CN116834756A - Vehicle mass and road gradient online estimation method, device and equipment - Google Patents

Abstract

The application relates to a vehicle mass and road gradient online estimation method, a device and equipment, which are suitable for an electric vehicle and a hub motor thereof, and are used for estimating the longitudinal force born by a tire by acquiring vehicle parameters, the output torque of the hub motor of the vehicle, the rotation speed of the wheel and the longitudinal acceleration measurement value of the vehicle, and establishing a longitudinal dynamics model of the vehicle by combining the longitudinal acceleration measurement value of the vehicle; according to a longitudinal dynamics model of the vehicle, designing a recursive least square method for adaptively introducing forgetting factors, calculating acceleration residual errors, judging whether the forgetting factors need to be introduced according to whether the acceleration residual errors are larger than a preset value, and estimating the vehicle mass and the road gradient on line; the method and the device realize simultaneous online estimation of the vehicle mass and the road gradient, can quickly converge and improve the stability of the estimation result when the running condition of the vehicle changes, and have high precision, strong working condition adaptability, quick convergence and good stability.

Description

A method, device and equipment for online estimation of vehicle mass and road gradient

Technical field

The invention relates to the field of vehicle technology, and in particular to a method, device and equipment for online estimation of vehicle mass and road gradient.

Background technique

With the improvement of electric vehicle intelligence, accurate estimation of vehicle mass and road gradient is very important for active safety and control of electric vehicles. However, when estimating the vehicle mass and slope, most of the methods currently use the engine output torque to be transmitted to the wheels to calculate the longitudinal force on the vehicle. Since the output torque needs to be transmitted through multiple wheels to reach the wheels, the calculated longitudinal force will be There is a bias and it does not apply to electric vehicles. Secondly, most existing technologies establish dynamic models based on vehicle acceleration to estimate vehicle mass and road gradient. The vehicle acceleration depends on the vehicle speed differential. When the vehicle speed estimation is inaccurate, it will affect the acceleration estimation. As a result, the accuracy of vehicle mass and road gradient is affected. At the same time, many algorithms in the existing technology estimate vehicle mass or road gradient offline, which is not conducive to real-time acquisition of vehicle status information and real-time control of vehicle active safety.

With the development of electric vehicles and wheel hub motors, the calculation of the tire longitudinal force can be directly obtained by calculating the wheel hub motor and wheel speed sensor. The obtained tire longitudinal force and acceleration are more accurate. Therefore, the present invention hopes to use the longitudinal acceleration measured by the acceleration sensor. Establish a vehicle longitudinal dynamics model to more conveniently and accurately estimate vehicle mass and road gradient online.

Contents of the invention

In order to solve the above technical problems, the present invention provides an online estimation method of vehicle mass and road gradient, which includes the following steps:

Step 1: Obtain vehicle parameters, and obtain the vehicle wheel hub motor output torque, wheel rotation angular velocity, and vehicle longitudinal acceleration measurement values through vehicle-mounted sensors;

Step 2: Use the hub motor to output torque and calculate the longitudinal force on each tire:

Among them, F _xi is the longitudinal force on each tire of the vehicle, Ti _is the output torque of each wheel hub motor, I _w is the rotational inertia of the wheel, ω _i is the rotation angular velocity of each wheel, and r _e is the effective rolling of the wheel. radius;

Add the longitudinal forces F _xi on each wheel to obtain the total longitudinal force on the tire F _xtotal ;

Calculate the driving air resistance F _aero :

Among them, C _D is the air resistance coefficient, A is the windward area, v is the vehicle speed,

Calculate the road resistance F _grade , including rolling resistance and grade resistance:

F _grade =Mg(μcosβ+sinβ)

The total longitudinal force F _xtotal on the tire is subtracted from the driving air resistance F _aero and the road resistance F _grade , and then the vehicle dynamics equation is obtained:

in, to accelerate the vehicle.

