Method for controlling processes in a motor vehicle

• US 5,713,332 A
• Filed: 09/25/1996
• Issued: 02/03/1998
• Est. Priority Date: 05/28/1994
• Status: Expired due to Fees

• Current Assignee: E. I. du Pont de Nemours and Company (Corteva, Inc.), Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Tolle, Henning, Adler, Walter, Ullrich, Thorsten, Schmitt, Manfred
• Primary Examiner(s): ARGENBRIGHT, TONY MICHAEL
• Application Number:
  US08/718,329
• Time in Patent Office: 496 Days
• Field of Search: 123/417, 123/480, 123/486, 123/674, 123/675, 180/197, 364/424.084, 364/424.085, 364/424.088, 364/431.052, 364/431.053, 364/431.062, 477/121, 477/138
• US Class Current: 701/59
• CPC Class Codes: B60T 2201/14 Electronic locking-differential
• CPC Class Codes: B60T 8/174 characterised by using special control logic, e.g. fuzzy logic , neural computing
• CPC Class Codes: B60T 8/175 Brake regulation specially adapted to prevent excessive wheel spin during vehicle acceleration, e.g. for traction control safety devices for propulsion unit control responsive to, or preventing, skidding of wheels B60K28/16
• CPC Class Codes: F02D 41/0215 in relation with elements of the transmission
• CPC Class Codes: F02D 41/2416 Interpolation techniques
• CPC Class Codes: F02D 41/2429 Methods of calibrating or learning
• CPC Class Codes: F02D 41/2441 characterised by the learning conditions
• CPC Class Codes: F02P 5/1502 using one central computing unit
• CPC Class Codes: G05B 13/0205 not using a model or a simulator of the controlled system
• CPC Class Codes: G05B 13/024 in which a parameter or coefficient is automatically adjusted to optimise the performance
• CPC Class Codes: G05B 13/0265 the criterion being a learning criterion

Abstract

A method for controlling processes in a motor vehicle is intended, for example, for open-loop/closed-loop control of combustion processes, transmission switching processes or braking processes. An engine characteristics map is formed by operating quantities of the respective process and is represented by a number of support points. In the corresponding control unit, for each support point a piece of information about the value of the map at the position of the support point is provided. In the open-loop/closed-loop control of the process, for a determined working point of the process at least one map value that is decisive for this working point is determined from the map and taking account of this, at least one control quantity for the emission of a control signal is formed. The map is automatically adapted to a changed process behavior. For the adaptation, a correction value is determined for a certain operating phase of the map. According to the present invention, additionally for each support point a piece of information about the position of the support point within the map is provided so that the support points are enterable independent of a fixed grid into the map. For adaptation of the map, a number of support points of the map lying around the determined correction value is chosen. These support points are then adapted, the support point correction being carried out distance-weighted with reference to the position of the determined correction value within the map.

Claims

1. A method for controlling at least one process in a motor vehicle, the motor vehicle including a control unit having an engine characteristics map, the method comprising the steps of:

forming the engine characteristics map with operating quantities of the at least one process, the engine characteristics map being represented by at least one support point being positioned at selected locations in the engine characteristics map independent of a fixed grid, each of the at least one support point corresponding to at least one respective engine characteristics value of the engine characteristics map at the selected locations of the at least one support point;

generating at least one of a control quantity and a model quantity for providing at least one of a control emission signal and a modeling emission signal;

determining the at least one engine characteristics value from the engine characteristics map as a function of at least one of the control and model quantities to obtain at least one working point of the at least one process;

adapting the engine characteristics map to a changed process behavior of the at least one process for determining a first correction value for a preselected operating phase of the at least one process;

selecting a first support amount of the at least one support point positioned near the first correction value for adapting the engine characteristics map;

calculating a second correction value for each of the selected support points as a function of the first correction value for the first support amount of the at least one support point; and

adaptively distance-weighting the first support amount of the at least one support point as a function of a position of the first correction value in the engine characteristics map.

2. The method according to claim 1, wherein the at least one process includes at least one of a combustion process, a transmission switching process and a braking process.

3. The method according to claim 1, wherein the step of adapting the engine characteristics map includes selecting the at least one support point for interpolating the at least one respective engine characteristics value at the selected location of the first correction value.

4. The method according to claim 3, further comprising the step of selecting a first amount of the at least one support point positioned near the at least one working point for interpolating the at least one respective engine characteristics value, the first amount of the at least one support point being uniformly distributed around the at least one working point.

5. The method according to claim 4, wherein the first amount of the at least one support point corresponds to map dimensions of the engine characteristics map, the map dimensions corresponding to an operating amount of the operating quantities increased by 1, and further comprising the step of forming an interpolation support point group using the first amount of the nearest adjacent points of the at least one support point.

6. The method according to claim 1, wherein the step of adapting the engine characteristics map is performed only in a stationary operating phase of the at least one process.

7. The method according to claim 6, wherein the stationary operating phase is defined by maintaining the at least one working point within a limited region of the engine characteristics map for predefined minimum map accesses.

8. The method according to claim 7, wherein the limited region is formed by interpolating an interpolation support point group of the at least one working point.

9. The method according to claim 6, wherein the stationary operating phase is defined by maintaining the working point within at least two adjacent limited regions of the engine characteristics map for the predefined minimum map accesses.

10. The method according to claim 7, wherein the step of adapting the engine characteristics map is performed at a completion of the stationary operating phase.

11. The method according to claim 10, wherein the completion of the stationary operating phase is defined by removing the at least one working point from the limited region, the at least one working point being removed from the limited region after the predefined minimum map accesses in the limited region.

