DE102020134605A1 - METHOD AND DEVICE FOR DETERMINING A TIME-VARYING CONTINUOUS MEASUREMENT AND VEHICLE - Google Patents

Abstract

The invention relates to a method for determining a time-varying, continuous measured variable, in which the measured variable is measured repeatedly with a measuring system, for example an optical sensor, and a moving average is formed from a number (averaging depth) of measured values of the measured variable. The method is characterized in that (a) a current measured value and at least one previous measured value are evaluated, (b) depending on the result of the evaluation in step (a), an averaging depth and/or a weighting of the measured values over which the average is made, is determined and (c) that the moving average is formed with the averaging depth determined in step (b) and/or the weighting determined in step (b). The invention also relates to a device for determining a time-varying, continuous measured variable and a vehicle.

Description

In a first aspect, the present invention relates to a method for determining a time-varying continuous measured variable, in which the measured variable is measured repeatedly with a measuring system, for example an optical sensor, and a moving average is formed from a number (averaging depth) of measured values of the measured variable.

In a further aspect, the present invention relates to a device for determining a time-varying, continuous measured variable, with a measuring system for repeatedly measuring the measured variable and with a control and evaluation unit which is set up to calculate a sliding one from a number (averaging depth) of measured values of the measured variable to form the mean.

In many applications, physical measured variables are recorded via sensors. The acquisition of the measurand is typically disturbed by noise. It is known (Poisson statistics) that in the case of repeated measurements, i.e. measurements in which the same measurement is carried out several times in succession in the same measurement situation, the noise can be reduced by a factor of √N by forming a moving average over N measured values. The number N of measured values to be averaged is referred to as the averaging depth. A typical case for a repetitive measurement, in which sliding averaging is used, is the distance measurement with an optical time-of-flight sensor to an object in a surveillance area.

In practical applications, continuous measured variables are often recorded. Such continuous measured variables are distinguished by the fact that they do not change suddenly due to the properties of the observed system. The measured variable as a function of time and also the first and second derivation of these measured variables with respect to time are therefore essentially continuous.

Digital processing of the measured values naturally results in a quantization of the measured values, which is all the finer the better the resolution of the analog-to-digital converter used. In the situations considered here, however, the measurement inaccuracy in a single measurement (noise) is generally greater than the resolution of the analog-to-digital converter used.

An example of a continuous metric is again measuring the distance to an object with a 1D range finder. The observed object can be stationary relative to the 1D range finder or moving at constant speed or accelerated. Because of its mass, however, the object cannot make any abrupt changes in its position if the sampling rate is high enough. This property applies to many practical applications.

With a high averaging depth, the noise can be significantly reduced in the case of slowly changing or static measured variables, for example in the case of a stationary object. The situation is more difficult for rapidly changing measured variables, such as objects moving relative to the sensor. The higher the averaging depth, the more the noise is reduced. Due to the different times of the measurements, which are averaged, there is a deviation of the averaged measured value compared to the real measured variable, which is referred to as lag.

So-called Kalman filters are known in the prior art. These are predictive filters that, based on a system model, discard erroneous measured values and, if necessary, extrapolate missing measured values.

Methods are known in financial mathematics that smooth out market movements and at the same time are intended to react to changes at an early stage. An example of this is Kaufman's Adaptive Moving Average.

An object of the invention can be seen as specifying a method, a device and a vehicle in which the quality of the measurement data is increased depending on the respective application.

This object is achieved by the method having the features of claim 1, the device having the features of 18 and the vehicle having the features of claim 22.

Preferred variants of the method according to the invention and advantageous refinements of the device according to the invention and of the vehicle according to the invention are explained below, in particular with reference to the dependent claims and the figures.

The device according to the invention can be set up in particular to carry out the method according to the invention.

The method of the type specified above is further developed according to the invention in that (a) a current measured value and at least one previous measured value are evaluated, (b) depending on the result of the evaluation in step (a) an averaging depth and/or a weighting of the measured values being averaged is specified and (c) that the moving average is formed with the averaging depth specified in step (b) and/or the weighting specified in step (b).