According to the vehicle longitudinal acceleration measurement value, a vehicle longitudinal dynamics model represented by the longitudinal acceleration measurement value is established:

The longitudinal acceleration measurement value a _sen obtained by the acceleration sensor and the vehicle acceleration The relationship is expressed as:

Among them, g is the acceleration of gravity, β is the road slope;

Substituting into the vehicle dynamics equation we get:

Further sorting out the vehicle longitudinal dynamics model represented by longitudinal acceleration measurements:

in, μ is the tire rolling resistance coefficient, M is the vehicle mass, and β is the road gradient;

Step 3: Based on the vehicle longitudinal dynamics model, design a recursive least squares method that adaptively introduces the forgetting factor, calculate the acceleration residual, and determine whether the forgetting factor needs to be introduced based on whether the acceleration residual is greater than the preset value, and estimate online Vehicle mass and road grade, including:

Convert the vehicle longitudinal dynamics model into a linear equation form:

y=φ ^T θ

Among them, y= _asen ,

Calculate the acceleration residual ε _k :

When the acceleration residual is less than the preset value, it means that the mass of the vehicle and the road gradient change very little or remain unchanged. At this time, the forgetting factor is not introduced in the recursive least squares method to make the estimation results smoother and reduce fluctuations. The estimation results are more stable and the target estimation parameter vector is solved online. The calculation steps are:

L(k)＝P(k-1)φ(k)(1+φ ^T (k)P(k-1)φ(k)) ^-1

P(k)＝(IL(k)φ ^T (k))P(k-1)

When the acceleration residual is greater than the preset value, a forgetting factor is introduced in the recursive least squares method to increase the weight of new data to quickly track changes in vehicle mass or road gradient, which can make the estimation results converge quickly and solve the target estimated parameter vector online. The calculation steps are:

L(k)＝P(k-1)φ(k)(λ+φ ^T (k)P(k-1)φ(k)) ^-1

Among them, P(k) is the covariance matrix at time k, L(k) is the gain matrix at time k, φ(k) is the coefficient vector at time k, is the target estimated parameter vector at time k, λ is the forgetting factor, 0＜λ＜1;

The preset value range is 0.05~0.3.

based on estimates Converted into estimates of vehicle mass and road gradient:

in, and/> are the target parameter estimation vectors/> The first and second elements of;

Step 4: Output the vehicle mass and road slope estimation results online in real time.

The invention also provides an online estimation device for vehicle mass and road gradient, including: an acquisition module, an estimation module and an output module;

In the acquisition module, the above-mentioned step 1 is processed to obtain vehicle parameters and vehicle sensor information;

In the estimation module, the above-mentioned steps 2 and 3 are processed, the longitudinal force on each tire is calculated according to the vehicle parameters and vehicle-mounted sensor information, and the vehicle longitudinal dynamic force represented by the longitudinal acceleration measurement value is established based on the vehicle longitudinal acceleration measurement value. learning model; calculate the acceleration residual, based on a recursive least squares method that adaptively introduces a forgetting factor, determine whether the forgetting factor needs to be introduced based on whether the acceleration residual is greater than the preset value, and estimate the vehicle mass and road gradient online;

In the output module, the above-mentioned step 4 is processed and the vehicle mass and road slope estimation results are output online.

The invention also provides an electronic device, including: a memory and a processor;

The memory is used to store executable instructions; the processor is used to implement the above online estimation method of vehicle mass and road gradient when executing the executable instructions stored in the memory.

The present invention also provides a computer-readable medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned online estimation method of vehicle mass and road gradient is implemented.

Beneficial effects of the present invention:

The invention provides an online estimation method, device and equipment for vehicle mass and road gradient, which is suitable for electric vehicles and their hub motors. It adopts a recursive least squares method that adaptively introduces a forgetting factor to achieve simultaneous estimation of vehicle mass and road gradient. Online estimation can quickly converge and improve the stability of the estimation results when the vehicle driving conditions change. The estimation results have high accuracy, strong adaptability to working conditions, fast convergence, and good stability.

Description of the drawings

Figure 1 is a schematic diagram of the overall flow of the vehicle mass and road gradient estimation method of the present invention;

Figure 2 is a schematic structural diagram of the vehicle mass and road gradient estimation device of the present invention;

Figure 3 is a vehicle driving speed change diagram according to the embodiment of the present invention;

Figure 4 is a residual change diagram when the vehicle is driving according to the embodiment of the present invention;

Figure 5 is a schematic diagram of the vehicle quality online estimation results provided by the embodiment of the present invention;

Figure 6 is a schematic diagram of the online estimation results of road slope provided by the embodiment of the present invention.

