Side slip angle estimation method, equipment, medium and product

Abstract

The embodiment of the invention relates to the technical field of automobile control, and discloses a side slip angle estimation method and device, a medium and a product. Establishing a first state equation based on the dynamic model, and further establishing a first finite memory estimator based on dynamics; obtaining a first FME estimation value based on dynamics according to the first finite memory estimator; establishing a second state equation based on the kinematics model, and further establishing a second finite memory estimator based on kinematics; calculating a gain value based on the first FME estimate and vehicle sensor data using a first neural network; the second finite memory estimator obtains a second FME estimated value based on kinematics according to the gain value; and fusing the first FME estimation value and the second FME estimation value by using a second neural network to obtain a final fusion estimation value of the side slip angle. The technical problem of low estimation precision of the side slip angle can be at least solved.

Description

Technical Field

[0001] The present application relates to the field of automobile control technology, and in particular to a method, device, medium and product for estimating a sideslip angle of a center of mass. Background Art

[0002] The vehicle's center of mass slip angle is a key parameter for the vehicle's lateral stability control, which directly affects driving safety, especially when turning at high speed. Traditional methods for estimating the center of mass slip angle are divided into two categories: estimation methods based on kinematic models and estimation methods based on dynamic models:

[0003] The kinematic estimation method is mainly based on the direct integration method of the lateral acceleration and yaw rate sensor signals to estimate the sideslip angle of the center of mass. It has very good robustness to changes in vehicle parameters, road adhesion conditions and driving control methods. However, this estimation method cannot avoid the accumulation of integral errors, has observability problems when the vehicle is driving in a straight line, and is sensitive to sensor noise, resulting in limited accuracy.

[0004] The estimation method based on dynamic model is the main method for estimating the sideslip angle of the center of mass at present. Its basic principle is to estimate the sideslip angle of the center of mass by establishing a state observer based on the relationship between the lateral force of the tire and the sideslip angle of the tire, based on the vehicle dynamic model and tire model. However, this method relies on the dynamic model and is sensitive to vehicle parameters (mass, tire stiffness, etc.), which are easily affected by load, road conditions, etc., and the nonlinear characteristics of tires are significant at high speeds, resulting in large model errors. Summary of the invention

[0005] One object of the present application is to provide a method, device, medium and product for estimating the sideslip angle of the center of mass, at least to solve the technical problem of low accuracy in estimating the sideslip angle of the center of mass.

[0006] To achieve the above objectives, some embodiments of the present application provide the following aspects:

[0007] In a first aspect, some embodiments of the present application also provide a method for estimating a sideslip angle at a center of mass, comprising establishing a first state equation based on a dynamics model, and then establishing a first finite memory estimator based on dynamics; obtaining a first finite memory estimator (FME) based on dynamics according to the first FME; establishing a second state equation based on a kinematic model, and then establishing a second finite memory estimator based on kinematics; using a first neural network to calculate a gain value based on the first FME estimate and vehicle sensor data; using the second finite memory estimator to obtain a second FME estimate based on kinematics according to the gain value; using a second neural network to fuse the first FME estimate and the second FME estimate to obtain a final fused estimate of the sideslip angle at a center of mass.

[0008] In a second aspect, some embodiments of the present application further provide an electronic device, comprising: one or more processors; and a memory storing computer program instructions, wherein the computer program instructions, when executed, cause the processor to perform the steps of the method described above.

[0009] In a third aspect, some embodiments of the present application further provide a computer-readable medium having computer program instructions stored thereon, wherein the computer program instructions can be executed by a processor to implement the method as described above.

[0010] In a fourth aspect, some embodiments of the present application further provide a computer program product, comprising a computer program / instruction, which implements the steps of the method described above when executed by a processor.

