Heavy vehicle dynamics state estimation method, system, equipment and medium

Abstract

The invention provides a heavy vehicle dynamic state estimation method, system and device and a medium, and belongs to the technical field of vehicle control, and the method comprises the steps: collecting a dynamic signal data flow of a heavy vehicle, dividing the dynamic signal data flow into a plurality of data collection intervals through a sliding window, taking the data acquisition interval closest to the current moment as a current data interval, and obtaining a dynamic signal in the current data interval; constructing a dynamic state estimation model based on a second-order Volterra series on the basis of the obtained vehicle front wheel rotation angle, the vehicle yaw velocity and the vehicle lateral velocity; solving the optimal coefficient matrix of the constructed dynamic state estimation model; and storing the solved optimal coefficient matrix into a nonvolatile memory of a vehicle control unit, and establishing a mapping relationship between a vehicle front wheel steering angle input sequence and a vehicle yaw velocity and a mapping relationship between the vehicle front wheel steering angle input sequence and a vehicle lateral velocity for estimating a dynamic state in real time.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of vehicle control, and particularly relates to a heavy vehicle dynamics state estimation method, system, device and medium. BACKGROUND

[0002] Real-time estimation of heavy vehicle dynamics state plays an important role in intelligent transportation and vehicle control. With the development of logistics industry and autonomous driving technology, there is an increasing demand for real-time and accurate perception of vehicle dynamics behavior. Heavy vehicles have complex dynamics characteristics due to their large load and high center of mass. Accurate estimation of yaw rate and lateral velocity is crucial to improve vehicle stability and safety. The industry is constantly seeking more efficient and reliable estimation methods to adapt to complex working conditions.

[0003] In the prior art, researchers have adopted various methods to solve the problem of vehicle dynamics state estimation. For example, the estimation method based on linear single-track model simplifies the vehicle dynamics equation to realize state prediction; nonlinear methods such as Kalman filter series algorithms are used to handle system nonlinearities; in addition, data-driven methods such as machine learning models learn the dynamics mapping relationship from historical data to try to improve estimation accuracy.

[0004] However, traditional linear models are difficult to capture the strong nonlinear dynamics characteristics of heavy vehicles, resulting in increased estimation error under high-speed steering and other working conditions; the parameter identification process of nonlinear models is complex and computationally expensive, which cannot meet the real-time requirements; and black box models such as deep learning lack physical interpretability, which is not conducive to the deployment of vehicle-mounted systems and functional safety verification.

[0005] Therefore, the present application provides a heavy vehicle dynamics state estimation method, system, device and medium. SUMMARY

[0006] The present application provides a heavy vehicle dynamics state estimation method, system, device and medium to at least solve the problems of insufficient estimation accuracy of traditional linear models due to the strong nonlinear dynamics characteristics of heavy vehicles, and high computational complexity, poor real-time performance, and weak interpretability of existing nonlinear or data-driven models, which are not conducive to real-time and reliable deployment of vehicle-mounted systems.

[0007] In a first aspect, the present application provides a heavy vehicle dynamics state estimation method, which comprises: Step S1: Collecting the dynamics signal data stream of the heavy vehicle, and dividing the dynamics signal data stream into a plurality of data collection intervals through a sliding window, and taking the data collection interval closest to the current time among the plurality of data collection intervals as the current data interval, and acquiring the dynamics signals in the current data interval, including the vehicle front wheel steering angle, the vehicle yaw rate, and the vehicle lateral velocity; Step S2: based on the obtained vehicle front wheel angle, vehicle yaw rate and vehicle lateral velocity, a dynamic state estimation model based on second-order Volterra series is constructed; Step S3: the optimal coefficient matrix of the constructed dynamic state estimation model is solved; Step S4: the optimal coefficient matrix obtained by solving is stored in the non-volatile memory of the vehicle control unit, and the mapping relationship between the vehicle front wheel angle input sequence and the vehicle yaw rate and the vehicle lateral velocity is established, which is used for real-time estimation of the dynamic state.

[0008] Further, in step S1, according to the pre-set time window length , the dynamic signal data stream is divided into several data collection intervals by using a sliding window with a sliding step of

[0009] Among them, is the initial time, is the current time; The time window length and the sliding step satisfy: ; In step S1, the data collection interval closest to the current time among the several data collection intervals of the time window length obtained by sliding window division is taken as the current data interval, that is, the time stamp of the current data interval satisfies , and the vehicle front wheel angle, vehicle yaw rate and vehicle lateral velocity in the current data interval are obtained.