The device of the type specified above is further developed according to the invention in that the control and evaluation unit is also set up to (A) evaluate a current measured value and at least one previous measured value, (B) depending on the result of the evaluation in step (A). to determine the averaging depth and/or a weighting of the measured values that are averaged, and (C) to form the moving average with the averaging depth determined in step (B) and/or the weighting determined in step (B).

A device according to the invention is present in a vehicle according to the invention.

Time-varying continuous measured variables are those measured variables that vary over time, i.e. are not constant, and can assume continuous values (in contrast to measured variables that can only assume certain, for example integer, values).

The term measuring system should be understood to mean all components that are necessary for measuring the measurand. In the case of an optical sensor, these are, for example, the light transmitter, the light receiver and suitable means for actuating and/or controlling these components as intended.

The control and evaluation unit can be, in particular, programmable, electronic components of a known type. For example, microcontrollers or comparable programmable components with analog and digital functions can be used.

The term current measured value is intended to refer to the most recent measured value that was measured with the measuring system.

Previous measured values refer to measured values that were measured in the past. A preceding measured value can in particular be a directly preceding measured value, ie the last measured value measured before the current measured value.

Only those measured values are referred to as previous measured values which, apart from the fact that the measured value itself can change, are collected in the same measurement situation as a current measured value. This is important, for example, with a two-dimensional optical scanner that can measure distances in different spatial directions. It is clear that with such a measuring device, only those measured values are used for averaging and in this sense can be previous measured values that are collected in one and the same spatial direction, and therefore in the same measurement situation.

In the context of the present description, the term “evaluation” should be understood to mean any type of quantitative evaluation and/or processing of the measurement data and possibly other data, in particular for the purpose of determining the averaging depth and/or the weighting for forming the moving average.

In the context of the present description, the term “vehicle” should be understood to mean all devices whose intended use consists of moving in a defined manner relative to the environment that can be controlled while driving. For example, they can be vehicles, in particular driverless vehicles, that are used in factory automation, such as stacker cranes.

Within the scope of the present description, the term sensor can be used to designate both the measurement components, ie the measurement system, and the device according to the invention itself.

As a basic idea of the present invention can be considered, in the generic method and the generic device, the parameters for the formation of a moving average, in particular the averaging depth and / or the weighting, depending on the respective measurement situation and in particular depending on a current measured value and from set previous measurement data.

In the method according to the invention, the measured variable is evaluated in order to adapt the averaging depth and/or the weighting to a particular application situation, in particular in an automated manner.

The measured values over which the moving average is formed according to the invention can contain in particular the current measured value and a number of previous measured values, optionally after application of a weighting function, more on this below.

The invention enables adaptive averaging for continuous and noisy measured variables. By adapting the parameters for the formation of the sliding according to the invention mean value, significant improvements in the quality of the processed measurement data can be achieved for many applications.

A further significant advantage can be seen in the fact that the method according to the invention and the device according to the invention make it easier to determine an appropriate compromise between reducing the noise on the one hand and reducing the tracking delay on the other for applications in which both static and dynamic situations occur .

A more specific adaptation of the parameters for the formation of the moving average is possible in a preferred variant of the method according to the invention, in which a history of the last M measured values before the current measured value is also evaluated in step (a). Each of the last M measured values can be evaluated. However, it is also possible to evaluate the last M measured values collectively.

For example, in step (a) an, in particular weighted, mean value can also be calculated over the last N measured values before the current measured value. N can be equal to M in particular.

In a further preferred variant of the method according to the invention, a noise of the measured values is determined from the previous measured values and also taken into account or evaluated in step (a). The noise provides an important guide to the choice of averaging depth. For example, there is no point in increasing the averaging depth when a change in readings over time is greater than the noise.

