Vehicle mass real-time estimation method and device

Abstract

The embodiment of the invention discloses a whole vehicle mass real-time estimation method and device, and the method comprises the steps: obtaining the vehicle speed and motor torque at the current moment, and calculating the acceleration of a vehicle in real time according to the vehicle speed at the current moment through a preset calculation method; obtaining an identification model of the whole vehicle mass according to a vehicle kinetic equation in combination with the matching relationship between the vehicle speed and the acceleration; and on the basis of combination of a forgetting factor and a gradient descent method, real-time online identification of identification parameter vectors is conducted on the identification model according to the motor torque, the vehicle speed and the acceleration, and the real-time whole vehicle mass is determined according to the identification parameter vectors. Through the above mode, the embodiment of the invention has high precision and reliability, and improves the accuracy and stability of vehicle mass estimation.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of autonomous driving vehicle control, and specifically to a method and device for real-time estimation of vehicle mass. Background Art

[0002] The vehicle mass is one of the most critical parameters in vehicle control, which determines the safety, stability and comfort of vehicle control. Especially for commercial vehicles such as buses and trucks, the need to carry people or goods causes their weight to fluctuate greatly. For conventional manned driving, the driver can intelligently judge the range of mass changes by visually observing and feeling the vehicle's response to the accelerator and brakes. In the chassis control of unmanned driving, if the intelligent judgment technology is limited and the change in the vehicle mass cannot be perceived in time, not only can a good driving experience not be achieved, but more importantly, in the case of heavy load and large inertia, there will be the risk of collision due to untimely braking and the risk of mechanical damage due to excessive braking. Therefore, real-time and accurate estimation of the vehicle mass is particularly critical for intelligent driving chassis control.

[0003] Vehicle mass estimation refers to calculating the vehicle mass through measurable signals (speed, torque) and designing a reasonable algorithm. In embedded controllers, real-time vehicle mass estimation requires low algorithm complexity, small amount of calculation, high stability, no need to store a large amount of data, and no singular values ​​during the operation. The specific difficulties in designing the vehicle mass estimation algorithm are: 1) Decoupling of rolling resistance, slope resistance, and wind resistance; 2) Filtering and denoising of input signals. Filtering will distort the original signal and cause delays, resulting in large fluctuations in the identification results; 3) The problem of calculation amount. Offline data sequence analysis can obtain stable results, but the amount of calculation is large and a large amount of data needs to be stored, which is not suitable for embedded applications. Online calculations usually use identification methods such as recursive least squares, which have a small amount of calculation, but it is difficult to obtain stable identification results.

[0004] The research on vehicle mass estimation has a long history, and a variety of estimation algorithms have emerged. These algorithms can be divided into three categories: time domain method, frequency domain method, and data model-based method. The time domain method is based on the vehicle longitudinal dynamics model. The identification model is obtained through mathematical derivation, and the vehicle mass is calculated using the least squares method or optimization method. This type of method has a simple principle and clear calculation steps. There are many theoretical and simulation studies. It can well deal with the decoupling of rolling resistance and slope resistance in difficulty 1, but there is no better method for the decoupling of wind resistance; and for difficulties 2 and 3, there is no scientific and reasonable method that can be applied to embedded occasions. The frequency domain method converts the problem of vehicle mass estimation into a frequency response problem between wheel speed and acceleration by deriving the longitudinal dynamics model, decoupling the influence of other time-varying variables, but requires high real-time and accuracy of acceleration measurement, and frequency analysis must store a large amount of data. Therefore, this type of method is more suitable for offline analysis and difficult to apply to real-time occasions. The method based on the data model needs to collect a large amount of data, train the data model, and obtain the black box relationship between the vehicle model and the input quantity. This type of method has large data storage and large amount of calculation, and is rarely used in practice. Therefore, it is necessary to seek a method for estimating the vehicle mass with stable and accurate identification results with small calculations. Summary of the invention

[0005] In view of the above problems, an embodiment of the present invention provides a method and device for real-time estimation of vehicle mass, which overcomes the above problems or at least partially solves the above problems.

