US 6,151,545 AGranted

Method for signal processing in a control system of motor vehicles

AssigneeWABCO GmbH

InventorsRainer Jollerichs, Thomas Lenz, Helmut Pannbacker, Henrich Riedemann, Detlef Schmidt

Life of this patent

Estimating the term…

Loading patent…

Abstract

A signal (S) has a value at every point in time. The values may change in time and may lie within or outside a predetermined basic value range (BVMIN, BVMAX). The signal (S) is processed so that control steps associated with the values of the signal (S) are carried out when certain values of the signal (S) occur outside the basic value range (BVMIN, BVMAX). It may then occur that the carrying out of the control steps may be triggered in an undesirable manner because of implausible values of the signal (S). Such an implausible signal behavior must be recognized rapidly and the undesirable effects of the implausible signal behavior must be decreased or totally prevented at an early moment. For this purpose a value of the signal (S) occurring with relative frequency is determined as a dominant value (DV) as well as a frequency value (FV), which indicates frequency of occurrence of values of the signal (S) situated in the vicinity of the dominant value (DV). The execution of the control steps associated with the given values of signal (S) is then in addition rendered dependent on the magnitude of the deviation of the dominant value (DV) from the basic value range (BVMIN, BVMAX) when the frequency value (FV) reaches at least a predetermined magnitude (F1, F2). A preferred area of application of the invention is electronic control and regulating systems in vehicles.

A method for signal processing carried out in a vehicle control system, comprising the steps of:

(a) determining the values of at least one first signal (S), which has a face value at any point in time, whereby the values change in time,
(b) locating the values of the first signal (S) within as well as outside a basic value range (BVMIN, BVMAX) determined for the first signal (S),
(c) determining a value of the first signal (S) appearing with relative frequency as a dominant value (DV),
(d) determining a frequency value (FV) which indicates the frequency of the occurrence of the values in the proximity of the dominant value (DV) of the first signal (S),
(e) carrying out control steps as a function of the occurrence of certain values of the first signal (S) outside the basic value range (BVMIN, BVMAX) and taking into account the dominant value (DV) and the frequency value (FV), in such manner that the carrying out of the control steps depends on the magnitude of deviation of the dominant value (DV) from the basic value range (BVMIN, BVMAX) when the frequency value (FV) reaches at least a predetermined magnitude,
wherein:
(f) in order to evaluate the first signal (S), time segments which follow each other in time are determined,
(g) a time segment may either run its course normally or be terminated early,
(h) the value of the first signal (S) which occurs first in a time segment is used as a comparison value (CV) for this time segment,
(i) a value range which surrounds the comparison value (CV) is selected as a comparison value range, and
(j) wherein at the occurrence of values of the first signal (S) outside the comparison value range, a time segment is terminated early,
and wherein,
(k) only comparison values (CV) found in normally ended time segments are taken into consideration,
(l) a first comparison value taken into consideration is used as a preferred value (PV),
(m) a value range which surrounds the preferred value (PV) is selected as the preferred value range
(n) wherein the frequency value (FV) is derived from a number of further comparison values (CV) taken into consideration and which are within the preferred value range, and
(o) wherein the dominant value (DV) is identical with the preferred value (PV).

14 more elements

Abstract

Claims16