[0010] Further, in step S2, based on the obtained vehicle front wheel angle, vehicle yaw rate and vehicle lateral velocity, a dynamic state estimation model based on second-order Volterra series is constructed, which specifically includes: The dynamic state estimation model of the vehicle yaw rate based on the second-order Volterra series is established, and its expression is:

[0011] Among them, is the vehicle yaw rate to be estimated; is the vehicle front wheel angle input sequence; is the zero-order kernel function; is the first-order kernel function; is the second-order kernel function; The dynamic state estimation model of the vehicle lateral velocity based on the second-order Volterra series is established, and its expression is: ​

[0012] wherein, is a lateral velocity of the vehicle to be estimated.

[0013] Further, in step S3, the optimal coefficient matrix of the constructed dynamic state estimation model is solved, specifically including: Step S301: Stacking the first-order kernel function and the second-order kernel function of the dynamic state estimation model as a coefficient matrix by column, and stacking the vehicle front wheel angle input sequence as an input matrix by row, constructing a linear system to be solved based on the input matrix, the dynamic state quantity matrix to be estimated, and the coefficient matrix; Step S302: Using the least squares method to initialize and solve the linear system to obtain an initial coefficient matrix; Step 303: Constructing an objective function and solving the objective function using a nonlinear optimization algorithm to obtain an optimal coefficient matrix.

[0014] Further, in step S301, the expression of the coefficient matrix is: ; wherein, denotes the coefficient matrix, denotes the first-order kernel function, denotes the second-order kernel function; The expression of the input matrix is:

[0015] wherein, denotes the input matrix, denotes the vehicle front wheel angle input sequence; The expression of the linear system is:

[0016] wherein, is a dynamic state quantity matrix to be estimated, that is, a measured yaw rate or lateral velocity matrix.

[0017] Further, in step S302, the least squares method is used to initialize and solve the linear system to obtain an initial coefficient matrix, and the expression is:

[0018] wherein, denotes the initial coefficient matrix.

[0019] Further, in step 303, the expression of the objective function is:

[0020] wherein, is a regularization coefficient.

[0021] In a second aspect, the embodiments of the present application also provide a system applied to the heavy vehicle dynamics state estimation method as described in the above aspects, and the system comprises: A data acquisition module acquires a dynamics signal data stream of the heavy vehicle, divides the dynamics signal data stream into a plurality of data acquisition intervals through a sliding window, takes a data acquisition interval closest to a current time as a current data interval, and acquires dynamics signals in the current data interval, including a vehicle front wheel steering angle, a vehicle yaw rate and a vehicle lateral speed. A model construction module constructs a dynamics state estimation model based on the second-order Volterra series based on the acquired vehicle front wheel steering angle, vehicle yaw rate and vehicle lateral speed. A model solving module solves an optimal coefficient matrix of the constructed dynamics state estimation model. A model application module stores the solved optimal coefficient matrix into a non-volatile memory of a vehicle control unit, establishes a mapping relationship between a vehicle front wheel steering angle input sequence and a vehicle yaw rate and a vehicle lateral speed, and is used for real-time estimation of dynamics states.

[0022] In a third aspect, an electronic device comprises a memory, a processor and a computer program stored on the memory and executable on the processor, and the processor implements the steps of the heavy vehicle dynamics state estimation method as described in the above aspects when executing the program.

[0023] In a fourth aspect, a storage medium has a computer program stored thereon, and the computer program implements the steps of the heavy vehicle dynamics state estimation method as described in the above aspects when executed by a processor.

[0024] From the above technical solutions, the present application has the following advantages: In the heavy vehicle dynamics state estimation method provided by the present application, a complete dynamics state estimation model is established, data is acquired through a sliding window, real-time updating and continuous processing of dynamics signal data are ensured, high-quality data basis is provided for subsequent modeling, on this basis, the second-order Volterra series is used to construct a dynamics state estimation model, the nonlinear characteristics and memory effect of the system are effectively captured by introducing a second-order kernel function, the estimation accuracy of the vehicle yaw rate and the vehicle lateral speed is improved, the timeliness of data acquisition is ensured, and accurate description of dynamics characteristics is realized.

