Method and system for predicting dynamic state of electric heavy-duty vehicle and terminal equipment

Abstract

The invention discloses a dynamic state prediction method and system for an electric heavy-duty vehicle and terminal equipment, and relates to the technical field of new energy vehicles. The method comprises the steps that heavy-load vehicle driving state information data are received and preprocessed, and the heavy-load vehicle driving state information data comprise the longitudinal acceleration, the transverse acceleration, the yaw velocity, the wheel rotating speed and the wheel rotating angle; and constructing a transverse speed model based on a Koopman operator according to the driving state information data, wherein the transverse speed model is used for predicting the transverse speed of the heavy-duty vehicle. Aiming at the problem that the transverse speed of the electric heavy-load vehicle is usually unmeasurable, a transverse speed model based on a Koopman operator is provided, the problem that the transverse speed cannot be obtained through a traditional method can be solved, a linear heavy-load vehicle dynamic model is provided to represent the nonlinear characteristics of the heavy-load vehicle, nonlinear modeling of a heavy-load vehicle system is avoided, and the method is suitable for the large-scale production of the electric heavy-load vehicle. And subsequent controller design is facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy vehicles, and in particular to a method, system and terminal device for predicting the dynamic state of an electric heavy-duty vehicle. Background Art

[0002] With the rapid development of the automobile industry and the increasing requirements for environmental protection and energy conservation, new energy vehicles, as an environmentally friendly and energy-saving vehicle with great development prospects, have received more and more attention and application. Most of the existing methods for modeling electric vehicle dynamics are based on the derivation of physical laws. This method requires the calibration of a large number of vehicle parameters, such as vehicle mass, inertia, center of mass position, etc., which is not only difficult to calibrate, but also not adaptable to working conditions. There are also machine learning methods for modeling vehicle dynamics, but this method cannot obtain lateral velocity, and this method is essentially nonlinear modeling, which poses challenges to subsequent controller design. Summary of the invention

[0003] The purpose of the present invention is to provide a method, system and terminal device for predicting the dynamic state of an electric heavy-duty vehicle, which not only solves the estimation of the lateral velocity, but more importantly, the model finally obtained is a linear model, which is convenient for subsequent controller design.

[0004] According to a first aspect of the present invention, in order to achieve the above-mentioned purpose, the present invention provides the following technical solution: a method for predicting the dynamic state of an electric heavy-duty vehicle, comprising the following steps:

[0005] Receiving heavy-load vehicle driving state information data, wherein the heavy-load vehicle driving state information data includes longitudinal acceleration, lateral acceleration, yaw angular velocity, wheel speed, and wheel angle;

[0006] A lateral velocity model based on the Koopman operator is constructed according to the driving status information data to predict the lateral velocity of the heavy-loaded vehicle;

[0007] Based on the nonlinear dynamic model and the predicted lateral velocity, a linear heavy-load vehicle dynamics model based on the Koopman operator is constructed to predict the dynamic state of the heavy-load vehicle.

[0008] Furthermore, the longitudinal acceleration, lateral acceleration, yaw angular velocity, wheel speed, and wheel angle in the heavy-load vehicle driving status information data are respectively collected by an acceleration sensor, yaw angular velocity sensor, speedometer, and angle sensor installed on the vehicle itself.

[0009] Furthermore, a lateral velocity model based on the Koopman operator is constructed according to the driving state information data to predict the lateral velocity of the heavy-loaded vehicle, as follows:

[0010] (31) Let the state vector xvy =[a xk ,a yk ,γ k ,w k ], the input vector is u vyk =δ f , define the state dimension-raising function of the Koopman operator for lateral velocity as σ k (x vyk ,u vyk ), its dynamic evolution is as follows:

[0011] σ k (x vyk+1 ,u vyk+1 )=σ k (g(x vyk ,u vyk ))=K vy σ k (x vyk ,u vyk );

[0012] (32) Assume that the Koopman operator is a set of matrices (A vy ,B vy ,C vy ,D vy ) to achieve, let p k =σ k (x vyk ,u vyk ), and design the following linear dynamics model:

[0013] p k+1 =A vy p k +B vy u vyk

[0014] v yk =C vy p k +D vy u vyk

[0015] In the formula, p k is the state of the kth step after dimensionality increase, p k+1 is the state of the k+1th step after dimensionality upgrade;

[0016] (33) Collect the signals from time 0 to time M and define them as follows:

[0017] P M =[p 0 p 1 …p M-1 ]

[0018] P M+1 =[p 1 p 2 …p M ]

[0019] U vyM =[u vy0 u vy1 …u vyM-1 ]

[0020] V yM =[v y0 v y1 …v yM-1 ]

[0021] Where P M is the vector matrix of states from 0 to M-1, P M+1 is the vector matrix of states from 1 to M; U vyM V is the vector matrix from 0 to M-1 for the steering input; yM is the vector matrix of lateral velocity from 0 to M-1; M is the data length;

[0022] The Koopman linear model of lateral velocity can be expanded into the following form through data:

[0023] P M+1 =A vy P M +B vy U vyM

[0024] V yM =C vy P M +D vy U vyM

[0025] The system matrix group (A vy , B vy , C vy , D vy ):

[0026]

[0027] In the formula ‖·‖ F Represents the Frobenius norm of the corresponding matrix; by solving the above formula, the lateral velocity can be predicted.

[0028] Furthermore, the construction of the nonlinear dynamic model is as follows:

[0029] (41) According to the laws of vehicle motion and Newtonian mechanics, the lateral and longitudinal coupling dynamics model of a traditional heavy-duty vehicle is described by the following equation:

[0030]

[0031] Where m is the mass of the vehicle, in kg; v x is the longitudinal velocity of the vehicle in m / s; v y is the lateral velocity of the vehicle, in m / s; γ is the yaw rate of the vehicle, in rad / s; δ f is the front wheel turning angle, in rad; I z is the yaw inertia, unit: kg·m 2 ; L f is the distance from the center of mass of the vehicle to the front axle, in meters; L r is the distance from the vehicle's center of mass to the rear axle, in meters; Refers to the longitudinal tire force of the front wheel, unit N; Refers to the longitudinal tire force of the rear wheel, unit N; Refers to the lateral tire force of the front wheel, unit N; Refers to the lateral tire force of the rear wheel, unit N;

[0032] (42) Select the state vector as x k =[v x ,v y ,γ k ], the input vector is u k =δ f , then the vehicle coupling dynamics model in (41) can be simplified to a general nonlinear dynamics model, expressed as follows:

[0033] x k+1 =f(x k ,u k )

[0034] In the formula, x k is the state vector, u k is the input vector.

[0035] Furthermore, based on the nonlinear dynamic model and the predicted lateral velocity, a linear heavy-load vehicle dynamic model based on the Koopman operator is constructed to predict the dynamic state of the heavy-load vehicle, as follows:

[0036] (51) The Koopman operator state dimension-raising function of heavy-load vehicle dynamics is defined as φ k (x k ,u k ), its dynamic evolution is as follows:

[0037] φ k (x k+1 ,u k+1 )=φ k (f(xk ,u k ))=K vd φ k (x k ,u k )

[0038] Where K vd is the Koopman operator; the state matrix x k Contains lateral velocity, which is obtained by the lateral velocity model based on the Koopman operator;

[0039] (52) Let z k =φ k (x k ,u k ), and assume that the Koopman operator K vd is a set of matrices (A vd ,B vd ,C vd ,D vd ) is realized, the above formula can be transformed into the general form of linear control system:

[0040] z k+1 =A vd z k +B vd u vdk

[0041] y k =C vd z k +D vd u vdk

[0042] Where z k ∈R q is the dimension-increased state vector of the original linear control system, (A vd ,B vd ,C vd ,D vd ) is the system matrix determined by the Koopman operator;

[0043] (53) By collecting the state data x of the linear control system k With input data u k , combined with the definition of the dimension-raising function, and define the following data matrix:

[0044] Z k =[z 0 z 1 …z N-1 ]

