Method and system for detecting and localizing sensor defects in motor vehicles
First Claim
1. A method for detecting and localizing sensor defects in a motor vehicle, the method comprising the acts of:
- determining a measuring signal (yk) for a description of a dynamic behavior of the motor vehicle by a sensor;
computing from a mathematical equivalent model, a state value (x−
k) which is assigned to the measuring signal (yk);
computing an estimation error (e−
k) of the state value (x−
k);
determining a residue (rk) from the difference between the measuring signal (yk) and a reference quantity (Ck*x−
k) corresponding to the measuring signal (yk) and formed from the state value (x−
k);
forming a characteristic parameter (ε
) by which sensor defects are detected and localized from the multiplication of the residue (rk) and the estimation error (e−
k); and
generating an error signal (EI, EL) when the characteristic parameter (ε
) exceeds a limit value (sk) to indicate that a defect in at least one of the pertaining sensors was detected (EI) and to indicate which one of the pertaining sensors is defective (EL).
8 Assignments
0 Petitions
Accused Products
Abstract
For detecting and localizing sensor defects in motor vehicles, a sensor determines a measuring signal for the description of the dynamic behavior of the motor vehicle. A state value assigned to the measuring signal is computed from a mathematical equivalent model. An estimation error of the state value is determined and a residue is determined from the difference between the measuring signal and a reference quantity which corresponds to the measuring signal and is formed from the state value. For clearly detecting and localizing the sensor defects, a characteristic parameter is formed by multiplying the residue and the estimation error, and an error signal is generated if the characteristic parameter exceeds a limit value.
23 Citations
25 Claims
-
1. A method for detecting and localizing sensor defects in a motor vehicle, the method comprising the acts of:
-
determining a measuring signal (yk) for a description of a dynamic behavior of the motor vehicle by a sensor;
computing from a mathematical equivalent model, a state value (x−
k) which is assigned to the measuring signal (yk);
computing an estimation error (e−
k) of the state value (x−
k);
determining a residue (rk) from the difference between the measuring signal (yk) and a reference quantity (Ck*x−
k) corresponding to the measuring signal (yk) and formed from the state value (x−
k);
forming a characteristic parameter (ε
) by which sensor defects are detected and localized from the multiplication of the residue (rk) and the estimation error (e−
k); and
generating an error signal (EI, EL) when the characteristic parameter (ε
) exceeds a limit value (sk) to indicate that a defect in at least one of the pertaining sensors was detected (EI) and to indicate which one of the pertaining sensors is defective (EL).- View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24)
-
3. The method according to claim 2, wherein the residue (rk) is computed from the measuring signal (yk) and the state value (x−
- k) according to the equation
-
4. The method according to claim 2, wherein an optimal filter estimation value (x+k) is determined from the rule
-
5. The method according to claim 1, wherein the residue (rk) is computed from the measuring signal (yk) and the state value (x−
- k) according to the equation
-
6. The method according to claim 5, wherein an optimal filter estimation value (x+k) is determined from the rule
-
7. The method according to claim 1, wherein an optimal filter estimation value (x+k) is determined from the rule
-
8. The method according to claim 7, wherein the amplification matrix (Kk) is a Kalman gain which is computed according to the equation
-
9. The method according to claim 8, wherein the estimation error (e−
-
k) is determined from the difference between the optimal filter estimation value (x+k−
1) and the prediction value (x−
k) according to the equation
-
k) is determined from the difference between the optimal filter estimation value (x+k−
-
10. The method according to claim 9, wherein the characteristic parameter (ε
- ) is determined from the equation
-
11. The method according to claim 7, wherein the estimation error (e−
-
k) is determined from the difference between the optimal filter estimation value (x+k−
1) and the prediction value (x−
k) according to the equation
-
k) is determined from the difference between the optimal filter estimation value (x+k−
-
12. The method according to claim 11, wherein the characteristic parameter (ε
- ) is determined from the equation
-
13. The method according to claim 1, wherein the limit value (sk) is adaptively calculated from the residue (rk) according to the rule
-
14. The method according to claim 1, wherein for several sensors the vector of measuring signals (yk) takes the form
-
15. The method according to claim 14, wherein a weighting matrix (Mk) has the form
wherein the coefficients (Mij) of the weighting matrix (Mk) are determined empirically. -
16. The method according to claim 15, wherein the vector of state values (x−
- k) has the form
-
17. The method according to claim 15, wherein the diagonal matrix (RMk), which contains the components of the residue (rk) on the main diagonal, has the form
-
k 0 0 0 0 0 r 2 k 0 0 0 0 0 r 3 k 0 0 0 0 0 r 4 k 0 0 0 0 0 r 5 k ] the estimation error (e−
k) only taking into account the differences of inherently dynamic state quantities.
-
-
18. The method according to claim 15, wherein an input vector (uk) has the form
-
19. The method according to claim 14, wherein the vector of state values (x−
- k) has the form
-
20. The method according to claim 19, wherein the diagonal matrix (RMk), which contains the components of the residue (rk) on the main diagonal, has the form
-
k 0 0 0 0 0 r 2 k 0 0 0 0 0 r 3 k 0 0 0 0 0 r 4 k 0 0 0 0 0 r 5 k ] the estimation error (e−
k) only taking into account the differences of inherently dynamic state quantities.
-
-
21. The method according to claim 19, wherein an input vector (uk) has the form
-
22. The method according to claim 14, wherein the diagonal matrix (RMk), which contains the components of the residue (rk) on the main diagonal, has the form
-
k 0 0 0 0 0 r 2 k 0 0 0 0 0 r 3 k 0 0 0 0 0 r 4 k 0 0 0 0 0 r 5 k ] the estimation error (e−
k) only taking into account the differences of inherently dynamic state quantities.
-
-
23. The method according to claim 22, wherein an input vector (uk) has the form
-
24. The method according to claim 14, wherein an input vector (uk) has the form
-
-
25. A system for detecting and localizing sensor defects in a motor vehicle, comprising:
-
sensors which feed measuring signals (yk) describing a dynamic behavior of the motor vehicle as input signals;
a computing unit which receives the measuring signals;
an evaluation unit coupled to an output side of the computing unit;
a computer readable medium for the computing unit and the evaluation unit, said computer readable medium having stored thereon program code segments that;
use a mathematical equivalent model in computing state values (x−
k) assigned to the measuring signals;
compute estimation errors (e−
k) of the state values;
determine residues (rk) from differences between the measuring signals and reference quantities corresponding to the measuring signals and formed from the state values;
form characteristic parameters (ε
) via multiplication of the residues and the estimation errors; and
generate error signals (EI, EL) when the characteristic parameters exceed limit values (sk) to indicate defects in pertaining sensors were detected (EI) and to indicate which of the pertaining sensors were defective (EL).
-
Specification