In a further advantageous variant of the method according to the invention, if the determined noise of the measured values is less than a specific upper limit to be defined, a derivation of the time profile of the measured values is formed and taken into account or evaluated in step (a). Typically, a control algorithm can work qualitatively in such a way that the depth of averaging is chosen to be smaller the larger the derivative of the time profile of the measured values is, ie the faster the measured values change. From a consideration of the first derivative of the measured variable, a change in the measured variable can also be detected at an early stage.

In this context, in a further important variant of the method according to the invention, a lag delay is determined from a number of previous measured values and also taken into account or evaluated in step (a). For example, a histogram can be formed from a number of previous measured values and the lag can be determined from the histogram. It is interesting, for example, whether or to what extent a distribution of the previous measured values is symmetrical to a mean value.

For example, in step (b) of the method according to the invention, the averaging depth can be adjusted in such a way that a lag is the same as the noise of the measured values.

In the case of a dynamic measurement variable, for example the distance of a moving object, the averaging depth can advantageously be selected in such a way that the lag does not significantly exceed the noise. In the case of a rapidly changing measurand, the noise can often be neglected compared to the change in the measurand, so that an averaging can be dispensed with in this case.

In the case of a largely static measured variable, the averaging depth can therefore be massively increased in order to significantly reduce the noise. In the case of rapid changes in the measured variable, on the other hand, the averaging depth can be reduced to 1 in order to minimize the lag.

In the case of slow changes in the measured variable, the averaging depth can be adjusted in such a way that the lag is in the order of magnitude of the noise of the measured values.

In addition to the averaging depth, the concept of weighting comes into play when the individual measured values used to form the mean values are each multiplied by a specific weighting factor. The concepts of averaging depth and weighting overlap to the extent that, given the same weighting of the measured values, different averaging depths can be viewed as weighting of the measured values with gate functions of different widths (the broader the gate function, the greater the averaging depth). In principle, there is freedom of design when choosing the appropriate weighting of the measured values.

In a simple variant of the method according to the invention, all previous measured values that are used for the sliding mean value are weighted equally when the mean value is formed.

In step (b) of the method, only the number N of previous measured values to be used for forming the moving average is then specified.

In principle, a function with which the measured values that are used to form the moving average are weighted can be of a largely arbitrary nature.

In variants of the method, measured values can be weighted all the more heavily the younger they are, ie the shorter the time since the relevant measurement times were obtained. For example, the measured values that are used for the sliding mean value can be weighted with a function when the mean value is formed, which function decreases with the time since the measuring times of the measured values to be weighted.

The more recent measured values are preferred even more if the measured values are weighted with a function which decreases exponentially with the time since the measurement times of the measured values to be weighted.

In a further variant of the method according to the invention, not all measured values from a previous time interval are used to form the moving average, but only a selection, for example every second, third or nth measured value; n is a natural number. This saves system resources such as storage space or computing capacity. This selection can also be designed in such a way that the time interval between the measured values used for the averaging varies, for example increases the longer the measured values are in the past. For practical reasons, the measured values are generally taken at times that are equidistant from one another, i. H. with a certain clock frequency. The evaluation and evaluation of the data is then comparatively simple.

Equidistant measurement times are not absolutely necessary, however. In principle, the measured values can be collected one after the other at points in time that have time intervals that vary from one another. This can then be taken into account in the evaluation, for example in the weighting of the measured values with a weighting function that depends on the time since a measurement point in time.

For many applications it can be assumed as a good approximation that the measurand is continuous and the noise of the measurement system is symmetrical and that an upper limit of the noise can be determined. The method according to the invention can in particular evaluate the current measured value, a history of the last M measured values and a mean value calculated over N measured values in order to decide on a change in the averaging depth (retain, increase, reduce).

In a preferred variant of the method according to the invention, the averaging depth is set to a maximum value if the measured value differs by less than a lower difference value to be defined from a, in particular immediately, previous measured value, several previous measured values or a function, in particular an average value, from several previous measured values . The lower difference value to be defined can preferably be selected to be as large as an upper limit of the noise. This variant of the method corresponds to a simple threshold value analysis.

