operation to count the number of elements in the set. The physical meaning of a multi-element non-empty subset is a composite fault category. It is assumed that under this composite fault category, it can be determined that one or more specific fault categories or combinations thereof occur simultaneously in one or more modules within the online steering device.

[0163] For an empty set in the fault_set of the discrimination framework, a confidence mass function value is also assigned to it. The confidence mass function value is 0, which means that the hypothesis mass(φ) = 0 is satisfied.

[0164] Furthermore, regarding the aforementioned discrimination framework:

[0165] fault_set= Fault_set1∪Fault_set2∪···∪Fault_set j It should also satisfy the following setting:

[0166] ;

[0167] Where, i∈{1,2,···,2 card(Fault_set) - card(Fault_set)-1}, 1≤I≤j,I∈N,j≥2,n∈{1,2,···,N1-1,N1}, where card(·) is the operation to count the number of elements in the set, and N is the set of positive integers.

[0168] In one embodiment, the confidence mass function value assigned to any element or subset (whether the set is empty or not) in the fault_set should have a union operation rule (summation) and an intersection operation rule (product).

[0169] Union operations include:

[0170] When an element or set assigned by a confidence mass value (regardless of whether the set is empty) is joined with another element or set assigned by a confidence mass value (regardless of whether the set is empty), the confidence mass value assigned to the new set after the union operation should be the sum of the confidence mass values ​​assigned to all elements or sets (regardless of whether the set is empty) participating in the union operation, as shown in the following equation:

[0171] ;

[0172] Where, α i This represents a set or set element, where n∈N, and N is the set of positive integers.

[0173] The intersection operation rules include:

[0174] When an element or set assigned by a confidence mass value (whether the set is empty or not) is intersected with another element or set assigned by a confidence mass value (whether the set is empty or not), the confidence mass value assigned to the new set after the intersection operation should be the product of the confidence mass values ​​assigned to all elements or sets (whether the sets are empty or not) participating in the intersection operation, as shown in the following equation:

[0175] ;

[0176] Where, α i This represents a set or set element, where n∈N, and N is the set of positive integers.

[0177] When using the fault_set framework described above to identify fault categories for all modules of the steer-by-wire system, two fault categories are assigned: "local ignorance state" (functional unit still functions normally) and "global ignorance state" (functional unit does not function normally). Before describing the local ignorance state and the global ignorance state, the following settings are first made:

[0178] These two states of ignorance, like each fault category element in the aforementioned discrimination framework, can be assigned a confidence mass function value (mass) with a value between [0,1]. The sum of all assigned confidence mass function values ​​should be a fixed value.

[0179] Local ignorance and global ignorance are described as follows:

[0180] A state of partial ignorance represents an extended fault category that cannot be encompassed by any single logically independent and mutually exclusive fault category, after removing the sum of the confidence mass function mass values ​​corresponding to each logically independent and mutually exclusive fault category, when the method can be certain that only one fault category occurs simultaneously within one or more functional units of the steer-by-wire system. The state of partial ignorance is denoted by Θ. L This means that the following relationship should be satisfied:

[0181] ;

[0182] Where j represents the number of modules in the steer-by-wire system for which fault category identification is to be performed.

[0183] A global state of ignorance represents an extended fault category obtained when the method cannot fully determine whether only one fault category occurs simultaneously in one or more modules of the steering-by-wire device. This is achieved by removing the sum of the confidence mass function values ​​corresponding to each logically independent and mutually exclusive fault category, and further removing the confidence mass function values ​​corresponding to the local state of ignorance. The remaining confidence mass function values ​​are the fault categories that cannot be covered by any non-empty proper subset of the discrimination frame fault_set. This is then combined with the local state of ignorance and the discrimination frame to form an extended fault category.

[0184] Global ignorance state, denoted by Θ P This means that the following relationship should be satisfied:

[0185] ;

[0186] Among them, there is Θ L ∩Θ P =Θ L , i∈{1,2,···,2 card(Fault_set) - card(Fault_set)-2}, and have card (fault_set) i )>1, where card(·) is the operation to count the number of elements in the set. Furthermore, according to the relationship in the above formula, α∩Θ P =The derivation of α, where α represents a set or a set element.

[0187] For the fault_set discrimination framework, it is also necessary to clarify that the sum of the confidence mass functions assigned to the fault categories corresponding to all its subsets should be as shown in the following formula, regardless of whether the subset is an empty set or whether it has multiple elements:

[0188] ;

[0189] Here, α represents a set or a set element.

[0190] The confidence mass function assignment relationship described by the above formula exists in the following two cases:

[0191] a) When a method can assure the user, through a specific operation, that only one definite type of fault occurs simultaneously within one or more modules of the steering-by-wire system, only a state of local ignorance exists. L At this point, the confidence mass function assignment relationship shown in the following equation is satisfied:

[0192] ;

[0193] Where α represents a set or a set element;

[0194] b) When the method, through a specific operation, prevents the user from accurately knowing whether only one specific fault category occurs simultaneously within one or more modules of the steering-by-wire device, a state of partial ignorance exists. L and the state of global ignorance Θ P At this point, the confidence mass function assignment relationship shown in the following equation is satisfied:

[0195] ;

[0196] Here, Θ represents an extended fault category known as "generalized ignorance state," which includes the local ignorance state Θ. L and the state of global ignorance Θ P There are two types of states of ignorance, and in the above formula, α represents a set or a set element.