[0006] According to one aspect of an embodiment of the present invention, a method for real-time estimation of vehicle mass is provided, the method comprising: obtaining a vehicle speed and a motor torque at a current moment, and calculating the acceleration of the vehicle in real time according to a preset calculation method based on the vehicle speed at the current moment; obtaining an identification model of the vehicle mass according to a vehicle dynamics equation in combination with a matching relationship between the vehicle speed and the acceleration; performing real-time online identification of an identification parameter vector of the identification model according to the motor torque, the vehicle speed and the acceleration based on a combination of a forgetting factor and a gradient descent method, and determining the real-time vehicle mass according to the identification parameter vector.

[0007] Optionally, the real-time calculation of the vehicle acceleration is performed according to a preset calculation method based on the vehicle speed at the current moment, including: determining initialization parameters; updating in real time a first state variable representing a smoothed value of the vehicle speed after filtering and a second state variable representing an approximate differential value of the vehicle speed based on the initialization parameters and the vehicle speed at the current moment by applying a preset fast convergence function; and using the updated second state variable as the calculated acceleration.

[0008] Optionally, the real-time calculation of the vehicle acceleration by applying a preset calculation method according to the vehicle speed at the current moment also includes: determining the number of delay cycles according to the initialization parameters; and determining whether the calculated vehicle acceleration corresponds to the vehicle speed at a moment before the number of delay cycles at the current moment.

[0009] Optionally, the method of obtaining the identification model of the vehicle mass according to the vehicle dynamics equation combined with the matching relationship between the vehicle speed and the acceleration includes: discretizing the vehicle dynamics equation and calculating the difference equation at two consecutive moments; constructing an initial identification model of the vehicle mass according to the difference equation; and correcting the initial identification model according to the matching relationship between the vehicle speed and the acceleration to obtain the identification model of the vehicle mass.

[0010] Optionally, constructing an initial identification model of the vehicle mass according to the differential equation includes: deforming the differential equation to obtain the following relationship:

[0011] Δf k =B(v k -v k-1 ) 2 -2B·V W (v k -v k-1 )+mΔa k

[0012]

[0013] Where Δf k is the difference in traction or braking force between time k and time k-1, v k is the vehicle speed at time k, V W is the wind speed, m is the vehicle mass; g is the acceleration due to gravity; f r is the rolling resistance coefficient; θ is the slope angle; C d is the air resistance coefficient; ρ is the air density; A p is the frontal area; Δa k is the acceleration difference between time k and time k-1; the deformed differential equation is written in matrix form to obtain the initial identification model of the vehicle mass:

[0014] The output of the recognition model represents the traction difference or braking force difference.

[0015] Optionally, the correcting the initial identification model according to the matching relationship between the vehicle speed and the acceleration to obtain the identification model of the vehicle mass includes: according to the matching relationship between the vehicle speed and the acceleration, substituting the vehicle speed at a time before the delay period number corresponding to the calculated vehicle acceleration at the current time into the initial identification model to obtain the identification model of the vehicle mass:

[0016]

[0017] Among them, k d is the number of delay cycles, For vehicle kk d The vehicle speed at the moment.

[0018] Optionally, the real-time online identification of the identification parameter vector of the identification model is performed based on the motor torque, the vehicle speed and the acceleration based on the forgetting factor combined with the gradient descent method, including: determining an initial identification parameter vector θ in the identification model, and calculating the traction force difference or braking force difference Y between time k and time k-1 based on the motor torque; determining an input vector X of the identification model based on the vehicle speed and the acceleration; if it is detected that the absolute value of the acceleration difference between time k and time k-1 is greater than a preset value, recursively performing real-time online identification on the identification model using the gradient descent method with a preset forgetting factor until the real-time vehicle mass is determined to be between the no-load mass and the fully loaded mass based on the identification parameter vector θ.

[0019] Based on the same inventive concept, a real-time estimation device for the mass of a vehicle is also provided, comprising: an acceleration calculation unit, used to obtain the vehicle speed and motor torque at the current moment, and calculate the acceleration of the vehicle in real time according to a preset calculation method based on the vehicle speed at the current moment; an identification model acquisition unit, used to obtain an identification model of the mass of the vehicle according to a vehicle dynamics equation combined with a matching relationship between the vehicle speed and the acceleration; a mass estimation unit, used to perform real-time online identification of an identification parameter vector of the identification model according to the motor torque, the vehicle speed and the acceleration based on a combination of a forgetting factor and a gradient descent method, and determine the real-time mass of the vehicle according to the identification parameter vector.