[0025] In terms of model solving, firstly, the initial solution of the coefficient matrix is quickly obtained through the least square method, providing a high-quality starting point for subsequent optimization; then, the objective function containing the regularization term is constructed, and the interior point method and other nonlinear optimization algorithms are used for fine solving, which on the one hand effectively suppresses the overfitting phenomenon, and on the other hand ensures the numerical stability of the solving process. This hierarchical optimization method significantly improves the convergence speed of the algorithm while ensuring the calculation accuracy.

[0026] By storing the optimal coefficient matrix obtained by optimization into the non-volatile memory of the vehicle control unit, the mapping relationship between the steering wheel input sequence and the dynamic state quantity is established, and the vehicle control unit is provided with stable and reliable dynamic state estimation capability. BRIEF DESCRIPTION OF DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the present application, the drawings required in the description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0028] Figure 1 The flowchart of the heavy vehicle dynamics state estimation method provided by the present application. DETAILED DESCRIPTION

[0029] In order to make the application purpose, features and advantages of the present application more obvious and easy to understand, the technical solutions protected by the present application will be described clearly and completely by using specific embodiments and drawings. Obviously, the embodiments described below are only some of the embodiments of the present application, not all. Based on the embodiments in the present patent, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present patent.

[0030] The embodiments of the present application provide a heavy vehicle dynamics state estimation method, system, device and medium, which solves the technical problems that the traditional linear model estimation accuracy is insufficient due to the strong nonlinear dynamics characteristics of heavy vehicles, and the existing nonlinear or data-driven model has high computational complexity, poor real-time performance and weak interpretability, which is not conducive to real-time and reliable deployment of vehicle-mounted systems.

[0031] The technical solutions of the embodiments of the present application will be described in detail below with reference to the drawings.

[0032] Figure 1 The flowchart of the heavy vehicle dynamics state estimation method provided by the embodiments of the present application is shown in FIG. Figure 1 As shown in FIG. Step S1: Collecting the heavy vehicle dynamics signal data stream, and dividing the dynamics signal data stream into several data collection intervals through a sliding window, and taking the data collection interval closest to the current time in the several data collection intervals as the current data interval, and acquiring the dynamics signal in the current data interval, including the vehicle front wheel angle, vehicle yaw rate, vehicle lateral speed.

[0033] When the heavy vehicle is in a running state, start the vehicle motion data acquisition system based on multi-source sensor fusion, and capture the dynamics signal data stream at a fixed sampling frequency according to the vehicle motion data acquisition system.

[0034] In step S1, according to the pre-set time window length , the dynamics signal data stream is divided into several data collection intervals using a sliding window with a sliding step of

[0035] Among them, is the initial time, is the current time; It should be noted that the time window length and the sliding step satisfy: ; Among them, the dynamics signal includes the vehicle front wheel angle, the vehicle yaw rate, and the vehicle lateral speed.

[0036] It should be further noted that the data collection interval closest to the current time in the several time window length data collection intervals obtained by the sliding window is taken as the current data interval, that is, the timestamp of the current data interval satisfies , and the dynamics signal in the current data interval is acquired, including the vehicle front wheel angle, the vehicle yaw rate, and the vehicle lateral speed.

[0037] The heavy vehicle has strong nonlinear characteristics during operation, especially under the influence of tire nonlinearity, and the vehicle yaw rate and vehicle lateral speed of the heavy vehicle have strong nonlinearity. The traditional first-order Volterra series only has a linear term, and cannot capture the cross-influence between different input signals and different time points. In order to better fit the dynamics state (vehicle yaw rate, vehicle lateral speed) of the heavy vehicle, a second-order Volterra series is used to estimate the vehicle yaw rate and vehicle lateral speed of the heavy vehicle.

[0038] The second-order Volterra series is used to process nonlinear systems with memory, that is, the current output of the second-order Volterra series depends not only on the current input, but also on the past input.

[0039] ​The second-order Volterra series represents the output at time k , i.e. the series sum of input sequences .

[0040] wherein, is a to-be-estimated dynamic state quantity; is a vehicle front wheel steering angle input sequence; is a zero-order kernel function; is a first-order kernel function; is a second-order kernel function.

