DE102020205767A1 - Method for determining at least one piece of information about a soil - Google Patents

Abstract

A method for determining at least one item of information on a floor (112) with an arrangement (100) which can be moved relative to the floor (112) by means of a drive (104), the drive (104) being controlled with a drive signal which has a course with a periodic component which causes a torque (106) with a periodic course to be impressed on at least one drive wheel (102) of the arrangement (100), this impressing causing the arrangement (100) to oscillate, with the oscillation characterizing signal is detected and evaluated in order to determine the at least one piece of information.

Description

The invention relates to a method for determining at least one item of information about a floor and an arrangement for carrying out the method.

State of the art

Information about a floor is understood here to mean both the nature of the floor and a distance from this floor. This also means that the geometric properties of the soil are understood. The acquisition of the information thus enables the properties of the soil to be determined.

The determination of the nature of a floor or a floor area as a property of the floor is regularly of importance in those cases in which this floor is subsequently to be treated, in particular cleaned. The texture includes both properties of the soil, such as the hardness of the soil, in particular of its surface, as well as the material used, which in turn affects the properties of the soil. For example, a distinction is made between a stone floor, a wooden floor and a carpeted floor, which are made differently, i.e. H. which have different properties. On the basis of the specific condition, conclusions can then be drawn as to the type of floor or the floor type, for example wood, carpet or stone. All of these soils should be treated in a suitable way, which is why the aim is to determine the nature and, in particular, different properties of different soil areas by means of an automated process.

According to the state of the art, a combination of several sensors is typically used in household robots to enable navigation and to ensure safety functions, such as preventing falls, for example using so-called cliff sensors.

In the case of vacuum cleaner robots or mopping robots, the aforementioned cliff sensors are known, which prevent the robot from falling down an abyss, usually a staircase, and being damaged. Mechanical sensors with a spring and a push button sensor are sometimes used for this purpose.

It should be noted that mechanical sensors can generally wear out. In the case of a vacuum cleaner robot or mopping robot, these can get dirty, hair can get caught and make the sensor unusable. The characterization of the floor surface with regard to the general suitability for a mopping robot or for the use of a carpet brush, however, is impossible with a simple touch sensor.

The use of optical sensors is also known in vacuum cleaner robots or mopping robots. These are based, for example, on the time-of-flight concept. The transit time of a light pulse is measured and the distance to the ground is determined. A stair nosing is detected by a larger measured distance or, if the distance is too great, by failure of the back measurement. The disadvantage of these sensors is the comparatively high cost. The characterization of the soil surface on the basis of the reflection properties is basically conceivable, but this is not implemented in current products.

In addition, the use of optical sensors is known in vacuum cleaner robots or mopping robots, which determine the distance through the degree of coupling between an LED and a photodetector. This is done in a similar way to diffuse sensors. It will be based on this 1 referenced.

Due to the functional principle of the sensor, the distance cannot be distinguished from the reflection coefficient. A dark floor at a distance of 50 mm generates the same measurement signal as a lighter floor at a distance of 60 mm. Soiling of the optics cannot be distinguished from the reflection coefficient either. The detection of a step is possible, but the determination of the soil properties is not.

A cleaning robot is also known which detects the edge of carpets by means of detectors which are attached to both wheels of the robot. The sensors used are mainly tactile.

Tactile sensors usually differentiate between the flexibility of the floor and can therefore only be used to detect softer materials, such as. B. carpets can be used. A more detailed differentiation, for example between a parquet or laminate floor, is not possible in this way.

Furthermore, a cleaning robot is known which enables a distinction between carpets and a smooth surface by measuring the speed of the brush. A reduction in the speed can indicate an increased mechanical resistance and thus a carpet.

However, the speed of the brush depends on many other factors. In addition to the Essentially, wear and tear and soiling of the brush change the speed of the surface. In addition, the brush speed should be adjustable according to the cleaning purpose.