[0020] Based on the same inventive concept, an embodiment of the present invention further proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the aforementioned method when executing the program.

[0021] Based on the same inventive concept, an embodiment of the present invention further proposes a computer storage medium, in which at least one executable instruction is stored, and the executable instruction enables a processor to execute the aforementioned method.

[0022] The embodiment of the present invention obtains the vehicle speed and motor torque at the current moment, and calculates the acceleration of the vehicle in real time according to the vehicle speed at the current moment by applying a preset calculation method; obtains an identification model of the vehicle mass according to the vehicle dynamics equation in combination with the matching relationship between the vehicle speed and the acceleration; performs real-time online identification of the identification parameter vector of the identification model according to the motor torque, the vehicle speed and the acceleration based on a combination of a forgetting factor and a gradient descent method, and determines the real-time vehicle mass according to the identification parameter vector, which has high precision and reliability and improves the accuracy and stability of vehicle mass estimation.

[0023] The above description is only an overview of the technical solution of the embodiment of the present invention. In order to more clearly understand the technical means of the embodiment of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiment of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0025] Figure 1 A schematic diagram showing a flow chart of a method for real-time estimation of vehicle mass provided by an embodiment of the present invention;

[0026] Figure 2 A schematic diagram showing a specific framework for real-time estimation of vehicle mass provided by an embodiment of the present invention is shown;

[0027] Figure 3 A schematic diagram showing the structure of a real-time vehicle mass estimation device provided by an embodiment of the present invention is shown;

[0028] Figure 4 A schematic diagram of an electronic device in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0029] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.

[0030] Figure 1FIG. 2 is a flow chart showing a method for real-time estimation of vehicle mass provided by an embodiment of the present invention. Figure 1 As shown, the vehicle mass real-time estimation method is applied to a server, including:

[0031] Step S11: Acquire the vehicle speed and motor torque at the current moment, and calculate the vehicle acceleration in real time according to the vehicle speed at the current moment using a preset calculation method.

[0032] The embodiment of the present invention uses a method combining acceleration estimation, differential identification model and gradient descent to estimate the vehicle mass. The specific framework is as follows: Figure 2 As shown, the input is the vehicle speed and motor torque, and the acceleration a at time k is calculated based on the vehicle speed. k and the acceleration a at time k-1 k-1 , and obtain the speed v at time k k and the velocity v at time k-1 k-1 , calculate the traction force or braking force according to the motor torque, and get the traction force or braking force f at time k k and the traction or braking force f at time k-1 k-1 Then, the constructed difference identification model is identified according to the obtained physical quantities to obtain the identification parameter vector of the difference identification model, thereby obtaining the vehicle mass m.

[0033] In step S11, optionally, the initialization parameters are first determined. Based on the experience used in practice, the initialization parameters shown in Table 1 are given. The larger the fast coefficient r, the faster the convergence, but the greater the fluctuation. The filter coefficient c0 is actually the number of cycles of the smoothed input signal, that is, the output will have a delay of c0 cycles. The larger the filter coefficient, the smoother the output, but the convergence will be slower.

[0034] In order to improve the convergence speed without causing output fluctuations, this patent proposes a method for predicting the input signal. That is, predict the input signal based on the differential state x2. The number of prediction cycles k pred The larger the value, the better the delay compensation, but the output fluctuations become larger and the details become less.

[0035] Considering the filter coefficient and the number of prediction cycles, the total number of delay cycles is: k d =c0-k pred The delay period plays an important role in the subsequent correction of the vehicle mass estimation. In order to reduce the delay time, the operation period T s It needs to be as small as possible and is set to 10ms.