[0045] Z k+1 =[z 1 z2 …z N ]

[0046] U k =[u 0 u 1 …u N-1 ]

[0047] Y k =[y 0 y 1 …y N-1 ]

[0048] Where N is the selected data length or data step length, Z k , Z k+1 , U k With Y k The data segments corresponding to the 0th to N-1th step of the dimension-upgrading state, the 1st to Nth step of the dimension-upgrading state, the 0th to N-1th step of the steering input, and the 0th to N-1th step of the output vector are respectively, where Z k With Z k+1 Only one time step is missing;

[0049] (54) Using the above linear control system relationship, the relationship between the data is as follows:

[0050] Z k+1 =A vd Z k +B vd U k

[0051] Y k =C vd Z k +D vd U k

[0052] The system matrix group (A vd ,B vd ,C vd ,D vd )

[0053]

[0054] Thus, the linear heavy-load vehicle dynamics model after dimensionality increase is obtained.

[0055] According to a second aspect of the present invention, the present invention provides an electric heavy-duty vehicle dynamics state prediction system, which is used to implement the above-mentioned electric heavy-duty vehicle dynamics state prediction method, comprising:

[0056] A receiving module is used to receive and pre-process the driving state information data of the heavy-loaded vehicle, wherein the driving state information data of the heavy-loaded vehicle includes longitudinal acceleration, lateral acceleration, yaw angular velocity, wheel speed, and wheel angle;

[0057] The first construction module is used to construct a lateral velocity model based on the Koopman operator according to the driving state information data, so as to predict the lateral velocity of the heavy-loaded vehicle;

[0058] The second construction module is used to construct a linear heavy-load vehicle dynamics model based on the Koopman operator based on the nonlinear dynamics model and the predicted lateral velocity, so as to realize the prediction of the dynamic state of the heavy-load vehicle.

[0059] According to a third aspect of the present invention, the present invention provides a terminal device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the memory stores a computer program capable of running on the processor, and when the processor loads and executes the computer program, the above-mentioned method for predicting the dynamic state of an electric heavy-load vehicle is adopted.

[0060] According to a fourth aspect of the present invention, the present invention provides a storage medium comprising computer executable instructions, wherein the computer executable instructions are used to execute the above-mentioned method for predicting the dynamic state of an electric heavy-duty vehicle when executed by a computer processor.

[0061] The present invention has at least the following beneficial effects:

[0062] 1. Aiming at the problem that the lateral velocity of electric heavy-duty vehicles is usually unmeasurable, the present invention proposes a lateral velocity model based on the Koopman operator using the Koopman operator, which can solve the problem that the traditional method cannot obtain the lateral velocity, and there is no need to obtain the full state parameters of the vehicle, such as mass, wheelbase inertia, etc.

[0063] 2. According to the lateral and longitudinal coupling dynamics model of traditional heavy-loaded vehicles, the present invention proposes a data-driven coupling dynamics linear model based on the Koopman operator to realize the prediction of the dynamic state of heavy-loaded vehicles. It does not require the configuration parameters of the heavy-loaded vehicles, such as mass, wheelbase inertia, etc., and uses a linear model to characterize the nonlinear characteristics of the heavy-loaded vehicles, avoiding nonlinear modeling of the heavy-loaded vehicle system and facilitating subsequent controller design.

[0064] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 It is a schematic diagram of the process of the method of the present invention;

[0066] Figure 2 It is a schematic diagram of the reconstruction effect of the lateral velocity model based on the Koopman operator in the present invention on the lateral velocity;

[0067] Figure 3 It is a schematic diagram of the prediction effect of the heavy-load vehicle dynamics model on the vehicle dynamics state in the present invention. DETAILED DESCRIPTION

[0068] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0069] Embodiment 1:

[0070] See also Figure 1-Figure 3 The present invention provides a technical solution: a method for predicting the dynamic state of an electric heavy-duty vehicle, comprising the following steps:

[0071] S1. Receive heavy-load vehicle driving status information data, wherein the heavy-load vehicle driving status information data includes longitudinal acceleration, lateral acceleration, yaw angular velocity, wheel speed, wheel angle;

[0072] It should be noted that the longitudinal acceleration, lateral acceleration, yaw rate, wheel speed, and wheel angle in the driving status information data of the heavy-loaded vehicle are collected by the acceleration sensor, yaw rate sensor, speedometer, and angle sensor installed on the vehicle itself;

[0073] S2. A lateral velocity model based on the Koopman operator is constructed according to the driving state information data to predict the lateral velocity of the heavy-loaded vehicle, as follows:

[0074] (S21) Considering that the lateral velocity is nonlinearly related to the lateral acceleration, yaw rate, wheel angle, wheel motion state and other parameters, that is:

[0075] v yk =g(a xk ,a yk ,γ k ,w k ,δ k )

[0076] This embodiment proposes to use the Koopman operator to learn the lateral velocity of the heavy-loaded vehicle;

[0077] (S22) Record the state vector x vy =[a xk ,ayk ,γ k ,w k ], the input vector is u vyk =δ f , define the state dimension-raising function of the Koopman operator for lateral velocity as σ k (x vyk ,u vyk ), its dynamic evolution is as follows:

[0078] σ k (x vyk+1 ,u vyk+1 )=σ k (g(x vyk ,u vyk ))=K vy σ k (x vyk ,u vyk )

[0079] Assume that the Koopman operator is a set of matrices (A vy ,B vy ,C vy ,D vy ) to achieve, let p k =σ k (x vyk ,u vyk ), and design the following linear dynamics model:

[0080] p k+1 =A vy p k +B vy u vyk

[0081] v yk =C vy p k +D vy u vyk

[0082] In the formula, p k is the state of the kth step after dimensionality increase, p k+1 is the state of the k+1th step after dimensionality upgrade;

[0083] (S23) The above design can be understood as designing a high-dimensional state p through the Koopman operator linear model. k The dynamic evolution of vyk Under the action of vy A series of signals contained in can realize the above matrix (A vy ,B vy ,Cvy ,D vy ) to obtain the prediction of lateral velocity;

[0084] First, collect the signals from time 0 to time M and define them as follows:

[0085] P M =[p 0 p 1 …p M-1 ]

[0086] P M+1 =[p 1 p 2 …p M ]

[0087] U vyM =[u vy0 u vy1 …u vyM-1 ]

[0088] V yM =[v y0 v y1 …v yM-1 ]

[0089] Where P M is the vector matrix of states from 0 to M-1, P M+1 is the vector matrix of states from 1 to M; U vyM V is the vector matrix from 0 to M-1 for the steering input; yM is the vector matrix of lateral velocity from 0 to M-1; M is the data length;

[0090] The Koopman linear model of lateral velocity can be expanded into the following form through data:

[0091] P M+1 =A vy P M +B vy U vyM

[0092] V yM =C vy P M +D vy U vyM

[0093] By constructing an optimization problem and using the least squares method, the system matrix group (A vy ,B vy ,C vy ,D vy ):

[0094]

[0095] In the formula ‖·‖ F represents the Frobenius norm of the corresponding matrix;

[0096] By solving the above optimization problem, the prediction of lateral velocity can be achieved;

[0097] S3. Based on the nonlinear dynamics model and the predicted lateral velocity, a linear heavy-load vehicle dynamics model based on the Koopman operator is constructed to predict the dynamic state of the heavy-load vehicle, as follows:

[0098] (S31) According to the vehicle motion law and Newton's mechanics, the lateral and longitudinal coupling dynamics model of a traditional heavy-duty vehicle can be described by the following formula:

[0099]

[0100] Where m is the mass of the vehicle, in kg; v x is the longitudinal velocity of the vehicle in m / s; v y is the lateral velocity of the vehicle, in m / s; γ is the yaw rate of the vehicle, in rad / s; δ f is the front wheel turning angle, in rad; I z is the yaw inertia, unit: kg·m 2 ; L f is the distance from the center of mass of the vehicle to the front axle, in meters; L r is the distance from the vehicle's center of mass to the rear axle, in meters; Refers to the longitudinal tire force of the front wheel, unit N; Refers to the longitudinal tire force of the rear wheel, unit N; Refers to the lateral tire force of the front wheel, unit N; Refers to the lateral tire force of the rear wheel, unit N;