[0197] In one embodiment, a confidence rule formulation method and a confidence rule activation mechanism, as described below, are used to provide a basis for determining the fault category of all modules of the steer-by-wire device using the fault_set discrimination framework, namely the confidence rule, and the selection method of the confidence rule is listed.

[0198] The confidence rules, which form the basis for the fault category identification process for all modules of the steer-by-wire system using the fault_set discrimination framework, are described below:

[0199] (1) Confidence rules are a number of time-varying logical judgments that describe the one-to-one correspondence between signals collected by the sensor network and after data dimensionality reduction at different times t and the probability functions P corresponding to various fault categories. There are a total of Last confidence rules, each symbolized by Rule. l (t) represents, where l∈{1,2,···,Last-1,Last}, and for any probability function P, its value is between 0 and 1.

[0200] (2) For the steer-by-wire device, regardless of whether fault diagnosis is to be added for any associated modules in the steer-by-wire device other than the steering gear mechanical module and the actuator motor module, the confidence mass function assigned to each multi-element non-empty subset should first be used to expand the set of all possible fault category sequences of the steer-by-wire device (i.e., the discrimination framework fault_set).

[0201] The extended result of the discriminant framework is a temporal probabilistic discriminant framework that combines time and probability sequences. There are a total of *Last* such frameworks, the same number as the number of confidence rules. Each individual temporal probabilistic discriminant framework is defined by P. k ({Fault n,p},t) means that t represents different times, and the subscript k represents the sequence number of the individual temporal probability discrimination frame (where k is the index of the frame). l∈{1,2,···,Last-1,Last} represents the possible fault categories, where the first subscript n represents the sequence number of the possible fault category, with values ​​ranging from 1 to N. p (N, N p ∈Z), N p Represents the total number of possible N. p There are several fault categories, i.e., n∈{1,2,...,N}. p -1, N p}, where the second term p represents the N possible occurrences. p All fault categories belong to the p-th module (p∈Z) of the steer-by-wire system, i.e., any module of the steer-by-wire system. The process of transforming the original discrimination frame into a temporal probabilistic discrimination frame involves the following two steps:

[0202] S1. The original discrimination framework fault_set is described by the following formula:

[0203] Fault_set= Fault_set1∪Fault_set2∪···∪Fault_set j ;

[0204] Among them, Fault_setj The sequence represents the fault classification sequence, where the subscript j represents the subsequence corresponding to the j-th module of the steer-by-wire device (j∈Z), that is, any module of the steer-by-wire device. The symbol ∪ represents the union operation on the elements of the set.

[0205] S2. Temporal Probabilistic Discrimination Framework P k ({Fault n,p},t) is described by the following formula:

[0206] P k ({Fault n,p},t)=

[0207]

[0208]

[0209] ;

[0210] Among them, there is {Fault} n,p}∈Θ L ;P （n,p），k (t)(p∈{1,2,···,j}) represents the probability function P value of the possible fault category with sequence number n that may occur in the p-th module of the steer-by-wire device under the k-th confidence rule, derived by the confidence quality function mass and the corresponding confidence cross operation. Then, P satisfies... （n,p），k (t)∈[0,1];

[0211] P ΘL，k (t) represents the possible fault category of the steer-by-wire device under the k-th confidence rule, derived from the confidence mass function mass and the corresponding confidence cross operation, which is a state of local ignorance. L The probability function P value satisfies P ΘL，k (t) ∈ [0,1]; and for any time t, any single temporal probability discrimination frame sequence number k has the following relationship for the normalized probability function P value:

[0212] ;

[0213] Among them, there is k l∈{1,2,···, Last-1, Last}, where the value of j depends on the number of relevant modules considered when performing fault diagnosis on any other module of the steer-by-wire system, excluding the steering gear mechanical module and the actuator motor module. It should be clarified that for any individual temporal probability discrimination frame sequence number k, the temporal probability discrimination frame P...k ({Fault n,p In},t), the locally ignorant state Θ L The existence of this fault category is necessary to describe the possible errors caused by sensor noise, errors, and the different levels of confidence in different functionally monitored sensors in the sensor network.

[0214] The fault diagnosis method disclosed in this application monitors the operating status of each functional unit in the steer-by-wire device using multiple functional monitoring sensors in a sensor network. When a fault is detected in a functional unit, the faulty functional unit is identified using fault identification data. The operating status of the steer-by-wire device is then evaluated based on steering function fault assessment parameters obtained from the fault identification data of the faulty functional unit, providing timely prompts for repair or maintenance. The fault identification data of the functional unit is obtained based on the unique confidence mass function (mass) value of each functional monitoring sensor that detected the fault. Since the fault identification data can not only identify the fault status of each functional unit in the steer-by-wire device but also the fault location, it greatly simplifies the fault identification method and accurately identifies the fault location of redundantly designed functional units.