[0041] Step S2: based on the acquired vehicle front wheel steering angle, vehicle yaw rate and vehicle lateral velocity, a dynamic state estimation model based on the second-order Volterra series is constructed, which is used to estimate the vehicle yaw rate and vehicle lateral velocity according to the vehicle front wheel steering angle; In step S2, based on the acquired vehicle front wheel steering angle, vehicle yaw rate and vehicle lateral velocity, a dynamic state estimation model based on the second-order Volterra series is constructed, specifically including: The dynamic state estimation model of the vehicle yaw rate is established based on the second-order Volterra series, and the expression is:

[0042] wherein, is a to-be-estimated vehicle yaw rate; is a vehicle front wheel steering angle input sequence; is a zero-order kernel function; is a first-order kernel function; is a second-order kernel function; represents a memory depth; The dynamic state estimation model of the vehicle lateral velocity is established based on the second-order Volterra series, and the expression is:

[0043] wherein, is a to-be-estimated vehicle lateral velocity.

[0044] Step S3: the optimal coefficient matrix of the constructed dynamic state estimation model is solved.

[0045] ​In one example embodiment, the first-order kernel function and the second-order kernel function are stacked column by column into a coefficient matrix, the vehicle front wheel steering angle input sequence is stacked row by row into an input matrix, and a model matrix to be solved is constructed; the least square method is used to initialize the coefficient matrix of the model to be solved; an objective function of the model to be solved is constructed, a regularization term is introduced to suppress overfitting of the optimization result, and the coefficient matrix is solved by a nonlinear optimization algorithm such as the interior point method.

[0046] It should be noted that in step S3, the optimal coefficient matrix of the constructed dynamic state estimation model is solved, and specifically includes: Step S301: The first-order kernel function and the second-order kernel function of the dynamic state estimation model are stacked column by column into a coefficient matrix, and the vehicle front wheel steering angle input sequence is stacked row by row into an input matrix, and a linear system to be solved is constructed based on the input matrix, the dynamic state quantity matrix to be estimated, and the coefficient matrix.

[0047] Step S302: The linear system is initialized and solved by the least square method to obtain an initial coefficient matrix.

[0048] Step 303: An objective function is constructed, and the objective function is solved by using a nonlinear optimization algorithm to obtain an optimal coefficient matrix.

[0049] According to the embodiments of the present application, after the dynamic state estimation model is constructed, the zero-order kernel function , the first-order kernel function , and the second-order kernel function in the dynamic state estimation model need to be solved.

[0050] In order to facilitate the solution of the first-order kernel function and the second-order kernel function , the first-order kernel function and the second-order kernel function are stacked column by column into a coefficient matrix , and the expression of the coefficient matrix is: ; and the vehicle front wheel steering angle input sequence is stacked row by row into an input matrix , and the expression of the input matrix is: .

[0051] In order to start with a better initial value for subsequent model solving, improve the calculation speed, and initialize the solution of the coefficient by using the least square method.

[0052] The input matrix and the coefficient matrix constitute a linear system, and the expression of the linear system is:

[0053] wherein, is the to-be-estimated dynamic state quantity matrix, i.e., a measured yaw rate or lateral velocity matrix.

[0054] An initial coefficient matrix is obtained by using a least square method , and an expression thereof is as follows: .

[0055] An initial coefficient matrix is obtained by using a least square method , and this high-quality starting point is provided for a nonlinear optimization algorithm to start its fine search process at an initial position close to a global optimal solution. The initial estimation greatly optimizes the performance of the subsequent algorithm, effectively avoids the optimization process from falling into a local optimal solution, significantly accelerates the convergence speed, and ensures that the finally solved coefficient matrix has high precision and stability, thereby laying a solid foundation for real-time and reliable estimation of the dynamic state of a heavy vehicle.

[0056] According to another embodiment of the present application, in order to more accurately and stably solve the coefficient matrix , a nonlinear optimization algorithm is introduced to solve the coefficient matrix .

[0057] In order to solve the coefficient matrix by using the nonlinear optimization algorithm, it is necessary to first construct an objective function of the nonlinear optimization algorithm. In constructing the objective function, in order to prevent overfitting of the solution of the nonlinear optimization algorithm, resulting in that the solved coefficient matrix loses actual significance, a regularization term is introduced in the objective function to suppress overfitting of the optimization result. Finally, the constructed objective function is as follows:

[0058] wherein, is a regularization coefficient.

[0059] For the constructed objective function, a nonlinear optimization algorithm of an interior point method is used to solve the coefficient matrix , and the corresponding coefficient of the second-order Volterra series is obtained.