Detailed ways

As shown in Figure 1: The present invention provides an online estimation method of vehicle mass and road gradient, which includes the following steps:

Step 1: Obtain vehicle parameters, and obtain the vehicle wheel hub motor output torque, wheel rotation angular velocity, and vehicle longitudinal acceleration measurement values through vehicle-mounted sensors;

Step 2: Calculate the longitudinal force F _xi on each tire through the wheel hub motor output torque _Ti :

Among them, F _xi is the longitudinal force on each tire of the vehicle, Ti _is the output torque of each wheel hub motor, I _w is the rotational inertia of the wheel, ω _i is the rotation angular velocity of each wheel, and r _e is the effective rolling of the wheel. radius;

Add the longitudinal forces F _xi on each wheel to obtain the total longitudinal force on the tire F _xtotal ;

Calculate the driving air resistance F _aero :

Among them, C _D is the air resistance coefficient, A is the windward area, v is the vehicle speed,

Calculate the road resistance F _grade , including rolling resistance and grade resistance:

F _grade =Mg(μcosβ+sinβ)

The total longitudinal force F _xtotal on the tire is subtracted from the driving air resistance F _aero and the road resistance F _grade , and then the vehicle dynamics equation is obtained:

in, to accelerate the vehicle.

According to the vehicle longitudinal acceleration measurement value, a vehicle longitudinal dynamics model represented by the longitudinal acceleration measurement value is established:

The longitudinal acceleration measurement value a _sen obtained by the acceleration sensor and the vehicle acceleration The relationship is expressed as:

Among them, g is the acceleration of gravity, β is the road slope;

Substituting into the vehicle dynamics equation we get:

Further sorting out the vehicle longitudinal dynamics model represented by longitudinal acceleration measurements:

in, μ is the tire rolling resistance coefficient, M is the vehicle mass, and β is the road gradient;

Step 3: Based on the vehicle longitudinal dynamics model, design a recursive least squares method that adaptively introduces the forgetting factor, calculate the acceleration residual, and determine whether the forgetting factor needs to be introduced based on whether the acceleration residual is greater than the preset value, and estimate online Vehicle mass and road grade, including:

Convert the vehicle longitudinal dynamics model into a linear equation form:

y=φ ^T θ

Among them, y= _asen ,

Calculate the acceleration residual ε _k :

When the acceleration residual is less than the preset value, it means that the mass of the vehicle and the road gradient change very little or remain unchanged. At this time, the forgetting factor is not introduced in the recursive least squares method to make the estimation results smoother and reduce fluctuations. The estimation results are more stable and the target estimation parameter vector is solved online. The calculation steps are:

L(k)＝P(k-1)φ(k)(1+φ ^T (k)P(k-1)φ(k)) ^-1

P(k)＝(IL(k)φ ^T (k))P(k-1)

When the acceleration residual is greater than the preset value, a forgetting factor is introduced in the recursive least squares method to increase the weight of new data to quickly track changes in vehicle mass or road gradient, which can make the estimation results converge quickly and solve the target estimated parameter vector online. The calculation steps are:

L(k)＝P(k-1)φ(k)(λ+φ ^T (k)P(k-1)φ(k)) ^-1

Among them, P(k) is the covariance matrix at time k, L(k) is the gain matrix at time k, φ(k) is the coefficient vector at time k, is the target estimated parameter vector at time k, λ is the forgetting factor, 0＜λ＜1;

The preset value range is 0.05~0.3.

based on estimates Converted into estimates of vehicle mass and road gradient:

in, and/> are the target parameter estimation vectors/> The first and second elements of;

Step 4: Output the vehicle mass and road slope estimation results online in real time.

As shown in Figure 2: The present invention also provides an online estimation device for vehicle mass and road slope, including: an acquisition module, an estimation module and an output module;

In the acquisition module, the above-mentioned step 1 is processed to obtain vehicle parameters and vehicle sensor information;

In the estimation module, the above-mentioned steps 2 and 3 are processed, the longitudinal force on each tire is calculated according to the vehicle parameters and vehicle-mounted sensor information, and the vehicle longitudinal dynamic force represented by the longitudinal acceleration measurement value is established based on the vehicle longitudinal acceleration measurement value. learning model; calculate the acceleration residual, based on a recursive least squares method that adaptively introduces a forgetting factor, determine whether the forgetting factor needs to be introduced based on whether the acceleration residual is greater than the preset value, and estimate the vehicle mass and road gradient online;

In the output module, the above-mentioned step 4 is processed and the vehicle mass and road slope estimation results are output online.