[0215] Those skilled in the art will understand that all or part of the functions of the various methods in the above embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above embodiments are implemented by computer programs, the program can be stored in a computer-readable storage medium, which may include: read-only memory, random access memory, disk, optical disk, hard disk, etc., and the program is executed by a computer to achieve the above functions. For example, the program can be stored in the memory of a device, and when the program in the memory is executed by the processor, all or part of the above functions can be achieved. In addition, when all or part of the functions in the above embodiments are implemented by computer programs, the program can also be stored in a server, another computer, disk, optical disk, flash drive, or external hard drive, etc., and can be downloaded or copied to the memory of a local device, or the system of the local device can be updated. When the program in the memory is executed by the processor, all or part of the functions in the above embodiments can be achieved.

[0216] The above examples illustrate the present invention only to aid in understanding it and are not intended to limit the scope of the invention. Those skilled in the art can make various simple deductions, modifications, or substitutions based on the principles of this invention.

Claims

1. A fault diagnosis method for a steer-by-wire device, characterized in that, include: In response to a preset fault detection command, fault diagnosis data collected by the sensor network is acquired; the sensor network includes multiple functional monitoring sensors, each of which is arranged in the steer-by-wire device to monitor the working status of each functional unit in the steer-by-wire device; wherein, each functional unit in the steer-by-wire device includes functional units added based on safety redundancy. The fault status of each functional unit in the steer-by-wire device is determined based on the fault diagnosis data. Based on the identification parameters of the functional monitoring sensors of the corresponding functional unit that has malfunctioned, the fault identification data of the functional unit that has malfunctioned is obtained; the identification parameters of each functional monitoring sensor include a preset confidence mass function (mass) value, and the confidence mass function (mass) values ​​corresponding to each functional monitoring sensor are different; A steering function fault assessment parameter is obtained based on the fault identification data of each of the functional units that malfunctioned. The steering function fault assessment parameters are displayed; the fault assessment parameters are used to evaluate and identify the functional implementation status of each functional unit of the steer-by-wire device; the steering function fault assessment parameters are real values. When the value of the fault assessment parameter is greater than a preset functional monitoring assessment threshold, the steering-by-wire device is prompted to undergo inspection or maintenance.

2. The fault diagnosis method as described in claim 1, characterized in that, Also includes: The functional unit that has malfunctioned is identified and displayed based on the steering function fault assessment parameters.

3. The fault diagnosis method as described in claim 2, characterized in that, Also includes: Based on the fault identification data of the functional unit that has malfunctioned, the identification parameters of the corresponding functional monitoring sensor are obtained, so as to identify and display the fault location of the functional unit that has malfunctioned based on the identification parameters of the functional monitoring sensor.

4. The fault diagnosis method as described in claim 3, characterized in that, The step of obtaining fault identification data of the malfunctioning functional unit based on the identification parameters of the functional monitoring sensor corresponding to the malfunctioning functional unit includes: The confidence mass function (mass) values ​​of all functional monitoring sensors identified as faulty in the functional unit are calculated according to a preset confidence mass function algorithm to obtain fault identification data of the functional unit. The confidence mass function algorithm ensures that after calculating the confidence mass function values ​​of any number of functional monitoring sensors in the functional unit, the fault identification data values ​​obtained by different functional monitoring sensors will be different. This is used to inversely deduce the identification parameters of the functional monitoring sensors involved in the calculation based on the values ​​of the fault identification data, and then identify the fault location of the functional unit that has malfunctioned based on the identification parameters of the functional monitoring sensors.

5. The fault diagnosis method as described in claim 4, characterized in that, The step of obtaining a steering function fault assessment parameter based on the fault identification data of each of the functional units that has failed includes: The fault identification data of each functional unit that has failed is calculated according to the confidence quality function algorithm to obtain the steering function fault assessment parameters.

6. The fault diagnosis method as described in claim 4, characterized in that, The confidence mass function operation rules include summation and / or product operations.

7. The fault diagnosis method as described in claim 4, characterized in that, The confidence mass function mass value is a real number greater than 0 and less than 1.

8. The fault diagnosis method as described in claim 4, characterized in that, The steer-by-wire device includes a steering mechanism module, an actuator motor module, and / or an associated module. The steering mechanism module includes mechanical structure functional units for realizing the steering function. The actuator motor module includes electromechanical control functional units for realizing the steering function. The associated module includes functional units for assisting and / or monitoring the steering mechanism module and the actuator motor module. The associated module includes sensors, a power supply, a communication and diagnostic link, and / or a control processor.

9. A computer-readable storage medium, characterized in that, The medium stores a program that can be executed by a processor to implement the fault diagnosis method as described in any one of claims 1 to 8.

10. A computer program product comprising a computer program and / or instructions, characterized in that, When the computer program and / or instructions are executed by the processor, they implement the fault diagnosis method as described in any one of claims 1 to 8.

Images