Data processing: vehicles – navigation – and relative location – Vehicle control – guidance – operation – or indication
Reexamination Certificate
2003-01-21
2003-11-11
Camby, Richard M. (Department: 3661)
Data processing: vehicles, navigation, and relative location
Vehicle control, guidance, operation, or indication
C701S036000, C381S071100
Reexamination Certificate
active
06647324
ABSTRACT:
FIELD OF THE INVENTION
The invention relates to a method for digitally filtering a signal burdened with noise for the purpose of filtering out the noise components, especially to determine a zero crossover of the signal. The invention also relates to a method for controlling the driving dynamic of a vehicle and a corresponding computer program and control system.
BACKGROUND OF THE INVENTION
Various methods for digitally filtering are known from the state of the art, especially the so-called Butterworth filter. Furthermore, so-called adaptive and predictive filters are known from the state of the art.
Also known from the state of the art are digital filters for the lowpass filtering of signals burdened with noise for filtering out the noise components. A common disadvantage of such digital filters is the phase shift of the filtered signal generated by the filtering with reference to the original signal. This phase shift is especially disadvantageous for control applications because a precise phase position is important. This applies also to control applications for vehicles, especially motor vehicles.
German patent publication 4,112,007 discloses a system for forming a signal for the chassis control of a vehicle wherein the relative movements between the vehicle body and the wheels are detected. Proceeding from this, the predicted spring deflection velocity at a future time point is determined. Based on this, the control of a vehicle variable is undertaken such as the control of adjustably designed dampers. The object is to adjust the damper characteristic in operating phases of lesser damping forces of the damper in order to minimize the adjusting noise of the damping.
It is here disadvantageous that a Taylor series development is used for predicting the spring deflection velocity. Such a Taylor series development is based, by definition, on derivations which intensify the noise component in the signal. For this reason, the series development must be interrupted early in this application so that only a very small number of support points can be considered. Furthermore, in this method, it is assumed that the considered support points are “true” (that is, not noise-burdened) signal values so that measuring inaccuracies lead to massive effects on the prediction.
SUMMARY OF THE INVENTION
It is an object of the invention to provide an improved method for digitally filtering a noise-containing signal by filtering out the noise component. It is a further object of the invention to provide an improved method for controlling the driving dynamic of a vehicle and to provide a corresponding computer program and control system.
The method of the invention for digitally filtering a noise-containing signal is for filtering out components of noise. The signal has a fundamental frequency and the method includes the steps of: inputting the noise-containing signal into a digital filter having a filter length (K) less than the period of the fundamental frequency; filtering out a signal component of an order (m) from the signal based on earlier scanning values of the signal within the filter length (K); and, outputting a filtered signal value.
A special advantage of the invention is that the filtered signal has only a slight or no phase shift with reference to the unfiltered noise-burdened signal. This applies especially to the filtering of the linearly approximated signal trace so that the time point of the crossover of the noise-burdened signal through a specific signal level can be especially precisely determined by evaluating the filtered signal. In this way, especially the time point of the zero crossover can be precisely determined.
A further special advantage of the invention is that an FIR filter (finite-pulse response) can be used so that the filter is stable in every case.
According to a preferred embodiment of the invention, an FIR filter of a specific filter length is used which is significantly less than the period of the fundamental frequency of the signal to be filtered. For example, the filter length is selected as ⅕ or as {fraction (1/10)} of the period of the fundamental frequency.
This is especially advantageous for such signals which are based on a vibration system since these signals, in general, contain a strong harmonic component of the fundamental frequency and this harmonic component can be linearly approximated in the region of the zero crossover. For such signals, a corresponding design of the filter for filtering out the linear signal component is advantageous so that the zero crossover of the signal can be determined especially accurately.
According to another preferred embodiment of the invention, the filtering takes place via interpolation based on earlier scanning values within a window length pregiven by the filter length. Alternatively or in addition thereto, the filtering takes place via extrapolation based on the same scanning values in order to arrive at an extrapolated future signal value. The extrapolation can be especially used to estimate future signal values, for example, the zero crossover or to compute signal values for specific earlier or future time points.
According to a further preferred embodiment of the invention, the filtering of a nonlinear signal component takes place such as a signal component of quadratic or higher order. For filtering out signal components of quadratic order, a quadratic interpolation algorithm is used in order to filter out corresponding nonlinear signal traces.
According to a further preferred embodiment of the invention, the filter length for filtering out a linear signal component is so selected that the signal trace of the signal, which is to be filtered, has essentially a linear trace within a relevant range. For a vibration signal, this applies to the region about the zero crossover of the signal and this crossover can be approximated via a linear function.
According to a preferred embodiment of the invention, the method of the invention for digitally filtering is used as a basis for the control of one or several driving-dynamic parameters of a vehicle. For the control of the damping, for example, the input signals from elevation position sensors are detected at specific scanning time points and the corresponding noise-containing signal is filtered and supplied to the control algorithm. For example, the control of the damping characteristic takes place adaptively based on the detection of the zero crossover of the signal.
REFERENCES:
patent: 5416844 (1995-05-01), Nakaji et al.
patent: 5559893 (1996-09-01), Krokstad et al.
patent: 5875411 (1999-02-01), Volkart et al.
patent: 6185300 (2001-02-01), Romsburg
Creutzburg Uwe
Hoinkaus Hermann
Rieckmann Andreas
Camby Richard M.
Continental Aktiengesellschaft
Ottesen Walter
LandOfFree
Method for digitally filtering a signal burdened with noise... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for digitally filtering a signal burdened with noise..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for digitally filtering a signal burdened with noise... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3137065