[0101] (S32) Select the state vector as x k =[v x ,v y ,γ k ], the input vector is u k =δ f , then the above vehicle coupling dynamics can be simplified to a general form of nonlinear dynamic model, expressed as follows:

[0102] x k+1 =f(x k ,u k )

[0103] For the above nonlinear heavy-duty vehicle system, the Koopman operator is used to implement data-driven linear vehicle dynamics modeling, and the Koopman operator state dimension-raising function of heavy-duty vehicle dynamics is defined as φ k (x k ,u k ), its dynamic evolution is as follows:

[0104] φ k (x k+1 ,u k+1 )=φ k (f(x k ,u k ))=K vd φ k (x k ,u k )

[0105] Where K vd is the Koopman operator. Note that the original state matrix x k The lateral velocity is included, which is usually an unmeasurable signal. Here, this parameter is obtained based on the lateral velocity model of the Koopman operator;

[0106] (S33) Let z k =φ k (x k ,u k ), and assume that the Koopman operator K vd is a set of matrices (A vd ,B vd ,C vd ,D vd ) is realized, the above formula can be transformed into the general form of linear control system:

[0107] z k+1 =A vd z k +B vd u vdk

[0108] y k =C vd z k +D vd u vdk

[0109] Where z k ∈R q is the dimension-increased state vector of the original linear control system, (A vd ,B vd ,C vd ,D vd ) is the system matrix determined by the Koopman operator;

[0110] (S34) By collecting the status data x of the system k With input data u k , combined with the definition of the dimension-raising function, and define the following data matrix:

[0111] Z k =[z 0 z 1 …z N-1 ]

[0112] Z k+1 =[z 1 z 2 …z N ]

[0113] U k =[u 0 u 1 …u N-1 ]

[0114] Y k =[y 0 y 1 …y N-1 ]

[0115] Where N is the selected data length or data step length, Z k , Z k+1 , U k With Y k The data segments corresponding to the 0th to N-1th step of the dimension-upgrading state, the 1st to Nth step of the dimension-upgrading state, the 0th to N-1th step of the steering input, and the 0th to N-1th step of the output vector are noted. k With Z k+1 Only one time step is missing;

[0116] (S35) Using the above linear system relationship, the relationship between the data is as follows:

[0117] Z k+1 =A vd Z k +B vd U k

[0118] Y k =C vd Z k +D vd U k

[0119] The system matrix group (A vd ,B vd ,C vd ,D vd )

[0120]

[0121] Therefore, through the data of the nonlinear electric heavy-duty vehicle system, the Koopman operator and the above-mentioned optimization algorithm, a linear heavy-duty vehicle dynamics model after dimensionality increase can be obtained.

[0122] Next, the technical solution of this embodiment is further verified in conjunction with specific embodiments:

[0123] To verify the double Koopman operator proposed in this embodiment, firstly, the Koopman operator for lateral velocity is constructed and optimized by using the lateral and longitudinal acceleration, yaw angular velocity, wheel speed and other signals of the heavy-loaded vehicle. Corresponding to the content of step S2, the final Koopman operator reconstructs the lateral velocity as shown in the following figure: Figure 2 As shown, it can be seen that on the training data set, the operator has a good fitting effect on the lateral velocity;

[0124] Based on the lateral velocity obtained by the Koopman operator, the heavy-duty vehicle dynamics modeling mentioned in S3 is carried out. The effects of the key dynamic states of the heavy-duty vehicle are as follows: Figure 3 As shown, it can also be seen that the three states in the heavy-load vehicle dynamics based on the double Koopman operator are close to the true values, verifying the good modeling effect of the data-driven algorithm.