Alternatively or additionally, in a further advantageous embodiment of the method according to the invention, the averaging depth can be set to a minimum value, in particular to 1, if the measured value has changed by more than an upper difference value to be defined from one, in particular immediately, previous measured value, several previous measured values or one Function, in particular an average value, differs from several previous measured values. An averaging depth of 1 means that the current measured value is output as is and no actual mean value is formed.

Alternatively or additionally, the averaging depth can be set variably, depending on how quickly the measured values vary. Variants of the method are therefore also preferred in which the averaging depth is determined variably depending on a difference in the measured value from one, in particular immediately, previous measured value, several previous measured values or a function, in particular an average value, from several previous measured values if the difference is greater is than the lower differential value and smaller than the upper differential value, the averaging depth expediently decreasing as the difference increases. In other words, the averaging depth can be set to be greater, the less the measured value differs from the previous measured values.

For some applications it may be important that a tracking error caused by averaging is constant even if the averaging depth changes. This can be achieved by always placing the averaging window with a dynamic width (dynamic averaging depth) symmetrically around a fixed point in time in the measured value history. When averaging over a maximum of M values (max. measurement value window) at an acquisition rate of F (n measurement values per time unit), the measurement values can be measured with the constant delay (M/2)/F from are given. This variant of the method according to the invention is therefore characterized in that an averaging window with a variable width is placed symmetrically around a fixed point in time in the averaging history.

A further group of variants of the method according to the invention is characterized in that additional information or data about the system, which directly or indirectly describe the measured variable, is used to set the averaging depth and/or the weighting. In particular, this can be measured data from other sensors that are used to measure the same measured variable (e.g. in a redundant arrangement of several identical sensors) or other measured variables, e.g. acceleration, rotational position. Typical practical applications are situations in which the sensors in question relate to a moving part of the application, for example a moving target, a moving sensor, for example on a vehicle according to the invention. These advantageous variants of the method according to the invention are characterized in that measurement data from another measurement system, in particular another sensor, for example an acceleration sensor or a rotary encoder, is used to set the averaging depth and/or the weighting of the previous measured values, which are used for averaging will.

For example, a vehicle according to the invention can have at least one additional sensor, in particular an acceleration sensor and/or a rotary encoder, whose measurement data is/are also used to determine the averaging depth.

Expediently, control signals that influence the measured variable, for example control signals to an actuator installed in the system, can also be included in the decision-making process regarding the averaging depth and/or weighting. Preferred embodiments of the method in this context are characterized in that an activity of, in particular mechanically active components, which are present in a moving object on or in which the measuring system is also arranged, in particular actuators, when determining the averaging depth and /or the weighting of the measured values to be used for the moving average according to step (b) is taken into account or evaluated.

In practice, discontinuous changes, i.e. jumps, in the measured value can occur sporadically, in particular due to external influences, even in the case of continuous measured variables. For example, this is the case with 1D rangefinders when the observed object is covered by an object moving into the measuring beam. Such jumps can be recognized and treated as a special case by suitable programming. If such a jump is detected, the averaging can be restarted, if necessary with adjustment of the averaging depth. For example, the last reading can be held until a new mean value from the target is available.

Furthermore, external sources of interference can lead to individual incorrect measurements in real systems (in the case of distance sensors, e.g. rain, dust, etc.).

In advantageous variants of the method according to the invention, the averaging depth can, for example, only be adjusted when a change in the measured values is detected with sufficient reliability and stability (e.g. for a certain number of consecutive measurements). Individual outliers of the measured values can be preferably filtered out.

In advantageous variants of the method according to the invention, a plausibility check can be carried out for the measured values and for measured values that are recognized as implausible, the last preceding mean value recognized as plausible can be output.

In principle, the method according to the invention and the device according to the invention can be used for any time-variable and continuous measured variables. The device according to the invention can advantageously be designed as an optical sensor for determining a distance to an object in a monitored area, with a transmitter for emitting transmitted light into the monitored area and with a receiver for detecting light reflected back from the monitored area, the control and evaluation unit is set up to control the transmitter and the receiver and to evaluate the light detected by the receiver and to determine a distance of the object from the sensor as a measured variable.