[0011] Compared with the related art, the solution provided in the embodiment of the present application improves the accuracy and robustness of the estimation by combining the advantages of the dynamic and kinematic estimation methods and using neural networks for fusion. The error of the traditional method is optimized by the neural network, and a high accuracy can still be maintained under large side slip angle conditions; it only relies on the inertial measurement unit, vehicle speed sensor and steering angle sensor, without the need for high-cost GNSS or tire sensors; it can adapt to changes in different vehicle parameters, such as load, tire pressure, etc., which is more advantageous for practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0013] Figure 1 A schematic diagram of a flow chart of a method for estimating a sideslip angle of a center of mass provided according to an embodiment of the present application;

[0014] Figure 2 A schematic diagram of the effect of a first neural network provided according to an embodiment of the present application;

[0015] Figure 3 A schematic diagram of the effect of a second neural network provided according to an embodiment of the present application;

[0016] Figure 4 A schematic diagram of a two-degree-of-freedom bicycle model provided according to an embodiment of the present application;

[0017] Figure 5 A schematic diagram of the structure of a neural network provided according to an embodiment of the present application;

[0018] Figure 6The present invention is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0020] First embodiment

[0021] The first embodiment of the present application relates to a method for estimating a center of mass sideslip angle. Figure 1 As shown, the method may include the following steps:

[0022] S101, establishing a first state equation based on a dynamics model, and then establishing a first finite memory estimator based on dynamics; obtaining a first FME estimation value based on dynamics according to the first finite memory estimator.

[0023] The bicycle model is used to establish the vehicle's dynamic equations, including the lateral force balance equation and the yaw moment balance equation, and to construct a state space model. The finite memory estimator (FME) is used to calculate the current first FME estimate, i.e., the dynamic center of mass sideslip angle estimate, based on the vehicle's dynamic parameters (such as tire cornering stiffness, vehicle speed, yaw rate, etc.), and is continuously updated to adapt to different vehicle states.

[0024] The estimation method based on the dynamic model can accurately describe the vehicle's motion characteristics and has high accuracy during steady-state driving (such as cornering at a constant speed). Since the dynamic estimation takes into account the relationship between the tire cornering force and the vehicle speed and yaw rate, the estimation result is more reliable in high-speed cornering or intense driving scenarios. However, the dynamic estimation is greatly affected by changes in vehicle parameters (such as load, tire pressure, and road friction coefficient). When the vehicle parameters change suddenly, the estimation error may increase.

[0025] S102, establishing a second state equation based on the kinematic model, and then establishing a second kinematic-based finite memory estimator;

[0026] The state equation is established using a kinematic model, which derives the estimated value of the center of mass slip angle based on the longitudinal acceleration, lateral acceleration and yaw rate of the vehicle. The finite memory estimator (FME) is used to calculate the current kinematic center of mass slip angle estimate in combination with historical state information. The kinematic estimation relies on the vehicle speed, acceleration and other data measured by the IMU (inertial measurement unit), avoiding the reliance on complex dynamic parameters.

[0027] The kinematic method mainly relies on sensor data (such as IMU measurements), is not affected by changes in parameters such as vehicle mass and tire stiffness, and is applicable to different types of vehicles. Compared with dynamic estimation, kinematic estimation can still work stably under low-speed and straight road conditions. Since kinematic estimation relies on sensor data, it is easily affected by noise and drift, resulting in increased estimation errors during long-term operation.

[0028] S103, using the first neural network to calculate a gain value based on the first FME estimation value and vehicle sensor data; the second finite memory estimator obtains a kinematics-based second FME estimation value according to the gain value.

[0029] The first neural network is used to calculate the gain value, which is used to optimize the state estimate of the kinematic model to make it closer to the true center of mass sideslip angle. The input of the first neural network includes: the first FME estimate (dynamic estimate) vehicle sensor data (vehicle speed, lateral acceleration, longitudinal acceleration, yaw rate, steering angle). The first neural network calculates the optimal gain value based on the input data and uses it to adjust the state equation of the kinematic model to improve the accuracy of the kinematic estimation. The second FME estimator calculates the optimized kinematic center of mass sideslip angle estimate under the action of the new gain value.