[0125] In summary, the present invention proposes a dynamic coupling modeling method for electric heavy-duty vehicles based on a double Koopman operator. Firstly, in view of the problem that the lateral velocity of electric heavy-duty vehicles is usually unmeasurable, a linear prediction model based on easily measurable parameters such as longitudinal acceleration, lateral acceleration, yaw angular velocity, and tire speed is proposed by using the Koopman operator; secondly, according to the lateral and longitudinal coupled dynamic model of traditional heavy-duty vehicles, a data-driven coupled dynamic linear model based on the Koopman operator is proposed to realize the prediction of the dynamic state of the heavy-duty vehicle. The above-mentioned cascade Koopman operator realizes the prediction of the lateral and longitudinal coupled dynamic motion state of the heavy-duty vehicle without a lateral velocity sensor.

[0126] Embodiment 2:

[0127] This embodiment provides a system for predicting the dynamic state of an electric heavy-duty vehicle, which is used to implement the method for predicting the dynamic state of an electric heavy-duty vehicle described in the first embodiment, including:

[0128] A receiving module is used to receive and pre-process the driving state information data of the heavy-loaded vehicle, wherein the driving state information data of the heavy-loaded vehicle includes longitudinal acceleration, lateral acceleration, yaw angular velocity, wheel speed, and wheel angle;

[0129] The first construction module is used to construct a lateral velocity model based on the Koopman operator according to the driving state information data, so as to predict the lateral velocity of the heavy-loaded vehicle;

[0130] The second construction module is used to construct a linear heavy-load vehicle dynamics model based on the Koopman operator based on the nonlinear dynamics model and the predicted lateral velocity, so as to realize the prediction of the dynamic state of the heavy-load vehicle.

[0131] Specifically, the above-mentioned receiving module, the first building module and the second building module can be embedded in a computer processing system. The computer calls the above-mentioned modules to complete the task of predicting the dynamic state of the heavy-loaded vehicle based on the above-mentioned electric heavy-loaded vehicle dynamic state prediction method; the above-mentioned receiving module, the first building module and the second building module can perform operations according to the specific steps given in the electric heavy-loaded vehicle dynamic state prediction method.

[0132] It should be noted that it should be understood that the division of the various modules of the above system is only the division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated, and these modules can all be implemented in the form of software calling through processing elements; they can also be all implemented in the form of hardware; some modules can also be implemented in the form of software calling through processing elements, and some modules can be implemented in the form of hardware. For example, the receiving module can be a separately established processing element, or it can be integrated in a certain chip of the above device. In addition, it can also be stored in the memory of the above device in the form of program code, and a certain processing element of the above device calls and executes the functions of the above signal processing module. The implementation of other modules is similar. In addition, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module above can be completed by the hardware integrated logic circuit in the processor element or the instructions in the form of software.

[0133] For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASIC), or one or more digital singnal processors (DSP), or one or more field programmable gate arrays (FPGA). For another example, when a module is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processors that can call program code. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0134] Embodiment three:

[0135] The present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The memory stores a computer program capable of running on the processor, and when the processor loads and executes the computer program, the above-mentioned method for predicting the dynamic state of an electric heavy-load vehicle is adopted.

[0136] It should be noted that the terminal device can be a computer device such as a desktop computer, a laptop computer or a cloud server, and the terminal device includes but is not limited to a processor and a memory. For example, the terminal device can also include input and output devices, network access devices and buses.

[0137] Furthermore, the processor may adopt a central processing unit (CPU). Of course, according to actual usage, other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. may also be adopted. The general-purpose processor may adopt a microprocessor or any conventional processor, etc., and the present application does not impose any restrictions on this.

[0138] Embodiment 4:

[0139] The present invention provides a storage medium containing computer executable instructions, and the computer executable instructions are used to execute the above-mentioned electric heavy-duty vehicle dynamic state prediction method when executed by a computer processor.

[0140] Among them, the computer program can be stored in a computer-readable medium, the computer program includes computer program code, the computer program code can be in the form of source code, object code, executable file or certain middleware, etc. The computer-readable medium includes any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the computer-readable medium includes but is not limited to the above-mentioned components.