[0060] Finally, the dynamic state estimation model of the yaw rate and the lateral velocity established based on the second-order Volterra series is solved.

[0061] Step S4: store the solved optimal coefficient matrix into the non-volatile memory of the vehicle control unit, establish the mapping relationship between the vehicle front wheel steering angle input sequence and the vehicle yaw rate and lateral velocity, and use it for real-time estimation of the dynamic state.

[0062] Store the solved second-order Volterra series coefficient matrix in the non-volatile memory of the vehicle control unit, establish a permanent mapping relationship between the vehicle front wheel steering angle input sequence and the dynamic state quantity (yaw rate and lateral velocity), and realize power failure protection of the yaw rate and lateral velocity estimation model by solidifying the complete parameter set containing the zero-order kernel function, the linear term of the first-order kernel function, and the cross term of the second-order kernel function.

[0063] Further, as a refinement and expansion of the above embodiment, in order to fully describe the specific implementation process in this embodiment, another heavy vehicle dynamic state estimation method is provided. In step S1, according to the pre-set time window length , the dynamic signal data stream is divided into several data collection intervals using a sliding window with a sliding step size of :

[0064] wherein is the initial time, is the current time; The time window length and the sliding step size satisfy: ; In step S1, the data collection interval closest to the current time among the several data collection intervals divided by the sliding window is taken as the current data interval, i.e. the time stamp of the current data interval satisfies , and the vehicle front wheel steering angle, vehicle yaw rate, and vehicle lateral velocity in the current data interval are obtained.

[0065] It should be further noted that in step S2, based on the obtained vehicle front wheel steering angle, vehicle yaw rate, and vehicle lateral velocity, a dynamic state estimation model based on second-order Volterra series is constructed, which specifically includes: The dynamic state estimation model of the vehicle yaw rate based on the second-order Volterra series is established, and its expression is:

[0066] wherein is the vehicle yaw rate to be estimated; is the vehicle front wheel steering angle input sequence; is the zero-order kernel function; is the first-order kernel function; is a second-order kernel function; A dynamic state estimation model of the lateral velocity of the vehicle is established based on a second-order Volterra series, and an expression of the dynamic state estimation model is:

[0067] wherein, is the lateral velocity of the vehicle to be estimated.

[0068] On the basis of the above-mentioned embodiments, in order to further improve the heavy vehicle dynamic state estimation method provided by the above-mentioned embodiments, in an implementable manner, in an embodiment, in step S3, the optimal coefficient matrix of the constructed dynamic state estimation model is solved, and specifically includes: Step S301: Stacking the first-order kernel function and the second-order kernel function of the dynamic state estimation model by column to obtain a coefficient matrix, and stacking the vehicle front wheel angle input sequence by row to obtain an input matrix, and constructing a linear system to be solved based on the input matrix, the dynamic state quantity matrix to be estimated, and the coefficient matrix. In step S301, an expression of the coefficient matrix is: ; wherein, denotes the coefficient matrix, denotes the first-order kernel function, denotes the second-order kernel function; An expression of the input matrix is:

[0069] wherein, denotes the input matrix, denotes the vehicle front wheel angle input sequence; An expression of the linear system is:

[0070] wherein, is the dynamic state quantity matrix to be estimated, that is, the measured yaw rate or lateral velocity matrix.

[0071] Step S302: The linear system is initialized and solved by using the least square method to obtain an initial coefficient matrix. In step S302, the linear system is initialized and solved by using the least square method to obtain an initial coefficient matrix, and an expression of the initial coefficient matrix is:

[0072] wherein, denotes the initial coefficient matrix.

[0073] Step 303: Construct the objective function and solve the objective function using a nonlinear optimization algorithm to obtain the optimal coefficient matrix.

[0074] In step 303, the expression for the objective function is:

[0075] in, This is the regularization coefficient.

[0076] This invention also provides a real-time estimation system for the dynamic state of heavy vehicles, the system comprising: The data acquisition module collects dynamic signal data streams from heavy vehicles and divides the dynamic signal data streams into several data acquisition intervals through a sliding window. The data acquisition interval closest to the current moment is taken as the current data interval, and the dynamic signals within the current data interval are acquired, including the vehicle's front wheel steering angle, vehicle yaw rate, and vehicle lateral speed. The model building module constructs a dynamic state estimation model based on the acquired vehicle front wheel steering angle, vehicle yaw rate, and vehicle lateral velocity. The model solving module solves for the optimal coefficient matrix of the constructed dynamic state estimation model; The model application module stores the obtained optimal coefficient matrix into the non-volatile memory of the vehicle control unit and establishes the mapping relationship between the vehicle's front wheel steering angle input sequence and the vehicle's yaw rate and lateral velocity, respectively, for real-time estimation of dynamic state.