The pamphlet EP 1 967 115 A2 describes a method for detecting a floor covering for a vacuum cleaner which has a fan which can be driven by an electric motor. This motor is brought from a first speed at a first operating point to a second operating point at a second speed. The floor covering is identified by comparing the current engine parameters with reference values.

From the pamphlet DE 691 20 176 T2 A method for recognizing carpets and stairs for a cleaning robot is known, in which an ultrasonic wave signal is transmitted during a first period of time by an ultrasonic wave signal transmitter and a period of time is measured from the time when an ultrasonic wave signal indication signal is generated and determined as a function thereof, whether there is a carpet. It is taken into account that the absorption and reflection behavior depends on the nature of the floor covering.

Disclosure of the invention

Against this background, a method according to claim 1 and an arrangement according to claim 12 are presented. Embodiments emerge from the dependent claims and from the description.

The method is used to determine at least one item of information about a floor and is carried out with an arrangement that can be moved relative to the floor by means of a drive. The drive is controlled with a drive signal which has a profile with a periodic component which causes a torque with a periodic profile to be impressed on at least one drive wheel of the arrangement, this impression causing the arrangement to oscillate. A signal characterizing the oscillation is then recorded and evaluated in order to determine the at least one item of information.

In one embodiment of the method, the signal characterizing the oscillation carries information on an amplitude of the oscillation. The amplitude of the oscillation can then be related to the course of the periodic component of the drive signal in order to determine the at least one item of information. As an alternative or in addition, the amplitude of the oscillation can be related to a signal characterizing the applied torque in order to determine the at least one item of information.

In one embodiment, a torque is impressed with a sinusoidal profile.

In this method, a torque, for example. A torque with a periodic z. B. sinusoidal curve, imprinted on at least one drive wheel or several drive wheels or on a drive axle and it is with a sensor, e.g. is set in relation to the applied torque in order to determine the at least one item of information.

ermittelt werden, wobei I die Intensität, sig eine Eigenschaft einer Anordnung, mit der das Verfahren durchgeführt wird, die von einem Winkel zwischen einem Sender und einem Empfänger des Abstandssensors abhängt. sig kann bspw. experimentell bestimmt werden. Es besteht eine Abhängigkeit von dem Winkel zwischen Sender und Empfänger und von der Strahlbreite.A floor distance z can be measured with an optical distance sensor, with a reflection coefficient R also being determined. This ground clearance z can be about $$\text{z}=-\text{I.}\text{'}\left(\text{z}\text{, R}\right)/\text{I.}\left(\text{z}\text{, R}\right)*\text{sig}/2+\text{zmax}$$

be determined, where I is the intensity, sig a property of an arrangement with which the method is carried out, which depends on an angle between a transmitter and a receiver of the distance sensor. sig can be determined experimentally, for example. There is a dependency on the angle between transmitter and receiver and on the beam width.

erfolgen, wobei

die Amplitude der Vibration und M das eingeprägte Drehmoment ist.The flexibility G of the soil can also be determined as information. This can, for example

take place, where

is the amplitude of the vibration and M is the applied torque.

With the determined flexibility G and the determined reflection coefficient R, a classification of the floor can be carried out, which is of interest to the user.

A calibration can also be performed first to achieve reliable results.

The method presented here enables at least one piece of information to be recorded about or about a floor. Information about a soil is understood here to mean both properties inherent in the soil, such as, for example, the degree of hardness, which relate to the nature of the soil as well as the distance of the arrangement with which the method is carried out or of a sensor or distance sensor in the arrangement to the floor or the floor surface, which in turn can be influenced by geometric properties of the floor.

The described arrangement is used to carry out the method and is, for example, integrated or designed as such in a robot, in particular in a household robot.