[0141] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0142] For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. When an element is referred to as being "assembled on", "installed on", "fixed on" or "set on" another element, it can be directly on the other element or there can also be a centered element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be a centered element at the same time. The terms "vertical", "horizontal", "up", "down", "left", "right" and similar expressions used herein are only for illustrative purposes and are not intended to be the only implementation method.

[0143] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0144] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

Claims

1. A method for predicting the dynamic state of an electric heavy-duty vehicle, characterized in that: The following steps are involved: Receiving heavy-load vehicle driving state information data, wherein the heavy-load vehicle driving state information data includes longitudinal acceleration, lateral acceleration, yaw angular velocity, wheel speed, and wheel angle; A lateral velocity model based on the Koopman operator is constructed according to the driving status information data to predict the lateral velocity of the heavy-loaded vehicle; Based on the nonlinear dynamic model and the predicted lateral velocity, a linear heavy-load vehicle dynamics model based on the Koopman operator is constructed to predict the dynamic state of the heavy-load vehicle.

2. The method for predicting the dynamic state of an electric heavy-duty vehicle according to claim 1, characterized in that: The longitudinal acceleration, lateral acceleration, yaw angular velocity, wheel speed, and wheel angle in the heavy-load vehicle driving status information data are collected by the acceleration sensor, yaw angular velocity sensor, speedometer, and angle sensor installed on the vehicle itself.

3. The method for predicting the dynamic state of an electric heavy-duty vehicle according to claim 2, characterized in that: According to the driving status information data, a lateral velocity model based on the Koopman operator is constructed to predict the lateral velocity of the heavy-loaded vehicle, as follows: (31) Let the state vector x vy =[a xk ,a yk ,γ k ,w k ], the input vector is u vyk =δ f , define the state dimension-raising function of the Koopman operator for lateral velocity as σ k (x vyk ,u vyk ), its dynamic evolution is as follows: s k (x vyk+1 ,u vyk+1 )=s k (g(x vyk ,u vyk ))=K vy s k (x vyk ,u vyk ); (32) Assume that the Koopman operator is a set of matrices (A vy ,B vy ,C vy ,D vy ) to achieve, let p k =σ k (x vyk ,u vyk ), and design the following linear dynamics model: p k+1 =A vy p k +B vy u vyk v yk =C vy p k +D vy u vyk In the formula, p k is the state of the kth step after dimensionality increase, p k+1 is the state of the k+1th step after dimensionality upgrade; (33) Collect the signals from time 0 to time M and define them as follows: P M =[p0 p1…p M-1 ] P M+1 =[p1 p2…p M ] IN vyM =[in vy0 in vy1 ... in vyM-1 ] V yM =[v y0 v y1 …v yM-1 ] Where P M is the vector matrix of states from 0 to M-1, P M+1 is the vector matrix of states from 1 to M; U vyM V is the vector matrix from 0 to M-1 for the steering input; yM is the vector matrix of lateral velocity from 0 to M-1; M is the data length; The Koopman linear model of lateral velocity can be expanded into the following form through data: P M+1 =A vy P M +B vy U vyM V yM =C vy P M +D vy U vyM The system matrix group (A vy , B vy , C vy , D vy ): In the formula ‖·‖ F Represents the Frobenius norm of the corresponding matrix; by solving the above formula, the lateral velocity can be predicted.