[0077] The above are embodiments of the real-time estimation system for the dynamic state of heavy vehicles provided in this disclosure. This real-time estimation system for the dynamic state of heavy vehicles and the methods for estimating the dynamic state of heavy vehicles in the above embodiments belong to the same inventive concept. For details not described in detail in the embodiments of the real-time estimation system for the dynamic state of heavy vehicles, please refer to the embodiments of the methods for estimating the dynamic state of heavy vehicles described above.

[0078] The heavy vehicle dynamics state estimation method provided in this application embodiment can be applied to electronic devices. Those skilled in the art will understand that the electronic device structure involved in the embodiments of this invention does not constitute a limitation on the electronic device. An electronic device may include more or fewer components than illustrated, or combine certain components, or have different component arrangements. In the embodiments of this invention, the electronic device includes, but is not limited to, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of this application described and / or claimed herein.

[0079] Electronic devices may include processors, external memory interfaces, internal memory, universal serial bus (USB) interfaces, charging management modules, power management modules, batteries, wireless communication modules, audio modules, speakers, microphones, sensor modules, buttons, cameras, displays, and SIM card interfaces, etc.

[0080] It is understood that the structures illustrated in the embodiments of this application do not constitute a specific limitation on the electronic device. In other embodiments of this application, the electronic device may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0081] A processor may include one or more processing units, such as: a central processing unit (CPU), an application processor (AP), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), a controller, memory, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural network processing unit (NPU). Different processing units may be independent devices or integrated into one or more processors.

[0082] The processor can serve as the nerve center and command center of an electronic device. The controller can generate operation control signals based on the instruction opcode and timing signals to control the fetching and execution of instructions.

[0083] The processor may also include memory for storing instructions and data. In some embodiments, the memory in the processor is a cache memory. This memory can store instructions or data that the processor has just used or that are used repeatedly. If the processor needs to use the instruction or data again, it can retrieve it directly from this memory. This avoids repeated accesses, reduces processor latency, and thus improves system efficiency.

[0084] An external storage interface (ESI) can be used to connect external memory cards, such as microSD cards, to expand the storage capacity of electronic devices. The external memory card communicates with the processor through the ESI to perform data storage functions, such as saving music and video files on the external memory card.

[0085] Internal memory can be used to store computer executable program code, which includes instructions. The processor executes various functional applications and data processing of electronic devices by running the instructions stored in internal memory. Internal memory can include a program storage area and a data storage area. Internal memory can include high-speed random access memory, and can also include non-volatile memory, such as at least one disk storage device, flash memory device, universal flash storage (UFS), etc.

[0086] Wireless communication functionality in electronic devices can be achieved through antennas, wireless communication modules, modem processors, and baseband processors.

[0087] Wireless communication modules can provide solutions for wireless communication applications in electronic devices, including wireless local area networks (WLANs) (such as wireless fidelity (Wi-Fi) networks), Bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), and infrared (IR) technologies.

[0088] Electronic devices can implement audio functions through audio modules, speakers, receivers, microphones, headphone jacks, and application processors.

[0089] Electronic devices can achieve shooting functions through ISPs, cameras, video codecs, GPUs, displays, and application processors.

[0090] Electronic devices can achieve display functions through GPUs, displays, and application processors.

[0091] A GPU is a microprocessor for image processing, connected to the display screen and application processor. GPUs are used to perform mathematical and geometric calculations for graphics rendering. A processor may include one or more GPUs, which execute program instructions to generate or modify display information.

[0092] A display screen is used to display images, videos, etc. A display screen includes a display panel.

[0093] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0094] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of devices, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0095] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces, apparatuses, or units, or they may be electrical, mechanical, or other forms of connection.

[0096] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a full understanding of embodiments of the invention. However, those skilled in the art will recognize that the technical solutions of the invention can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of the invention.