With the presented method, the determination of the distance between a sensor and a reflector, typically the ground, is thus made possible in an embodiment in order to detect steps or abysses and additional ground properties, for example. The recorded soil properties can be used to classify the soil, i. H. to determine the type of soil. For example, different floors can be divided into the categories of carpet, plastic, wood and stone floors. In a separate step, the floors can also, if necessary, first be categorized as hard, i.e. H. wipeable, and soft, d. H. cannot be wiped off, be divided.

For this purpose, the coupling between a light source and an optical detector is determined in accordance with a reflex light scanner with intersecting transmitting or receiving apertures, whereby by varying the distance to the reflector, typically the floor, the distance as an absolute value, which is a cliff detection is required, can be distinguished from the reflection coefficient. The distance variation is stimulated by the robot's drive and measured back by means of a rotation rate sensor. By establishing the relationship between movement and excitation, conclusions can be drawn about the flexibility of the soil. The information about the reflection coefficient and the compliance can then be used to classify the soil type.

• - Es wird eine genauere Messung des Abstands zum Boden unabhängig von der Reflexionseigenschaft des Bodens ermöglicht. Auf diese Weise können Treppen und Türschwellen besser erkannt werden.
• - Es wird als neue Funktion eine Klassifizierung des Bodentyps mit vorhandenen Sensoren ermöglicht. Daraus ergibt sich eine bessere Gesamtfunktion hinsichtlich eines optimierten Einsatzes der Staubsaugerbürste und einer Sicherstellung, das nur bei dafür geeigneten Böden gewischt wird.
• - Es wird eine kostengünstige Lösung bereitgestellt, da nur im geringem Umfang oder sogar gar keine zusätzliche Hardware erforderlich ist.

The method described and the arrangement presented have a number of advantages, at least in some of the versions:

• A more precise measurement of the distance to the ground is made possible, regardless of the reflective properties of the ground. In this way, stairs and door thresholds can be better recognized.
• - As a new function, the soil type can be classified using existing sensors. This results in a better overall function with regard to an optimized use of the vacuum cleaner brush and a guarantee that only suitable floors are wiped.
• A cost-effective solution is provided, since only a small amount or even no additional hardware is required.

Further advantages and configurations of the invention emerge from the description and the accompanying drawings.

It goes without saying that the features mentioned above and those yet to be explained below can be used not only in the respectively specified combination, but also in other combinations or on their own, without departing from the scope of the present invention.

Figure list

• 1 shows a prior art optical distance sensor.
• 2 shows qualitative signal curves of the optical distance sensor in a graph 1 depending on the ground clearance for two different ground reflectivities.
• 3 shows an embodiment of an arrangement for performing the method.
• 4th shows, in a graph, qualitative signal profiles of an optical distance sensor in an embodiment of the method presented.
• 5 shows a classifier.

Embodiments of the invention

The invention is shown schematically on the basis of embodiments in the drawings and is described in detail below with reference to the drawings.

1 shows a distance sensor according to the prior art, generally designated by the reference number 10 is designated. In this case, the aperture of the transmission beam and the aperture of the receiving optics intersect at a distance of 40 mm.

2 shows in a graph 30th , on its abscissa 32 the floor distance z and on its ordinate 34 the intensity I is plotted, a first signal curve 40 with a ground reflectivity R ₁ and a second signal curve 42 with a floor reflectivity R ₂ , where R ₁ > R ₂ .

The intensity I measured at the detector thus depends on the distance from the reflector, in this case the distance from the ground, and on the reflectivity or the reflection coefficient R. If the distance is too close, the coupling between transmitter and receiver is almost zero, and if the distance is too large, likewise. in the The coupling is maximal in the area in which the beams cross.

