ESTIMATION OF THE LOAD OF A VEHICLE

• US 20110066322A1
• Filed: 12/03/2007
• Published: 03/17/2011
• Est. Priority Date: 12/03/2007
• Status: Active Grant

• Current Assignee: Nira Dynamics AB
• Original Assignee: Nira Dynamics AB
• Inventors: Andersson, Andreas, Josefsson, Jonas, Karlsson, Rickard
• Granted Patent: US 8,630,767 B2
• Time in Patent Office: Days
• US Class Current: 701/35
• CPC Class Codes: B60G 2400/52 in tyre
• CPC Class Codes: B60G 2400/60 Load
• CPC Class Codes: B60G 2800/70 Estimating or calculating vehicle parameters or state variables
• CPC Class Codes: B60T 2250/02 Vehicle mass
• CPC Class Codes: B60T 2270/86 Optimizing braking by using ESP vehicle or tire model
• CPC Class Codes: B60T 8/172 Determining control parameters used in the regulation, e.g. by calculations involving measured or detected parameters B60T8/17551 takes precedence
• CPC Class Codes: G01G 19/086 wherein the vehicle mass is dynamically estimated

Abstract

A method, an apparatus and a computer program product for estimating the current load of a vehicle wherein a filter bank including a filter for different weight classes, each filter implementing a vehicle model for estimating the current mass of the vehicle. Based on vehicle data indicative of a current driving situation of the vehicle, and filter parameter specific for the respective weight class, each filter provides a load estimation value as filter specific estimation of the current load of the vehicle.

Claims

1. A method of estimating the current load of a vehicle, comprising:

defining at least two weight classes;

providing a filter bank including, for each of the at least two weight classes, a filter, each filter implementing a vehicle model for estimating the current mass of the vehicle;

supplying, to each filter, at least one vehicle data indicative of a current driving situation of the vehicle, and at least one current filter parameter specific for the respective weight class;

determining, by means of each filter, a load estimation value as filter specific estimation of the current load of the vehicle;

determining, for each load estimation value, a current indication whether it can be assumed that the current load of the vehicle belongs to the respective weight class; and

selecting, among all load estimation values, the load estimation value having the best current indication as estimation of the current load of the vehicle for the current driving situation and/or weighing the current indications for the weight classes to obtain an overall load estimation value for the filter bank, wherein the overall load estimation value is used as load estimation for the current load of the vehicle for the current driving situation.

2. The method of claim 1, wherein

supplying, to each filter, at least one vehicle data, and at least one current filter parameter;

determining, by means of each filter, a load estimation value;

determining, for each load estimation value, a current indication; and

selecting, among all load estimation values, the mass estimation for which the best current indication is determined;

are repeated.

3. The method of claim 2, wherein the repetition is carried out in predefined time intervals and/or each time the actual mass of the vehicle is to be estimated.

4. The method of claim 1, wherein each filter comprises at least one of a model-based filter, a Kalman filter, an extended Kalman filter, an unscented Kalman filter, a constraint extended Kalman filter, a particle filter, a sigma-point filter, point-mass filter, a grid-based filter.

5. The method of claim 1, wherein the vehicle model includes at least one of the following parameters:

Gearbox efficiency (η

_tm);

engine torque (T_e);

gearbox damping constant (ζ

)

engine angular velocity (ω

_e);

engine inertia (J_e);

wheel angular velocity (ω

_w);

wheel angular acceleration ({dot over (ω

)}_w);

a gear ratio (i_g);

a final drive gear ratio (i_gf);

wheel inertia (J_e);

air density (ρ

);

aerodynamic drag coefficient (C_d);

vehicle aerodynamic front area (A_f);

vehicle speed (v);

a previously determined load estimation value ({circumflex over (m)}) from the respective filter;

a previously determined indication whether it can be assumed that the current load of the vehicle belongs to the respective weight class from the respective filter;

the gravitational constant (g);

rolling resistance coefficient (C_r); and

wheel radius (r).

6. The method of claim 1, wherein supplying at least one vehicle data indicative of the current driving situation includes at least one of

supplying an acceleration sensor value (y) indicating a current vehicle acceleration;

supplying an engine torque value (T_e) indicating a current torque generated by the vehicle'"'"'s engine;

supplying at least one wheel angular velocity value (w_w) indicating a current angular velocity of at least one vehicle'"'"'s wheel; and

supplying an engine angular velocity value (ω

_e) indicating a current angular velocity of the vehicle'"'"'s engine; and

supplying a final drive ratio value (i_fd) indicating a current final drive ratio.

7. The method of claim 1, wherein supplying at least one current filter parameter specific for the respective weight class includes supplying at least one the following filter parameters:

a previous load estimation value of the respective filter;

a previous mean of the respective filter;

a previous covariance of the respective filter; and

a previous probability of the respective filter.

8. The method of claim 7, wherein the previous covariance includes at least one of a covariance of a model error of the respective vehicle model and a covariance of measurement noise associated to the at least one vehicle data.

9. The method of claim 7, wherein

the previous mean is an initial mean assumed for the respective vehicle model or is a mean of the respective vehicle model resulting from a previously carried out determining a load estimation value;

and/or

the previous covariance is an initial covariance assumed for the respective vehicle model or is a covariance of the respective vehicle model resulting from a previously carried out determining a load estimation value; and

/or

the previous probability is an initial probability assumed for the respective vehicle model or a probability of the respective vehicle model resulting from a previously carried out step of determining a current probability.