[0036] Table 1 Acceleration calculation method

[0037]

[0038]

[0039] After determining the initialization parameters, a preset fast convergence function is applied to update in real time the first state variable representing the smoothed value of the vehicle speed after filtering and the second state variable representing the approximate differential value of the vehicle speed according to the initialization parameters and the vehicle speed; the updated second state variable is used as the calculated acceleration. The core function in the acceleration calculation method is u=FasrFun(x1,x2,r,h0). The differential function is realized by applying discrete optimal control theory to make the state track the input signal as quickly as possible, and avoid the flutter phenomenon when the system enters a steady state. The function FasrFun is derived using numerical calculation, and the specific process is shown in Table 2.

[0040] Table 2 FasrFun function for calculating acceleration

[0041]

[0042]

[0043] In the embodiment of the present invention, the delay period number k is also determined according to the initialization parameter. d =c0-k pred ; Determine that the calculated vehicle acceleration corresponds to the vehicle speed at the moment before the delay cycle number at the current moment, that is, the calculated acceleration a at the k moment k Actually it is with kk d The vehicle speed at that moment.

[0044] Step S12: obtaining an identification model of the vehicle mass according to a vehicle dynamics equation combined with a matching relationship between the vehicle speed and the acceleration.

[0045] In the embodiment of the present invention, the design of the vehicle estimation algorithm includes three parts: identification model derivation, identification model correction and real-time estimation algorithm. In step S12, the vehicle dynamics equation is optionally discretized, and the difference equation at two consecutive moments is calculated; the initial identification model of the vehicle mass is constructed according to the difference equation; the initial identification model is corrected according to the matching relationship between the vehicle speed and the acceleration to obtain the identification model of the vehicle mass.

[0046] The vehicle dynamics equation represents the balance between driving force and driving resistance, including rolling resistance, wind resistance, slope resistance, and acceleration resistance, as shown below:

[0047]

[0048] Where, m is the mass of the vehicle, in kg; g is the acceleration due to gravity; f ris the rolling resistance coefficient; θ is the slope angle; C d is the air resistance coefficient; ρ is the density of air, generally ρ=1.2258N·s 2 m -4 ; A p is the windward area, in m 2 ; V r is the relative speed between the car and the wind (vehicle speed when there is no wind), in m / s; a is the acceleration at time k, in m / s 2 , f is the traction force output by the motor or the braking force fed back, usually given by the motor torque T e The calculation is as follows:

[0049]

[0050] Among them, T e is the motor torque, i is the total transmission ratio, η is the transmission efficiency, and R is the wheel radius.

[0051] The above vehicle dynamics equation is deformed and the variables are replaced to obtain the deformed vehicle dynamics equation:

[0052] f=A+B·V r 2 +ma

[0053]

[0054] The variable in variable A is the slope angle θ and the rolling resistance coefficient f r , both remain basically unchanged in a single cycle. Similarly, variable B does not contain time variables and remains unchanged in a single cycle. Based on this, the deformed vehicle dynamics equation is discretized, and the equations at two consecutive moments are subtracted to obtain the differential equation:

[0055] Δf k =B((v k +V W ) 2 -(v k-1 +V W ) 2 )+mΔa k

[0056] Where Δf k is the traction force difference or braking force difference at time k, v k is the speed of the vehicle at time k, V W is the wind speed, the positive direction is defined as the same direction as the vehicle speed, Δa k is the acceleration difference at time k. k +V W The relative speed V between the car and the wind is specifiedr Among them, v k The vehicle speed at the current moment, in m / s.

[0057] From equation a 2 -b 2 =(ab) 2 +2b(ab), the difference equation is transformed to obtain the following relationship:

[0058] Δf k =B(v k -v k-1 ) 2 -2B·V W (v k -v k-1 )+mΔa k

[0059] Let C = 2B·V W , the transformed difference equation is:

[0060]

[0061] The deformed differential equation is written in matrix form to obtain the initial identification model of the vehicle mass:

[0062]

[0063] Wherein, θ is an identification parameter vector in the identification model, X is an input vector of the identification model, and Y is an output of the identification model, which represents a traction force difference or a braking force difference.

[0064] The input of the initial identification model is X = [(v k -v k-1 ) 2 (v k -v k-1 ) Δa k ], the output of the initial identification model Y = Δf k , the identification parameter vector of the initial identification model is θ=[BC m] T By identifying the parameter vector as θ, the vehicle mass can be obtained.