[0097] The aforementioned electronic device implements the method of heavy vehicle dynamic state estimation of this application, which includes: Step S1: Acquiring the dynamic signal data stream of the heavy vehicle, dividing the dynamic signal data stream into several data acquisition intervals through a sliding window, and taking the data acquisition interval closest to the current time as the current data interval, and acquiring the dynamic signals in the current data interval, including the vehicle's front wheel steering angle, vehicle yaw rate, and vehicle lateral velocity; Step S2: Based on the acquired vehicle's front wheel steering angle, vehicle yaw rate, and vehicle lateral velocity, constructing a dynamic state estimation model based on a second-order Volterra series; Step S3: Optimizing the constructed dynamic state estimation model... Step S4: The optimal coefficient matrix obtained by the solution is stored in the non-volatile memory of the vehicle control unit. The mapping relationship between the input sequence of the front wheel steering angle of the vehicle and the yaw rate and the lateral rate of the vehicle is established for real-time estimation of the dynamic state. This application adopts a technical solution that combines sliding window data acquisition with second-order Volterra series. Under the premise of ensuring real-time performance, it effectively captures the strong nonlinear dynamic characteristics of heavy vehicles. At the same time, it significantly reduces the computational complexity through matrix modeling and hierarchical optimization strategy. Finally, it realizes high-precision and interpretable dynamic state estimation, and solves the core contradiction of insufficient accuracy of traditional linear models and difficulty in vehicle deployment of complex nonlinear models.

[0098] The storage medium provided in this application stores a program product capable of implementing a method for estimating the dynamic state of heavy vehicles.

[0099] The method for estimating the dynamic state of heavy vehicles includes: acquiring dynamic signal data streams of heavy vehicles, dividing the dynamic signal data streams into several data acquisition intervals using a sliding window, and selecting the data acquisition interval closest to the current moment as the current data interval; acquiring the dynamic signals within the current data interval, including the vehicle's front wheel steering angle, vehicle yaw rate, and vehicle lateral velocity; constructing a dynamic state estimation model based on a second-order Volterra series based on the acquired vehicle's front wheel steering angle, vehicle yaw rate, and vehicle lateral velocity; solving for the optimal coefficient matrix of the constructed dynamic state estimation model; storing the obtained optimal coefficient matrix in the non-volatile memory of the vehicle control unit; and establishing a mapping relationship between the vehicle's front wheel steering angle input sequence and the vehicle yaw rate and vehicle lateral velocity, respectively, for real-time estimation of the dynamic state.

[0100] This invention establishes a complete dynamic state estimation model and collects data through a sliding window, ensuring real-time updates and continuous processing of dynamic signal data, providing a high-quality data foundation for subsequent modeling. On this basis, a second-order Volterra series is used to construct the dynamic state estimation model. By introducing a second-order kernel function, the nonlinear characteristics and memory effect of the system are effectively captured, improving the estimation accuracy of vehicle yaw rate and vehicle lateral velocity. This ensures both the timeliness of data acquisition and the accurate description of dynamic characteristics.

[0101] In terms of model solving, the initial solution of the coefficient matrix is ​​first obtained quickly by the least squares method, providing a high-quality starting point for subsequent optimization. Then, an objective function containing regularization terms is constructed, and nonlinear optimization algorithms such as the interior point method are used for fine solution. This effectively suppresses overfitting and ensures the numerical stability of the solution process. This hierarchical optimization method significantly improves the convergence speed of the algorithm while ensuring computational accuracy.

[0102] By storing the optimized coefficient matrix into the non-volatile memory of the vehicle control unit, a mapping relationship between the steering wheel input sequence and dynamic state variables is established, providing the vehicle control unit with a stable and reliable dynamic state estimation capability.

[0103] In some possible implementations, the heavy vehicle dynamic state estimation method of this disclosure can be implemented as a program product comprising program code that, when run on a terminal device, causes the terminal device to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure.

[0104] The storage medium disclosed herein may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0105] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0106] Any changes, modifications, substitutions, and variations made to the embodiments without departing from the principles and spirit of the present invention still fall within the protection scope of the present invention.

Claims

1. A method of heavy vehicle dynamics state estimation, characterized by, The method comprises: Step S1: collecting a heavy vehicle dynamics signal data stream, and dividing the dynamics signal data stream into a plurality of data collection intervals through a sliding window, and taking a data collection interval closest to a current time as a current data interval, and acquiring dynamics signals in the current data interval, including vehicle front wheel steering angle, vehicle yaw rate, and vehicle lateral speed; Step S2: based on the acquired vehicle front wheel steering angle, vehicle yaw rate, and vehicle lateral speed, constructing a dynamics state estimation model based on a second-order Volterra series; Step S3: solving an optimal coefficient matrix of the constructed dynamics state estimation model; Step S4: storing the solved optimal coefficient matrix in a non-volatile memory of a vehicle control unit, and establishing a mapping relationship between a vehicle front wheel steering angle input sequence and vehicle yaw rate and vehicle lateral speed, for real-time estimation of dynamics states.