All light sources that provide the light to be used with the desired wavelength and the desired intensity can be used as transmitters. Semiconductor components, in particular light-emitting diodes, VCSELS or semiconductor lasers, are preferably used in the visible range or infrared range.

All components can be used as receivers, which sufficient sensitivity to detect the reflected light in the area. In turn, semiconductor components, in particular photodiodes, for example in CMOS or CCD technology, are preferably used.

The device according to the invention can be designed in particular as an optical sensor according to the transit time principle, with the transmitter for emitting light pulses into the monitored area and with the receiver for detecting light pulses reflected back from the monitored area, with the control and evaluation unit being set up to evaluate the signals from the receiver detected light pulses and for determining the distance of the object from the sensor from a transit time of the light pulses reflected back from the object.

In principle, the present invention can advantageously be used in optical 1D range finders based on the transit time principle. Also advantageous is an embodiment of the device according to the invention, which is designed as an optical 2D scanner with a rotor and a stator. For example, it can be a 2D scanner that EP 2237063 A1 kind of act. In this respect, the contents of this document are intended to form part of the subject matter of the present application.

• 1: eine schematische Darstellung eines ersten Ausführungsbeispiels einer erfindungsgemäßen Vorrichtung;
• 2: eine schematische Darstellung eines zweiten Ausführungsbeispiels einer erfindungsgemäßen Vorrichtung und
• 3: eine schematische Darstellung eines dritten Ausführungsbeispiels einer erfindungsgemäßen Vorrichtung.

Further advantages and features of the present invention are described in connection with the accompanying figures. It shows:

• 1 : a schematic representation of a first exemplary embodiment of a device according to the invention;
• 2 : a schematic representation of a second embodiment of a device according to the invention and
• 3 : a schematic representation of a third exemplary embodiment of a device according to the invention.

Components that are the same and have the same effect are generally identified in the figures with the same reference symbols.

This in 1 The schematically illustrated first exemplary embodiment of a device 100 according to the invention for determining a time-varying, continuous measured variable has, as essential components, a measuring system 20, 40 for repeatedly measuring the measured variable d and a control and evaluation unit 50, which is set up to calculate from a number of previous measured values of the measure to form a moving average.

Specifically, device 100 is embodied as an optical sensor for determining a distance d (measured variable) from an object 10 in a monitored area 30, and the measuring system has a transmitter 20 for emitting transmitted light 22 into the monitored area 30 and a receiver 40 for detecting from the Monitoring area 30 reflected light 24 on. The control and evaluation unit 50 is set up to control the transmitter 20 and the receiver 40, to evaluate the light detected by the receiver 40 and to determine a distance d of the object 10 from the sensor as a measured variable.

In particular, the device 100 can be an optical sensor based on the transit time principle. The transmitter 20 is then used to emit light pulses 26 into the monitored area 30 and light pulses reflected back from the monitored area 30 , in particular by the object 10 , are detected with the receiver 40 .

The control and evaluation unit 50 is then additionally set up to evaluate the light pulses detected by the receiver 40, in particular their time of receipt, and to determine the distance d of the object 10 from the sensor from the propagation time of the light pulses radiated back from the object 10. The transit time is the difference between the time t2 when a light pulse is received and the time t1 when it is emitted. The distance d of the object can then be calculated using the speed of light c: d = c(t2-t1)/2. The technologies that are used here are basically known.

The transmitter 20, the receiver 40 and the control and evaluation unit 50 are in the in 1 example shown housed in a housing 60. Connection lines 52 shown schematically can be used for data exchange, ie in particular for outputting the measured values and/or the sliding mean value of the measured values and also for supplying energy.