[0065] In the embodiment of the present invention, the initial identification model is a linear regression model. The identification model eliminates the two variables of slope resistance and rolling resistance in the form of difference, and the remaining variables are constants in a short time. This greatly facilitates the algorithm design of parameter identification.

[0066] According to actual test experience, there is another key factor that affects the accuracy of the initial identification model. That is, the acceleration a in the input variable kand speed v k The matching relationship of speed v k In vehicle control, it is calculated from the wheel speed, which has high real-time performance. The acceleration value is usually obtained from a sensor based on the gyroscope principle, or calculated by an estimation algorithm. In either case, a stable value can only be obtained after filtering and denoising. Filtering and denoising will inevitably cause delays. Therefore, it is very important to determine the acceleration delay period. The embodiment of the present invention estimates the acceleration through an algorithm, and the number of delay cycles can be adjusted to solve this key problem. It can be seen from step S11 that the number of delay cycles is k d The current acceleration a k , which actually corresponds to k d The speed value before time Therefore, k d The speed before the moment is substituted into the initial identification model to obtain the corrected final identification model. That is, according to the matching relationship between the vehicle speed and the acceleration, the vehicle speed before the delay cycle number of the current moment corresponding to the calculated vehicle acceleration is substituted into the initial identification model to obtain the identification model of the vehicle mass:

[0067]

[0068] Among them, k d is the number of delay cycles, v k-kd For vehicle kk d The vehicle speed at the moment.

[0069] Step S13: Based on the forgetting factor combined with the gradient descent method, the identification model is subjected to real-time online identification of the identification parameter vector according to the motor torque, the vehicle speed and the acceleration, and the real-time vehicle mass is determined according to the identification parameter vector.

[0070] Optionally, determine the initial identification parameter vector θ in the identification model, θ = [BCm] T , and calculate the traction force difference or braking force difference Y at time k and time k-1 according to the motor torque; determine the input vector X of the identification model according to the vehicle speed and the acceleration, X=[(v k -v k-1 ) 2 (v k -v k-1 )Δa k ]; If the acceleration difference Δa between time k and time k-1 is detected kThe absolute value of is greater than the preset value, and the gradient descent method is used to perform recursive real-time online identification on the identification model with the preset forgetting factor until the real-time vehicle mass is determined to be between the unloaded mass and the fully loaded mass according to the identification parameter vector θ. The vehicle mass is the third item of the identification parameter vector θ. Based on the aforementioned identification model, the real-time estimation algorithm of the vehicle mass is shown in Table 3.

[0071] Table 3 Real-time vehicle mass estimation algorithm

[0072]

[0073] The embodiment of the present invention combines the forgetting factor with the gradient descent method for online identification. The calculation is performed in a recursive manner. The operation of each cycle is very small, and there is no need to save a large amount of historical data. It is suitable for embedded computing. A key to the identification algorithm is the screening of valid data. In the embodiment of the present invention, according to actual test experience, the condition |Δa is met. k |>0.1 greatly improves the accuracy and stability of identification. This condition can also avoid the situation where the denominator in the identification formula is zero, theoretically ensuring the reliable operation of identification. min ) and full load mass (m max ), output the vehicle mass value at that time.

[0074] The embodiment of the present invention proposes a vehicle mass estimation method that combines acceleration estimation, differential identification model, and gradient descent, which improves the estimation accuracy and stability of the vehicle mass. The method has a small amount of calculation, does not need to store a large amount of data, and is suitable for operation in an embedded controller. Specifically, in the acceleration calculation method, a method for predicting the input signal is proposed, and the acceleration estimation can theoretically obtain the delay time between the acceleration and the speed, realize accurate input signal time synchronization, and improve the convergence speed of the result; the number of delay cycles between the acceleration and the vehicle speed is theoretically determined, and the time synchronization of the input signal of the vehicle mass identification model is realized; the differential model of vehicle mass identification completely decouples rolling resistance, slope resistance, and wind resistance, and converts the identification model into a linear relationship between torque and speed; in the gradient descent algorithm of vehicle mass estimation, the screening of input signals and the bounded judgment of identification results are added, which improves the stability of the estimation.