The invention also provides an electronic device, including: a memory and a processor;

The memory is used to store executable instructions; the processor is used to implement the above online estimation method of vehicle mass and road gradient when executing the executable instructions stored in the memory.

The present invention also provides a computer-readable medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned online estimation method of vehicle mass and road gradient is implemented.

In one embodiment of the present invention, the preset value is 0.2; in order to verify the estimation effect of the estimation method of the present invention, CarSim and Simulink are jointly simulated, where the real mass of the vehicle is 1412kg, the initial road slope is 0°, and 200m After the road slope changes to 5°, the vehicle speed change chart is shown in Figure 3, and the estimation method is simulated and verified. The residual change diagram when the vehicle is driving is shown in Figure 4. The estimation results of the vehicle mass and road slope are shown in Figures 5 and 6 respectively. It can be seen from the figure that after the vehicle starts, the algorithm starts to estimate the acceleration residual ε When _k changes drastically, the recursive least squares method adaptively introduces a forgetting factor so that the estimation results converge quickly. The results show that both the vehicle mass and the road gradient estimates converge quickly, and the mass estimation results converge around the vehicle's true mass of 1412kg, and the road The slope estimation result converges around the true value of 0°; at about 11 seconds, due to the change in the slope of the road surface where the vehicle is traveling, the vehicle starts to go up a 5° slope, so the acceleration residual ε _k also changes drastically. At this time, the recursive least squares method is adaptively introduced. Forgetting factor to quickly converge to tracking changes, vehicle mass and road grade estimates will have shorter fluctuations. When the acceleration residual ε _k is less than the preset value, it means that the vehicle mass or road surface gradient changes little or remains unchanged. The recursive least squares method does not introduce a forgetting factor and improves the stability of the estimation results. It can be seen from the results that both estimators will quickly converge and stabilize near the true values. During the entire estimation process, the maximum error in the estimated vehicle mass is 8kg, and the maximum error in the estimated road gradient is 0.16 degrees. The estimation results show that, The estimation method has the advantages of high accuracy of estimation results, fast convergence, strong adaptability to working conditions, and good stability.

Contents not described in detail in this specification belong to the prior art known to those skilled in the art.

The above descriptions are only examples of this specification and are not intended to limit this application. To those skilled in the art, various modifications and variations may be made to this application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this application shall be included in the scope of the claims of this application.

Claims (9)