12. The method according to claim 9, the completion of the stationary operating phase is defined by removing the at least one working point from the two adjacent limited regions after the predefined minimum map accesses in the limited region.

13. The method according to claim 6, wherein a plurality of the first correction value is repeatedly computed for the engine characteristics map during the stationary operating phase, and wherein the plurality of the first correction value is combined to form a representative correction value.

14. The method according to claim 13, wherein the representative correction value is formed by at least one of averaging the plurality of the first correction value and averaging a respective position of each of the plurality of the first correction value being averaged.

15. The method according to claim 13, wherein the representative correction value is formed by low-pass filtering of the plurality of the first correction value.

16. The method according to claim 6, wherein a prespecified number of plurality of the first correction value are disregarded for combining at a beginning of the stationary operating phase.

17. The method according to claim 6, wherein the step of adapting the engine characteristics map includes selecting the at least one support point for forming an interpolation support point group to form a representative correction value.

18. The method according to claim 17, wherein the second correction value of the interpolation support point group of the representative correction value is determined according to:

space="preserve" listing-type="equation">Δ

Y.sub.i =γ

δ

ω

.sub.i Δ

Y.sub.ges

I=1, . . . , n+1,

wherein ##EQU2## γ

being a learning factor, Δ

Y_ges being the first correction value, and

d_j being distances between the at least one support point.

19. The method according to claim 1, wherein in the step of adapting the engine characteristics map, the first correction value is further determined by changing the at least one respective engine characteristics value of a predetermined amount of the at least one support point for determining if the engine characteristics map can be adapted, and wherein a support point density of the at least one support point is modified by at least one of displacing each of the at least one support point and modifying the predetermined amount of the at least one support point, the predetermined amount of the at least one support point being modified by at least one of adding the predetermined amount of additional support points and deleting the predetermined amount of the at least one support point, the support point density being modified if the engine characteristics map cannot be adapted.

20. The method according to claim 19, wherein the first correction value is provided to form at least one of an averaged absolute value and a low-pass-filtered absolute value for comparing with a predetermined limit value, the at least one respective engine characteristics value of the first support amount of the at least one support point being fixed if at least one of the averaged and low-pass-filtered absolute values exceeds the specified limit value.

21. The method according to claim 20, wherein the support point density in a region of the first correction value is increased by displacing the at least one support point of a interpolation support point group formed by the at least one support point, the at least one support point being displaced in a direction of a first correction value when the absolute value exceeds the specified limit value.

22. The method according to claim 21, wherein the at least one support point include displacements of the interpolation support point group of the first correction value, the displacements being calculated according to:

space="preserve" listing-type="equation">Δ

.sub.xi =inc sign(d.sub.i),

wherein d_i is a correction polarity.

23. The method according to claim 21, wherein the additional support points are generated at the corrective position of the first correction value in the engine characteristics map.

24. The method according to claim 19, wherein the predetermined amount of the at least one support point having a minimum information content is selected for being deleted.

25. The method according to claim 24, wherein the minimum information content is determined using an absolute error value obtained when at least one of omitting the at least one support point and interpolating one the predetermined locations of the at least one support point.

26. The method according to claim 24, wherein the predetermined amount of the at least one support point is modified only if the predetermined amount of the additional support points being added is greater than the predetermined amount of the support point being deleted.

27. The method according to claims 23, wherein the first correction value is positioned closer to one of the at least one support point, the first correction value being positioned closer than a predetermined minimum distance, the predetermined amount of the at least one support point not being added, the at least one support point including existing support points being adapted according to:

space="preserve" listing-type="equation">X.sub.i.sup.adapt =β

X.sub.i.sup.alt +(1-β

)X.sub.i.sup.neu

where I=1, . . . ,n 0≦

β

≦

1

space="preserve" listing-type="equation">X.sub.i.sup.adapt =β

Y.sub.alt +(1-β

)Y.sub.neu,

wherein β

is a forgetting factor, and X_i^alt, Y_alt, X_i^neu and Y_neu are the existing support points.

28. The method according to claim 27, wherein nearest-lying one of the at least one support point is corrected and another one of the at least one support point is not entered at the position of the first correction value when a ratio of the first correction value to the minimum distance to the nearest-lying one of the at least one support point exceeds a specified value.

29. The method according to claim 1, wherein the step of adapting the engine characteristics map includes determining if a plurality of the first correction value produces a common tendency, and determining at least one of an additive correction parameter and a multiplicative correction parameter being offset with a read-out map value during later map accesses.

30. The method according to claim 29, wherein the additive correction parameter is determined only from the plurality of the first correction value being formed in regions of the engine characteristics map in which an absolute value of the read-out map value is formed below a predefined upper limit.

31. The method according to claim 30, wherein the multiplicative correction parameter is determined only from the plurality of the first correction value being formed in the regions of the engine characteristics map in which the absolute value of the read-out map value is formed above a predefined lower limit.

32. The method according to claim 31, wherein the additive correction parameter is determined according to:

space="preserve" listing-type="equation">c.sub.0 (k)=c.sub.0 (k-1)+q0.sub.add Δ

Y(k)+q1.sub.add Δ

Y(k-1),

wherein q0_add and q1_add are control parameter constants of a proportional integral control algorithm, and Δ

Y(k) and Δ

Y(k-1) are deviations for map accesses.

33. The method according to claim 31, wherein the multiplicative correction parameter is determined according to:

space="preserve" listing-type="equation">c.sub.1 (k)=c.sub.1 (k-1)+q0.sub.mult Δ

Y(k)+q1.sub.mult Δ

Y(k-1),

wherein q0_mult and q1_mult are control parameter constants of a proportional integral control algorithm, and Δ

Y(k) and Δ

Y(k-1) are the deviations for map accesses.