[0030] By introducing neural network optimization, the error accumulation phenomenon of the kinematic method is reduced and the stability of long-term estimation is improved. Under different working conditions, the neural network can automatically adjust the gain value to make the kinematic estimation more accurate. Compared with directly using high-precision sensors (such as GNSS), the kinematic estimation method using neural network optimization has lower computational complexity and is suitable for real-time applications.

[0031] S104, using a second neural network to fuse the first FME estimation value and the second FME estimation value to obtain a final fused estimation value of the center of mass sideslip angle.

[0032] The second neural network is used to fuse the dynamic estimate and the optimized kinematic estimate to obtain a more accurate center of mass sideslip angle estimation result. The input of the second neural network includes the first FME estimate (dynamic estimate) and the second FME estimate (optimized kinematic estimate). The second neural network calculates the final fusion estimate through weighted summation and nonlinear mapping. This estimate combines the steady-state accuracy of the dynamic model and the robustness of the kinematic model, and can maintain high accuracy under different vehicle conditions.

[0033] The advantages of fusion dynamics and kinematics make the final estimated value closer to the true value than a single method, and can maintain a high estimation accuracy under different vehicle conditions (such as low speed / high speed, wet / dry road surface, etc.). The use of neural networks for fusion calculations reduces the computational overhead of using complex dynamic models alone for estimation, making the method more suitable for real-time control systems.

[0034] It is not difficult to find that compared with the related art, the scheme provided by the embodiment of the present application successfully integrates the dynamic and kinematic models through the FME estimation method optimized by the neural network, while maintaining high accuracy, the robustness and adaptability of the estimation are improved; the dynamic estimation provides steady-state accuracy, the kinematic estimation provides short-term dynamic adaptability, and the neural network integrates the advantages of both. Only basic sensors such as IMU and steering wheel angle sensor are used, and GNSS or tire force sensors are not required, which reduces costs. This method is particularly suitable for applications such as autonomous driving, vehicle stability control (ESP), and intelligent chassis control, and can effectively improve vehicle driving safety and comfort.

[0035] Second embodiment

[0036] The second embodiment of the present application relates to a method for estimating the sideslip angle of the center of mass. The second implementation is an improvement made on the basis of the first embodiment, and the specific improvements are:

[0037] Further, the use of the first neural network includes: the vehicle sensor data includes vehicle speed, lateral acceleration, longitudinal acceleration, yaw angular velocity and steering angle; a gain value is calculated based on the first FME estimation value and the vehicle sensor data; and the gain value in the second finite memory estimator is replaced according to the gain value, thereby obtaining the second FME estimation value.

[0038] like Figure 2 As shown, the input to neural network 1 includes the FME estimates based on dynamics And the sensor measurement value V x 、a x 、a y , r and δ. The output of neural network 1 is C nn, which is used to replace the gain C in the measurement equation of the kinematic state-space model; the kinematics-based FME uses the resulting Cnn to estimate the sideslip angle

[0039] Furthermore, the training of the first neural network includes: using a back propagation algorithm to adjust a weight parameter according to an error between the second FME estimation value and a true sideslip angle provided in the training data.

[0040] Furthermore, the training of the second neural network includes: using a back propagation algorithm to adjust a weight parameter according to an error between the center of mass sideslip angle fusion estimate and a true sideslip angle provided in training data.

[0041] like Figure 3 As shown, neural network 2 is used to fuse the estimation based on dynamics and kinematics. During the training process, neural network 2 is used to and As input, output the fused estimate Then, calculate the output of neural network 2 and the true sideslip angle β tr The error between them is calculated and neural network 2 is trained to reduce the error.

[0042] Furthermore, the dynamic model includes: deriving the dynamic model through the lateral force and yaw moment balance equations of the bicycle model.

[0043] Because the distance between the left and right wheels of a vehicle is much smaller than its turning radius, the vehicle dynamics model is usually simplified to a two-degree-of-freedom bicycle model to represent the lateral motion of the vehicle.