According to the invention, the control and evaluation unit 50 is also set up to carry out a method according to the invention. Specifically, control and evaluation unit 50 is set up to (A) evaluate a current measured value and at least one previous measured value, (B) depending on the result of the evaluation in step (A), an averaging depth and/or a weighting of the previous measured values via which is averaged, and (C) forming the moving average with the averaging depth specified in step (B) and/or the weight specified in step (B).

In detail, all variants of the method according to the invention described above can be used.

2 shows a second embodiment of a device 200 according to the invention. The only difference to the device 100 in 1 is that in the housing 60 there is also an acceleration sensor 70 whose measurement data are also used to determine the averaging depth and/or the weighting of the measurement values for the formation of the moving average.

3 shows an exemplary embodiment of a vehicle 400 according to the invention. The vehicle 400 according to the invention has a device 300 according to the invention, which is a 2D scanner with a rotor 90 and a stator 94 in the example shown. The device 300 according to the invention can be used to protect the vehicle 400 from collisions. The transmitter 20 and the receiver 40 are arranged on the rotor 90 . Also located on the stator 94, shown schematically, is a rotary encoder 80, with which the rotary position of the rotor 90 relative to the stator 94 can be measured. The transmitter 20 and the receiver 40 are connected to the control and evaluation unit 50 for data and energy exchange. This can be done, for example, via an optical path that is formed in a hollow shaft 92 that is coaxial with the axis of rotation. The technologies used here are known. In the housing 60 there is also, as in 2 , an acceleration sensor 70 which, like the rotary encoder 80, is connected to the control and evaluation unit 50.

During operation, the rotor 90 rotates rapidly in a basically known manner about the axis of rotation which is coaxial with the hollow shaft 92 . The distances to the objects present in different directions are measured in each case. in the in 3 In the situation shown, the transmitter 20 and the receiver 40 aim at an object 10 in the surveillance area 30 at an angle φ (azimuth angle) to the horizontal.

According to one of the methods according to the invention, the control and evaluation unit 50 determines an averaging depth based on the measurement data of the optical sensor 300 and outputs the averaged measurement values. In particular, the measurement data 70 from the acceleration sensor 70 and/or the rotary encoder 80 can also be used to determine the averaging depth/or the weighting of the measured values.

Reference List

10

object

20

light transmitter

30

surveillance area

40

recipient

50

Control and evaluation unit

52

Output/input of the control and evaluation unit 50

60

Housing

70

accelerometer

80

encoder

90

rotor

92

Hollow shaft/axis of rotation of the rotor 90

94

stator

100

device according to the invention

200

device according to the invention

300

device according to the invention

400

vehicle according to the invention

i.e

Distance object - sensor

φ

azimuthal angle

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• EP 2237063 A1 [0064]

Claims (23)