4. The method for predicting the dynamic state of an electric heavy-duty vehicle according to claim 3, characterized in that: The construction of the nonlinear dynamic model is as follows: (41) According to the laws of vehicle motion and Newtonian mechanics, the lateral and longitudinal coupling dynamics model of a traditional heavy-duty vehicle is described by the following equation: Where m is the mass of the vehicle, in kg; v x is the longitudinal velocity of the vehicle in m / s; v y is the lateral velocity of the vehicle, in m / s; γ is the yaw rate of the vehicle, in rad / s; δ f is the front wheel turning angle, in rad; I z is the yaw inertia, unit: kg·m 2 ; L f is the distance from the center of mass of the vehicle to the front axle, in meters; L r is the distance from the vehicle's center of mass to the rear axle, in meters; Refers to the longitudinal tire force of the front wheel, unit N; Refers to the longitudinal tire force of the rear wheel, unit N; Refers to the lateral tire force of the front wheel, unit N; Refers to the lateral tire force of the rear wheel, unit N; (42) Select the state vector as x k =[v x ,v y ,γ k ], the input vector is u k =δ f , then the vehicle coupling dynamics model in (41) can be simplified to a general nonlinear dynamics model, expressed as follows: x k+1 =f(x k ,u k ) In the formula, x k is the state vector, u k is the input vector.

5. The method for predicting the dynamic state of an electric heavy-duty vehicle according to claim 4, characterized in that: Based on the nonlinear dynamics model and the predicted lateral velocity, a linear heavy-load vehicle dynamics model based on the Koopman operator is constructed to predict the dynamic state of the heavy-load vehicle, as follows: (51) The Koopman operator state dimension-raising function of heavy-load vehicle dynamics is defined as φ k (x k ,u k ), its dynamic evolution is as follows: φ k (x k+1 ,u k+1 )=φ k (f(x k ,u k ))=K vd φ k (x k ,u k ) Where K vd is the Koopman operator; the state matrix x k Contains lateral velocity, which is obtained by the lateral velocity model based on the Koopman operator; (52) Let z k =φ k (x k ,u k ), and assume that the Koopman operator K vd is a set of matrices (A vd ,B vd ,C vd ,D vd ) is realized, the above formula can be transformed into the general form of linear control system: z k+1 =A vd z k +B vd u vdk y k =C vd z k +D vd u vdk Where z k ∈R q is the dimension-increased state vector of the original linear control system, (A vd ,B vd ,C vd ,D vd ) is the system matrix determined by the Koopman operator; (53) By collecting the state data x of the linear control system k With input data u k , combined with the definition of the dimension-raising function, and define the following data matrix: WITH k =[z0 z1…z N-1 ] WITH k+1 =[z1 z2…z N ] IN k =[u0 u1…u N-1 ] Y k =[y0 y1…y N-1 ] Where N is the selected data length or data step length, Z k , Z k+1 , U k With Y k The data segments corresponding to the 0th to N-1th step of the dimension-upgrading state, the 1st to Nth step of the dimension-upgrading state, the 0th to N-1th step of the steering input, and the 0th to N-1th step of the output vector are respectively, where Z k With Z k+1 Only one time step is missing; (54) Using the above linear control system relationship, the relationship between the data is as follows: Z k+1 =A vd Z k +B vd U k Y k =C vd Z k +D vd U k The system matrix group (A vd ,B vd ,C vd ,D vd ) Thus, the linear heavy-load vehicle dynamics model after dimensionality increase is obtained.

6. A system for predicting the dynamic state of an electric heavy-duty vehicle, used to implement the method for predicting the dynamic state of an electric heavy-duty vehicle according to any one of claims 1 to 5, characterized in that: include: A receiving module is used to receive and pre-process the driving state information data of the heavy-loaded vehicle, wherein the driving state information data of the heavy-loaded vehicle includes longitudinal acceleration, lateral acceleration, yaw angular velocity, wheel speed, and wheel angle; The first construction module is used to construct a lateral velocity model based on the Koopman operator according to the driving state information data, so as to predict the lateral velocity of the heavy-loaded vehicle; The second construction module is used to construct a linear heavy-load vehicle dynamics model based on the Koopman operator based on the nonlinear dynamics model and the predicted lateral velocity, so as to realize the prediction of the dynamic state of the heavy-load vehicle.

7. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: The memory stores a computer program that can be run on the processor. When the processor loads and executes the computer program, the method for predicting the dynamic state of an electric heavy-duty vehicle according to any one of claims 1 to 5 is adopted.

8. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions are used to perform the electric heavy-duty vehicle dynamic state prediction method according to any one of claims 1 to 5 when executed by a computer processor.

Images