[0075] In summary, the real-time estimation method of the vehicle mass of the embodiment of the present invention obtains the vehicle speed and motor torque at the current moment, and calculates the acceleration of the vehicle in real time according to the preset calculation method based on the vehicle speed at the current moment; obtains the identification model of the vehicle mass according to the vehicle dynamics equation combined with the matching relationship between the vehicle speed and the acceleration; based on the forgetting factor combined with the gradient descent method, the identification parameter vector of the identification model is identified in real time online according to the motor torque, the vehicle speed and the acceleration, and the real-time vehicle mass is determined according to the identification parameter vector, which has high precision and reliability, and improves the accuracy and stability of vehicle mass estimation.

[0076] The above specific embodiments of the present invention are described. In some cases, the actions or steps recorded in the embodiments of the present invention can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0077] Based on the same concept, the embodiment of the present invention also provides a real-time estimation device for vehicle mass. Figure 3 As shown, the real-time vehicle mass estimation device includes: an acceleration calculation unit, an identification model acquisition unit and a mass estimation unit.

[0078] An acceleration calculation unit, used to obtain the vehicle speed and motor torque at the current moment, and calculate the vehicle acceleration in real time according to the vehicle speed at the current moment using a preset calculation method;

[0079] An identification model acquisition unit, used to acquire an identification model of the vehicle mass according to a vehicle dynamics equation combined with a matching relationship between the vehicle speed and the acceleration;

[0080] The mass estimation unit is used to perform real-time online identification of the identification parameter vector of the identification model according to the motor torque, the vehicle speed and the acceleration based on a combination of a forgetting factor and a gradient descent method, and determine the real-time vehicle mass according to the identification parameter vector.

[0081] For the convenience of description, the above devices are described as various modules according to their functions. Of course, when implementing the embodiments of the present invention, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0082] The above-mentioned device is applied to the corresponding method in the aforementioned embodiment and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0083] Based on the same inventive concept, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in any one of the above embodiments is implemented.

[0084] An embodiment of the present invention provides a non-volatile computer storage medium, wherein the computer storage medium stores at least one executable instruction, and the computer executable instruction can execute the method described in any of the above embodiments.

[0085] Figure 4 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 401, a memory 402, an input / output interface 403, a communication interface 404, and a bus 405. The processor 401, the memory 402, the input / output interface 403, and the communication interface 404 are connected to each other in communication within the device through the bus 405.

[0086] The processor 401 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solution provided by the method embodiment of the present invention.

[0087] The memory 402 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 402 can store an operating system and other application programs. When the technical solution provided by the method embodiment of the present invention is implemented by software or firmware, the relevant program code is stored in the memory 402 and called and executed by the processor 401.

[0088] The input / output interface 403 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0089] The communication interface 404 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0090] The bus 405 comprises a pathway for transmitting information between the various components of the device (eg, the processor 401 , the memory 402 , the input / output interface 403 , and the communication interface 404 ).

[0091] It should be noted that, although the above device only shows the processor 401, the memory 402, the input / output interface 403, the communication interface 404 and the bus 405, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiment of the present invention, and does not necessarily include all the components shown in the figure.

[0092] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present disclosure is limited to these examples. Based on the concept of the present disclosure, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present invention as described above, which are not provided in detail for the sake of simplicity.

[0093] This application is intended to cover all such substitutions, modifications and variations that fall within the broad scope of all embodiments. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present invention should be included in the scope of protection of this disclosure.

Claims

1. A real-time estimation method for vehicle mass, characterized in that: The method comprises: Obtaining the vehicle speed and motor torque at the current moment, and calculating the vehicle acceleration in real time according to the vehicle speed at the current moment using a preset calculation method; Obtaining an identification model of the vehicle mass according to a vehicle dynamics equation combined with a matching relationship between the vehicle speed and the acceleration; Based on the forgetting factor combined with the gradient descent method, the identification model is subjected to real-time online identification of the identification parameter vector according to the motor torque, the vehicle speed and the acceleration, and the real-time vehicle mass is determined according to the identification parameter vector.