[0044] like Figure 4 Figure 2 shows a two-degree-of-freedom bicycle model where the X-axis is aligned with the longitudinal direction of the vehicle. x is the vehicle speed, V y is the lateral velocity, r is the yaw rate, a and b are the distances from the center of gravity to the front and rear axles, respectively, and F yf and F yr are the lateral tire forces acting on the front and rear axles of the vehicle, δ is the steering angle, and β is the sideslip angle at the center of mass.

[0045] In order to estimate the sideslip angle of the center of mass, a state estimation algorithm based on a state space model is usually used. The state space model of linear discrete time is usually expressed as:

[0046] x k =F k-1 x k-1 +B k-1 u k-1 +w k-1 ,

[0047] y k =H k x k +D k u k +v k ,

[0048] Among them, x k is the state vector at discrete time step k, y k is the measurement vector, u k is the control input, w k-1 is the process noise, v k is the measurement noise. Assume that both process and measurement noise are white Gaussian noise with zero mean. The corresponding matrix F k-1 ,B k-1 ,H k and D k , we obtain a discrete-time state-space model with appropriate dimensions. We will now use the kinematics and dynamics of a bicycle model to derive a state-space model of the above form.

[0049] The dynamic model is derived using the lateral force and yaw moment balance equations from the bicycle model and is expressed as follows:

[0050]

[0051] Ir&=aF yf -bF yr ,

[0052] Where m is the vehicle mass and I is the vehicle's moment of inertia (vertical axis). Front and rear axle lateral tire force F yf and F yr It can be expressed as follows:

[0053] F yr =-2C f α f ,

[0054] F yf =-2C r α r ,

[0055] Among them, C f and C r denote the cornering stiffness of the front and rear wheels respectively. In addition, α f and α r are the slip angles of the front and rear wheels respectively, and their expressions are as follows:

[0056]

[0057] Define the state vector, measurement vector and control input as follows:

[0058]

[0059] Where ay is the lateral acceleration, which can be measured using an inertial measurement unit (IMU). The system of equations is transformed into the following state space model:

[0060]

[0061] The state space equation is a continuous time (CT) state space model. We define the discrete time state quantity and observation quantity and control input as:

[0062]

[0063] Convert it into a discrete-time state-space model:

[0064] x k =F k-1 x k-1 +B k-1 u k-1 +w k-1 ,

[0065] y k =H k x k +D k u k +v k ,

[0066] in:

[0067]

[0068] Where T is the sampling time.

[0069] Furthermore, the kinematic model includes: deriving the kinematic model through the longitudinal acceleration and lateral acceleration of the bicycle model:

[0070]

[0071] Among them, a x and a y are the longitudinal and lateral accelerations respectively.

[0072] Rewritten in state space form:

[0073]

[0074] The observability matrix corresponding to the state equation is expressed as:

[0075]

[0076] Considering that when the vehicle is traveling in a straight line, the yaw rate r is close to zero, it is difficult to ensure observability when using a kinematic model. Therefore, the following kinematic equation is used:

[0077]

[0078] Assuming uniform motion, a x becomes zero, the above expression can be rewritten as:

[0079]

[0080] Therefore, we can calculate the Integrate to obtain β. Considering the disadvantage of error accumulation in the direct integration method, a state estimator is used to replace direct integration.

[0081] Define the state quantity, control input and observation quantity as x=β,

[0082] and the measured y = β measured

[0083] The kinematic equations are expressed in state space form as follows:

[0084]

[0085] y=x,

[0086] Transforming the above equation into discrete state space form yields:

[0087]

[0088] y k =Cx k +v k ,

[0089] In the continuous state space, the observed value is the center of mass sideslip angle. Here, the weight obtained by using the neural network multiplied by the estimated value of the center of mass sideslip angle obtained based on the dynamic model is used as the observed value to achieve the estimation of the center of mass sideslip angle.