A method for determining a time-varying, continuous measured variable, in which the measured variable is repeatedly measured using a measuring system, for example an optical sensor, and a moving average is formed from a number (averaging depth) of measured values of the measured variable, characterized in that (a) a current measured value and at least one previous measured value are evaluated, (b) that depending on the result of the evaluation in step (a), an averaging depth and/or a weighting of the measured values, over which the averaging is carried out, is determined and (c) that the moving average with the in Step (b) specified averaging depth and / or the in step (b) specified weighting is formed. procedure after claim 1 , characterized in that in step (a) a history of the last M measured values before the current measured value is also evaluated. procedure after claim 1 or 2 , characterized in that in step (a) an, in particular weighted, mean value is also calculated over the last N measured values before the current measured value. Procedure according to one of Claims 1 until 3 , characterized in that a noise of the measured values is determined and taken into account or evaluated in step (a). procedure after claim 4 , characterized in that a derivation of the time profile of the measured values is formed and taken into account or evaluated in step (a) if the noise is less than an upper limit to be defined. Procedure according to one of Claims 1 until 5 , characterized in that a lag delay is determined from a number of measured values and evaluated in step (a). Procedure according to one of Claims 1 until 6 , characterized in that all measured values that are used for the sliding mean value are weighted equally when the mean value is formed. Procedure according to one of Claims 1 until 6 , characterized in that the measured values that are used for the moving average are weighted with a function that decreases, in particular linearly or exponentially, with the time that the measurement times of the measured values to be weighted are in the past. Procedure according to one of Claims 1 until 6 , characterized in that only a selection of the measured values obtained in a previous time interval, in particular only every second, third or nth measured value, is used for forming the moving average. Procedure according to one of Claims 1 until 9 , characterized in that in step (b) the averaging depth is adjusted in such a way that a lag is equal to the noise of the measured values. Procedure according to one of Claims 1 until 10 , characterized in that the averaging depth is set to a maximum value if the current measured value differs by less than a lower difference value to be defined from one, in particular directly, previous measured value, several previous measured values or a function, in particular an average value, from several previous measured values . Procedure according to one of Claims 1 until 11 , characterized in that the averaging depth is set to a minimum value, in particular to 1, if the current measured value differs by more than an upper difference value to be defined from one, in particular immediately, previous measured value, several previous measured values or a function, in particular an average value, differs from several previous readings. Procedure according to one of Claims 1 until 12 , characterized in that the averaging depth is set variably depending on a difference between the current measured value and a, in particular immediately, previous measured value, several previous measured values or a function, in particular an average value, from several previous measured values if the difference is greater than the lower difference value and is less than the upper difference value, with the averaging depth decreasing as the difference increases. Procedure according to one of Claims 1 until 13 , characterized in that an averaging window with variable width is placed symmetrically around a fixed point in time in the averaging history. Procedure according to one of Claims 1 until 14 , characterized in that measurement data from a further measuring system, in particular a further sensor, for example an acceleration sensor (70) or a rotary encoder (80), for setting the averaging depth and / or the weighting of the measured values that are averaged. Procedure according to one of Claims 1 until 15 , characterized in that an activity of, in particular mechanically effective components, which are present in a moving object (200, 400) on or in which the measuring system (20, 40) is also arranged, in particular of actuators, when determining the averaging depth and/or the weighting of the measurements to be used for the moving average according to step (b). Procedure according to one of Claims 1 until 6 , characterized in that a plausibility check is carried out for the measured values and in that for measured values which are recognized as implausible, the last preceding mean value recognized as plausible is output. Device for determining a time-varying continuous measured variable, with a measuring system (20, 40) for repeatedly measuring the measured variable and with a control and evaluation unit (50) which is set up to calculate a moving average from a number (averaging depth) of measured values of the measured variable to form, characterized in that the control and evaluation unit is also set up to (A) evaluate a current measured value and at least one previous measured value, (B) depending on the result of the evaluation in step (A) an averaging depth and / or to determine the weighting of the measured values to be averaged and (C) to form the moving average with the averaging depth determined in step (B) and/or the weighting determined in step (B). device after Claim 18 , which is designed as an optical sensor for determining a distance to an object (10) in a monitored area (30), with a transmitter (20) for emitting transmitted light (22) into the monitored area (30) and with a receiver (40) for detecting light (24) reflected back from the monitored area (30), the control and evaluation unit (50) being set up to control the transmitter (20) and the receiver (40) and to evaluate the light detected by the receiver (40). and for determining a distance (d) of the object (10) from the sensor as a measured variable. device after claim 19 which is designed as an optical sensor based on the transit time principle, with the transmitter (20) for emitting light pulses (26) into the monitored area (30) and with the receiver (40) for detecting light pulses (28 ), wherein the control and evaluation unit (50) is set up to evaluate the light pulses detected by the receiver (40) and to determine the distance (d) of the object (10) from the sensor from a transit time of the object (10) reflected back light pulses. device after claim 19 or 20 , which is designed as a 2D scanner with a rotor (90) and a stator (94). Vehicle with a device according to one of claims 18 until 21 . vehicle after Claim 22 , characterized in that there is at least one further sensor, in particular an acceleration sensor (70) and/or a rotary encoder (80), whose measurement data is/are also used to determine the averaging depth.

Images