2. The method according to claim 1, characterized in that The step of calculating the acceleration of the vehicle in real time according to the vehicle speed at the current moment by using a preset calculation method comprises: Determine initialization parameters; Applying a preset fast convergence function to update in real time a first state variable representing a smoothed value of the vehicle speed after filtering and a second state variable representing an approximate differential value of the vehicle speed according to the initialization parameters and the vehicle speed at the current moment; The updated second state variable is used as the calculated acceleration.

3. The method according to claim 2, characterized in that The method of calculating the acceleration of the vehicle in real time according to the vehicle speed at the current moment by applying a preset calculation method also includes: Determine the number of delay cycles according to the initialization parameter; It is determined that the calculated vehicle acceleration corresponds to the vehicle speed at a time before the delay period number at the current time.

4. The method according to claim 1, characterized in that: The identification model for obtaining the vehicle mass according to the vehicle dynamics equation combined with the matching relationship between the vehicle speed and the acceleration includes: Discretize the vehicle dynamics equations and calculate the difference equations at two consecutive moments; constructing an initial identification model of the vehicle mass according to the differential equation; The initial identification model is corrected according to the matching relationship between the vehicle speed and the acceleration to obtain an identification model of the vehicle mass.

5. The method according to claim 4, characterized in that The initial identification model of the vehicle mass is constructed according to the differential equation, including: The difference equation is transformed to obtain the following relationship: Δf k =B(v k -v k-1 ) 3 -2B V W (v k -v k-1 )+mΔa k Where Δf k is the difference in traction or braking force between time k and time k-1, v k is the vehicle speed at time k, V W is the wind speed, m is the vehicle mass; g is the acceleration due to gravity; f r is the rolling resistance coefficient; θ is the slope angle; C d is the air resistance coefficient; ρ is the air density; A p is the frontal area; Δa k is the acceleration difference between time k and time k-1; The deformed differential equation is written in matrix form to obtain the initial identification model of the vehicle mass: Wherein, θ is an identification parameter vector in the identification model, X is an input vector of the identification model, and Y is an output of the identification model, which represents a traction force difference or a braking force difference.

6. The method according to claim 5, characterized in that The step of correcting the initial identification model according to the matching relationship between the vehicle speed and the acceleration to obtain the identification model of the vehicle mass includes: According to the matching relationship between the vehicle speed and the acceleration, the vehicle speed at the time before the delay period number corresponding to the calculated vehicle acceleration at the current time is substituted into the initial identification model to obtain the identification model of the vehicle mass: Among them, k d is the number of delay cycles, For vehicle kk d The vehicle speed at the moment.

7. The method according to claim 1, characterized in that The real-time online identification of the identification parameter vector of the identification model based on the combination of the forgetting factor and the gradient descent method according to the motor torque, the vehicle speed and the acceleration includes: Determine an initial identification parameter vector θ in the identification model, and calculate a traction force difference or a braking force difference Y at time k and time k-1 according to the motor torque; Determining an input vector X of the identification model according to the vehicle speed and the acceleration; If it is detected that the absolute value of the acceleration difference between time k and time k-1 is greater than a preset value, the identification model is recursively identified in real time online using a gradient descent method with a preset forgetting factor until the real-time vehicle mass is determined to be between the empty mass and the fully loaded mass according to the identification parameter vector θ.

8. A real-time vehicle mass estimation device, characterized in that: The device comprises: An acceleration calculation unit, used to obtain the vehicle speed and motor torque at the current moment, and calculate the vehicle acceleration in real time according to the vehicle speed at the current moment using a preset calculation method; An identification model acquisition unit, used to acquire an identification model of the vehicle mass according to a vehicle dynamics equation combined with a matching relationship between the vehicle speed and the acceleration; The mass estimation unit is used to perform real-time online identification of the identification parameter vector of the identification model according to the motor torque, the vehicle speed and the acceleration based on a combination of a forgetting factor and a gradient descent method, and determine the real-time vehicle mass according to the identification parameter vector.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method according to any one of claims 1 to 7 is implemented.

10. A computer storage medium, characterized in that: The storage medium stores at least one executable instruction, and the executable instruction enables the processor to execute the method according to any one of claims 1 to 7.

Images