2. The method of claim 1, wherein, In step S1, according to a pre-set time window length , a sliding window with a sliding step of is used to divide the kinetic signal data stream into a plurality of data collection intervals: wherein, is the initial time, is the current time; The time window length and the sliding step length satisfy: ； In step S1, the data collection interval closest to the current time among the data collection intervals of the lengths of the time windows divided by the sliding window is taken as the current data interval, that is, the timestamp of the current data interval satisfies , the front wheel steering angle, the vehicle yaw rate and the vehicle lateral speed in the current data interval are obtained.

3. The method of claim 2, wherein, In step S2, based on the acquired vehicle front wheel steering angle, vehicle yaw rate, and vehicle lateral speed, a dynamics state estimation model based on a second-order Volterra series is constructed, specifically including: A dynamics state estimation model of vehicle yaw rate is established based on a second-order Volterra series, and its expression is: wherein is the vehicle yaw rate to be estimated; is the vehicle front wheel steering angle input sequence; is the zeroth order kernel function; is the first order kernel function; is the second order kernel function; A dynamics state estimation model of vehicle lateral speed is established based on a second-order Volterra series, and its expression is: wherein, is the vehicle lateral velocity to be estimated.

4. The method of claim 3, wherein, In step S3, the optimal coefficient matrix of the constructed dynamics state estimation model is solved, specifically including: In step S301, the first-order kernel function and the second-order kernel function of the dynamics state estimation model are stacked by column to form a coefficient matrix, and the vehicle front wheel steering angle input sequence is stacked by row to form an input matrix, and a linear system to be solved is constructed based on the input matrix, the dynamics state quantity matrix to be estimated, and the coefficient matrix; In step S302, the linear system is initialized and solved by using the least square method, to obtain an initial coefficient matrix; In step 303, a target function is constructed, and the target function is solved by using a nonlinear optimization algorithm, to obtain an optimal coefficient matrix.

5. The method of claim 4, wherein, In step S301, the expression of the coefficient matrix is: ； wherein denotes a coefficient matrix, denotes a first order kernel function, denotes a second order kernel function; The expression of the input matrix is: wherein, denotes the input matrix, denotes the vehicle front wheel angle input sequence; The expression of the linear system is: wherein, is the matrix of the state variables to be estimated, i.e. the matrix of the measured yaw rate or lateral velocity.

6. The method of claim 5, wherein, In step S302, the linear system is initialized and solved by using the least square method, to obtain an initial coefficient matrix, and its expression is: wherein denotes the initial coefficient matrix.

7. The method of claim 6, wherein, In step 303, the expression of the target function is: wherein is a regularization coefficient.

8. A system for applying to the method for estimating the dynamic state of a heavy vehicle according to any one of claims 1 to 7, characterized in that, The system comprises: A data collection module collects a heavy vehicle dynamics signal data stream, and divides the dynamics signal data stream into a plurality of data collection intervals through a sliding window, and takes a data collection interval closest to a current time as a current data interval, and acquires dynamics signals in the current data interval, including vehicle front wheel steering angle, vehicle yaw rate, and vehicle lateral speed; A model construction module constructs a dynamics state estimation model based on a second-order Volterra series based on the acquired vehicle front wheel steering angle, vehicle yaw rate, and vehicle lateral speed; The model solving module solves the optimal coefficient matrix of the constructed dynamic state estimation model; The model applying module stores the solved optimal coefficient matrix into the non-volatile memory of the vehicle control unit, and establishes the mapping relationship between the vehicle front wheel angle input sequence and the vehicle yaw rate and the vehicle lateral velocity, which is used for real-time estimation of the dynamic state.

9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor implements the steps of the heavy vehicle dynamic state estimation method according to any one of claims 1-7 when executing the program.

10. A storage medium having stored thereon a computer program, characterized in that The computer program implements the steps of the heavy vehicle dynamic state estimation method according to any one of claims 1-7 when executed by the processor.

Images