[0090] Furthermore, the method is as follows Figure 5 As shown, the first neural network includes an input layer including 6 input nodes, an output layer including an output node, and a hidden layer including 5 nodes; the second neural network includes an input layer including 2 input nodes, an output layer including an output node, and a hidden layer including 3 nodes; the first neural network and the second neural network's hidden layer aggregate the information received by the input layer by weighted summation. The calculation formula is as follows:

[0091]

[0092] Among them, w ij is the connection weight between the ith input node and the jth output node. The aggregated input is processed by the sigmoid activation function as follows:

[0093]

[0094] Where n h is the number of hidden nodes. Finally, the output node processes the values ​​from the hidden nodes into a weighted sum to generate the final output c as follows:

[0095]

[0096] Among them, w j is the weight connected to the jth hidden node.

[0097] Furthermore, the data used to train the neural network comes from the automotive simulation software CarSim. By simulating a vehicle driving on a curve at a speed of 90-140 km / h, the data required for neural network training, including the true center of mass sideslip angle and sensor measurements, was obtained.

[0098] The mean absolute error (MAE) is used as the loss function, which is calculated as follows:

[0099]

[0100] Among them, is n d The total number of training data samples, and Y i are the predicted value (output of the neural network) and the target value of the i-th sample data. 90% of the training data is used for training and the remaining 10% is used for validation. The learning rate is set to 0.02.

[0101] It is not difficult to find that in the solution provided by the embodiment of the present application, basic sensors (IMU, steering wheel angle sensor, etc.) are used, and high-cost GNSS or tire sensors are not required, thereby reducing the cost of the system, having strong adaptability, being suitable for mass-produced vehicles, and ensuring low hardware costs. 3. The present invention has good robustness, can adapt to changes in vehicle parameters (load, tire pressure, etc.), and can still maintain high estimation accuracy under different working conditions

[0102] The step division of the above methods is only for clear description. When implemented, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this application; adding insignificant modifications to the algorithm or process or introducing insignificant designs without changing the core design of the algorithm and process are all within the scope of protection of this application.

[0103] In addition, some embodiments of the present application also provide an electronic device. The electronic device may be a digital computer in various forms, such as a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, etc. The electronic device may also be a mobile device in various forms, such as a personal digital processing, a cellular phone, a smart phone, a wearable device, and other similar computing devices.

[0104] The electronic device includes: one or more processors; and a memory storing computer program instructions, wherein when the computer program instructions are executed, the processor executes the steps of the method provided in any one or more of the above embodiments. Figure 6 An exemplary structural diagram of the electronic device is disclosed. Figure 6 As shown, the electronic device includes: one or more processors 1101, memory 1102, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some other embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Among them, the components shown in this article, their connections and relationships, and their functions are only used as examples, and are not intended to limit the implementation of the present application described and / or required herein.

[0105] The electronic device may further include: an input device 1103 and an output device 1104. The processor 1101, the memory 1102, the input device 1103 and the output device 1104 may be connected via a bus or other means. Figure 6 The example of connecting through bus is taken in the following.

[0106] The input device 1103 can receive input digital or character information, and generate key signal input related to the user settings and function control of the electronic device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick and other input devices. The output device 1104 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The display device may include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display and a plasma display. In some embodiments, the display device may be a touch screen.

[0107] To provide interaction with a user, the electronic device may be a computer. The computer has: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball), through which the user can provide input to the computer. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0108] In the embodiments of the present application, a computer program / instruction is stored on a computer-readable medium, and when the computer program / instruction is executed by a processor, the steps of the method provided by any one or more of the above embodiments are implemented. The computer-readable medium may be included in the electronic device described in the above embodiments; or it may exist independently without being assembled into the device. The above computer-readable medium carries one or more computer-readable instructions.

[0109] The memory 1102 can be used as a non-transient computer-readable storage medium, which can be used to store non-transient software programs, non-transient computer executable programs and modules. The processor 1101 executes various functional applications and data processing of the server by running the non-transient software programs, instructions and modules stored in the memory 1102, so as to implement the program instructions / modules corresponding to the method provided by any one or more embodiments in the embodiments of the present application.