1. An online estimation method of vehicle mass and road gradient, characterized by: including the following steps: Step 1: Obtain vehicle parameters, and obtain the vehicle wheel hub motor output torque, wheel rotation angular velocity, and vehicle longitudinal acceleration measurement values through vehicle-mounted sensors; Step 2: Calculate the longitudinal force on each tire through the wheel hub motor output torque; According to the vehicle longitudinal acceleration measurement value, a vehicle longitudinal dynamics model represented by the longitudinal acceleration measurement value is established: Among them, F _xi is the longitudinal force on each tire of the vehicle, Ti _is the output torque of each wheel hub motor, I _w is the rotational inertia of the wheel, ω _i is the rotation angular velocity of each wheel, and r _e is the effective rolling of the wheel. Radius; C _D is the air resistance coefficient, A is the windward area, v is the vehicle speed, μ is the tire rolling resistance coefficient, M is the vehicle mass, and β is the road gradient; Step 3: Calculate the acceleration residual based on the vehicle longitudinal dynamics model, determine whether the forgetting factor needs to be introduced in the recursive least squares method based on whether the acceleration residual is greater than the preset value, and solve the target estimated parameter vector online: When the acceleration residual is less than the preset value, no forgetting factor is introduced in the recursive least squares method; When the acceleration residual is greater than the preset value, a forgetting factor is introduced in the recursive least squares method; The estimated target parameter vector is converted into an estimation result of vehicle mass and road gradient; Step 4: Output the vehicle mass and road slope estimation results online in real time. 2. A method for online estimation of vehicle mass and road gradient according to claim 1, characterized in that: in step 2, the longitudinal force on each tire is calculated by outputting torque from the hub motor: Among them, F _xi is the longitudinal force on each tire of the vehicle, Ti _is the output torque of each wheel hub motor, I _w is the rotational inertia of the wheel, ω _i is the rotation angular velocity of each wheel, and r _e is the effective rolling of the wheel. radius; Add the longitudinal forces F _xi on each wheel to obtain the total longitudinal force on the tire F _xtotal ; Calculate the driving air resistance F _aero : Among them, C _D is the air resistance coefficient, A is the windward area, v is the vehicle speed, Calculate the road resistance F _grade , including rolling resistance and grade resistance: F _grade =Mg(μcosβ+sinβ) The total longitudinal force F _xtotal on the tire is subtracted from the driving air resistance F _aero and the road resistance F _grade , and then the vehicle dynamics equation is obtained: in, Accelerate the vehicle; According to the vehicle longitudinal acceleration measurement value, a vehicle longitudinal dynamics model represented by the longitudinal acceleration measurement value is established: The longitudinal acceleration measurement value a _sen obtained by the acceleration sensor and the vehicle acceleration The relationship is expressed as: Among them, g is the acceleration of gravity, β is the road slope; Substituting into the vehicle dynamics equation we get: Organize and obtain the vehicle longitudinal dynamics model represented by longitudinal acceleration measurements: in, μ is the tire rolling resistance coefficient, M is the vehicle mass, and β is the road gradient. 3. A method for online estimation of vehicle mass and road gradient according to claim 1, characterized in that: the acceleration residual calculation method in step 3 is as follows: Convert the vehicle longitudinal dynamics model into a linear equation form: y=φ ^T θ Among them, y= _asen , Calculate the acceleration residual ε _k : 4. An online estimation method of vehicle mass and road gradient according to claim 1, characterized in that: in step 3, when the acceleration residual is less than the preset value, no forgetting factor is introduced in the recursive least squares method, and the online estimation method is Solve for the target estimated parameter vector The calculation steps are: L(k)＝P(k-1)φ(k)(1+φ ^T (k)P(k-1)φ(k)) ^-1 P(k)＝(IL(k)φ ^T (k))P(k-1) When the acceleration residual is greater than the preset value, a forgetting factor is introduced in the recursive least squares method to solve the target estimated parameter vector online. The calculation steps are: L(k)＝P(k-1)φ(k)(λ+φ ^T (k)P(k-1)φ(k)) ^-1 Among them, P(k) is the covariance matrix at time k, L(k) is the gain matrix at time k, φ(k) is the coefficient vector at time k, is the target estimated parameter vector at time k, λ is the forgetting factor, 0<λ<1. 5. An online estimation method of vehicle mass and road gradient according to claim 4, characterized in that: the preset value is obtained through debugging and ranges from 0.05 to 0.3. 6. A method for online estimation of vehicle mass and road gradient according to claim 1, characterized in that: in step 3, according to the estimated Converted into estimates of vehicle mass and road gradient: in, and/> are the target parameter estimation vectors/> the first and second elements. 7. An online estimation device for vehicle mass and road gradient, characterized by: including an acquisition module, an estimation module and an output module; In the described acquisition module, step 1 of the online estimation method of vehicle mass and road gradient described in any one of claims 1-6 is processed to acquire vehicle parameters and vehicle-mounted sensor information; In the estimation module, steps 2 and 3 of an online estimation method of vehicle mass and road gradient described in any one of claims 1-6 are processed, and the longitudinal force of each tire is calculated based on the vehicle parameters and vehicle-mounted sensor information. Force, based on the vehicle longitudinal acceleration measurement value, establish a vehicle longitudinal dynamics model represented by the longitudinal acceleration measurement value; calculate the acceleration residual, based on a recursive least squares method that adaptively introduces a forgetting factor, based on whether the acceleration residual is greater than the preset value to determine whether it is necessary to introduce a forgetting factor and estimate vehicle mass and road gradient online; In the output module, step 4 of the online vehicle mass and road gradient estimation method described in any one of claims 1 to 6 is processed, and the vehicle mass and road gradient estimation results are output online. 8. An electronic device, characterized by: including: a memory and a processor; The memory is used to store executable instructions; the processor is used to implement the vehicle mass and road gradient online estimation method described in any one of claims 1-6 when executing the executable instructions stored in the memory. 9. A computer-readable medium, characterized in that: a computer program is stored thereon, and when the program is executed by a processor, the online estimation method of vehicle mass and road gradient described in any one of claims 1-6 is implemented.