10. The method of claim 1, wherein determining a current probability that the actual mass of the vehicle belongs to the respective weight class includes using the Bayes'"'"' theorem.

11. The method of claim 1, wherein determining a current probability that the actual mass of the vehicle belongs to the respective weight class includes calculating the current probability on the basis of a likelihood of the respective load estimation value and a previously calculated probability of the respective weight class.

12. The method of claim 11, wherein the likelihood of the respective load estimation value is determined on the basis of acceleration of the vehicle by exciting the respective vehicle model with at least two different loads of the vehicle wherein the actual vehicle'"'"'s mass without load is known.

13. The method of claim 1, wherein determining a load estimation value includes determining a weight related function, for the respective vehicle model, of a residual between the current load of the vehicle and an load estimation value obtained from the respective vehicle model being excited with the current load of the vehicle.

14. The method of claim 1, wherein determining a load estimation value includes determining, for each vehicle model, an uncertainty for the respective load estimation value.

15. The method of claim 1, wherein the uncertainty is determined by at least one of:

an operational mode of a gearbox of the vehicle;

acceleration magnitude; and

engine torque magnitude.

16. The method of claim 1, wherein the vehicle model includes a model for the vehicle'"'"'s acceleration and a model for a vehicle acceleration value provided by a sensor means of the vehicle, the vehicle acceleration value indicating a current acceleration of the vehicle.

17. The method of claim 16, wherein the model for the vehicle'"'"'s acceleration is based on the following equation:

18. The method of claim 16, wherein the model for the vehicle acceleration value is based on the following equation:

{dot over (v)}+g sin α

/γ

_m+gC_rγ

_m

wherein

{dot over (v)} is the vehicle acceleration;

g is the gravitational constant;

α

is an angle between the sensor means'"'"' measurement axis and the earth'"'"'s gravity field;

C_r is a rolling resistance coefficient; and

γ

_m is a mass factor for acceleration of the vehicle.

19. The method of claim 16, wherein the model for the output of a sensor means is based on the following equation:

y_acc+gC_r/γ

_m

wherein

y_acc is a measured acceleration of the vehicle;

g is the gravitational constant;

C_r is a rolling resistance coefficient; and

γ

_m is a mass factor for acceleration of the vehicle.

20. The method of claim 16, wherein the mass factor y. is represented by:

21. An apparatus for estimation of the current load of a vehicle, comprising:

a filter bank including at least two filters, each filter implementing a vehicle model for estimating the mass of the vehicle and being associated to one of at least two weight classes;

an input for supplying, to each filter, at least one vehicle data indicative of a current driving situation of the vehicle, and at least one current filter parameter specific for the respective weight class;

wherein

each filter is adapted to determine a load estimation value as filter specific estimation of the current load of the vehicle;

determination means being adapted to determine, for each load estimation value, a current indication whether it can be assumed that the current load of the vehicle belongs to the respective weight class; and

means being adapted to select, among all load estimation values, the load estimation value for which the best current indication is determined as estimation of the current load of the vehicle for the current driving situation and/or being adapted to weigh the current indications for the weight classes to obtain an overall load estimation value for the filter bank, wherein the overall load estimation value is used as load estimation for the current load of the vehicle for the current driving situation.

22. The apparatus of claim 21, wherein each filter comprises at least one of a model-based filter, a Kalman filter, an extended Kalman filter, an unscented Kalman filter, a constraint extended Kalman filter, a particle filter, a sigma-point filter, point-mass filter, a grid-based filter.

23. The apparatus of claim 21, wherein the input for supplying at least one vehicle data comprises at least one of

an input for receiving an acceleration sensor value (y) indicating a current vehicle acceleration provided by a vehicle sensor;

an input for receiving an engine torque value (T_e) indicating a current torque generated by the vehicle'"'"'s engine; and

an input for receiving at least one wheel angular velocity value (w_w) indicating a current angular velocity of at least one vehicle'"'"'s wheel.

an input for receiving an engine angular velocity value (ω

_e) indicating a current angular velocity of the vehicle'"'"'s engine; and

an input for receiving a final drive ratio value (i_fd) indicating a current final drive ratio.

24. A computer program product comprising program code for carrying out a method, when executed on a processing system, of estimating the current load of a vehicle, comprising:

defining at least two weight classes;

providing a filter bank including, for each of the at least two weight classes, a filter, each filter implementing a vehicle model for estimating the current mass of the vehicle;

supplying, to each filter, at least one vehicle data indicative of a current driving situation of the vehicle, and at least one current filter parameter specific for the respective weight class;

determining, by means of each filter, a load estimation value as filter specific estimation of the current load of the vehicle;

determining, for each load estimation value, a current indication whether it can be assumed that the current load of the vehicle belongs to the respective weight class; and

selecting, among all load estimation values, the load estimation value having the best current indication as estimation of the current load of the vehicle for the current driving situation and/or weighing the current indications for the weight classes to obtain an overall load estimation value for the filter bank, wherein the overall load estimation value is used as load estimation for the current load of the vehicle for the current driving situation.

25. The computer program product of claim 24, comprising program code for carrying out, when executed on a processing system, a method of estimating the current load of a vehicle,

wherein the operations of

supplying, to each filter, at least one vehicle data, and at least one current filter parameter;

determining, by means of each filter, a load estimation value;

determining, for each load estimation value, a current indication; and

selecting, among all load estimation values, the mass estimation for which the best current indication is determined;

are repeated.

26. The computer program product of claim 24 being stored on a computer readable storage medium or being stored in a storage of a processing system.