[0110] The memory 1102 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 1102 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory 1102 may optionally include a memory remotely arranged relative to the processor 1101, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0111] It should be noted that the computer-readable medium described in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device.

[0112] Computer readable media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, modules of programs or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0113] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0114] In the above-described embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. For example, it can be implemented by using an application specific integrated circuit (ASIC), a general-purpose computer or any other similar hardware device. In certain embodiments, the software program of the present application can be executed by a processor to implement the above steps or functions. Similarly, the software program of the present application (including relevant data structures) can be stored in a computer-readable recording medium, for example, a RAM memory, a magnetic or optical drive or a floppy disk and similar devices. In addition, some steps or functions of the present application can be implemented by hardware, for example, as a circuit that cooperates with a processor to perform each step or function.

[0115] The computer program product provided in the embodiment of the present application includes one or more computer programs / instructions, and when the computer program / instructions are executed by the processor, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from a computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.

[0116] The flow chart or block diagram in the accompanying drawings shows the possible architecture, function and operation of the equipment, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated system for hardware that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0117] The scope of the present application is defined by the appended claims rather than the above description, and it is intended that all changes that fall within the meaning and scope of the equivalent elements of the claims are included in the present application. Any figure mark in the claims should not be regarded as limiting the claims involved. In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the device claim may also be implemented by one unit or device through software or hardware. The words "first", "second", etc. are only used to distinguish the description, and do not indicate any particular order, nor can they be understood as indicating or implying relative importance.

[0118] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily mention changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims, and the above embodiments should be regarded as exemplary and non-restrictive.

Claims

1. A method for estimating a sideslip angle of a center of mass, characterized in that: The method comprises: Establishing a first state equation based on the kinetic model, and then establishing a first finite memory estimator based on kinetics; obtaining a first FME estimation value based on kinetics according to the first finite memory estimator; Establishing a second state equation based on the kinematic model, and then establishing a second kinematic-based finite memory estimator; Using a first neural network, a gain value is calculated based on the first FME estimate and vehicle sensor data; the second finite memory estimator obtains a second kinematics-based FME estimate based on the gain value; The first FME estimation value and the second FME estimation value are fused using a second neural network to obtain a final fused estimation value of the center of mass sideslip angle.

2. The method according to claim 1, characterized in that The using the first neural network comprises: The vehicle sensor data includes vehicle speed, lateral acceleration, longitudinal acceleration, yaw rate and steering angle; calculating a gain value based on the first FME estimate and the vehicle sensor data; The gain value in the second finite memory estimator is replaced according to the gain value, thereby obtaining the second FME estimation value.

3. The method according to claim 2, characterized in that The training of the first neural network includes: Using the back-propagation algorithm, the weight parameters are adjusted according to the error between the second FME estimate and the true sideslip angle provided in the training data.

4. The method according to claim 1, characterized in that The training of the second neural network includes: A back propagation algorithm is used to adjust the weight parameter according to the error between the center of mass sideslip angle fusion estimate and the true sideslip angle provided in the training data.

5. The method according to claim 1, characterized in that The dynamic model includes: deriving the dynamic model through the lateral force and yaw moment balance equations of the bicycle model.

6. The method according to claim 1, characterized in that The kinematic model includes: deriving the kinematic model through the longitudinal acceleration and the lateral acceleration of the bicycle model.

7. The method according to any one of claims 1 to 6, characterized in that: The method comprises: The input layer of the first neural network includes 6 input nodes, the output layer includes one output node, and the hidden layer includes 5 nodes; The input layer of the second neural network includes 2 input nodes, the output layer includes one output node, and the hidden layer includes 3 nodes; The hidden layers of the first neural network and the second neural network aggregate the information received by the input layer by weighted summation.

8. An electronic device, characterized in that: The electronic device comprises: one or more processors; and A memory storing computer program instructions, which, when executed, cause the processor to perform the steps of the method as claimed in any one of claims 1 to 7.

9. A computer readable medium having a computer program / instructions stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images