3 shows a schematic representation of an arrangement for carrying out the method, denoted overall by the reference number 100 and is designed as a household robot and is located on a floor 112 is located. The illustration shows a drive wheel 102 that from a drive 104 is driven. This can be on the drive wheels 102 a torque M 106 imprint. An inertial sensor 108 can adjust the rate of rotation about a transverse axis, ie parallel to a drive axis 110 , measure up. The measurement process can be carried out at a standstill or while driving. The procedure at standstill is described below:

/M(ω) bestimmt werden. Dazu müssen die Sensorsignale aufgezeichnet werden und mittels FFT in den Frequenzraum übertragen werden. Die weitere Berechnung kann dann wahlweise an einer Frequenzstückstelle oder an mehreren geführt werden. Die Frequenz der Anregung beträgt in Ausgestaltung 1 bis 100 Hz und ist an die Masseverhältnisse der Anordnung 100 angepasst. Das Drehmoment M 106 bewirkt eine Drehschwingung bzw. Vibration der Anordnung 100 um die Antriebsachse 110. Die Schwingung bzw. Vibration wird von dem Inertialsensor 108 erfasst bzw. gemessen und hängt vom eingeprägten Drehmoment M 106 und von der Nachgiebigkeit des Bodens 112 an einem Kontaktpunkt 114 zwischen dem Boden 112 und einem Stützrad 116 ab. Der Inertialsensor 108 stellt ein entsprechendes Sensorsignal bereit. Dieses Sensorsignal kann mittels Filterung oder Synchrondemodulation auf die gesuchte Amplitude

der Schwingung reduziert werden und anschließend mit dem bekannten eingeprägten Drehmoment M 106 in Relation gesetzt werden und somit mit Gleichung 1 ein Maß für die Nachgiebigkeit G des Bodens 112 bestimmt werden:

The drive 104 By means of appropriate control, it creates, for example, a sinusoidal torque on the drive axle 110 or the drive wheels. The course does not have to correspond to a pure sine. Any other periodic function is conceivable because the fundamental harmonic can be determined using an FFT. In the case of non-mono-frequency excitation, G (ω) =

/ M (ω) can be determined. To do this, the sensor signals must be recorded and transferred to the frequency domain using FFT. The further calculation can then optionally be carried out at one frequency segment point or at several. In an embodiment, the frequency of the excitation is 1 to 100 Hz and is dependent on the mass ratios of the arrangement 100 customized. The torque M 106 causes a torsional oscillation or vibration of the arrangement 100 around the drive axis 110 . The oscillation or vibration is from the inertial sensor 108 detected or measured and depends on the applied torque M 106 and the flexibility of the ground 112 at a contact point 114 between the ground 112 and a support wheel 116 away. The inertial sensor 108 provides a corresponding sensor signal. This sensor signal can be filtered or synchronized to the desired amplitude

the vibration can be reduced and then with the known applied torque M 106 be put in relation and thus with equation 1 a measure for the flexibility G of the soil 112 be determined:

In the front part of the arrangement 100 is on the ground 112 facing optical distance sensor 120 . The distance α between the distance sensor 120 and the ground 112 is nominal, for example, 50 mm, practically 5 to 80 mm. Due to the nature of the ground, for example the presence of a step or the wheels sinking in on soft ground, the value is initially unknown. Under the influence of the torsional vibration, depending on the reflection coefficient R, a different DC component or AC component is to be expected.

4th shows in a graph 230 , on its abscissa 232 the floor distance z and on its ordinate 234 the intensity I is plotted, a first signal curve 240 with a ground reflectivity R ₁ and a second signal curve 242 with a floor reflectivity R ₂ , where R ₁ = 1 and R ₂ = 0.8. The illustration shows the signal curve of the optical distance sensor under the influence of the torsional vibration.

erreicht werden, wobei R der gesuchte Reflexionskoeffizient der Oberfläche, z der gesuchte Abstand zum Boden, zmax der bekannte Abstand bei maximaler Intensität, d. h. Kreuzungspunkt der Sende- und Empfangsapertur, und sig eine bekannte Eigenschaft der Anordnung beschreibt, welche dominant vom Winkel zwischen Sender und Empfänger abhängt.With a suitable design of the optics, a sensor characteristic can be adjusted accordingly $$\text{I.}\left(\text{z}\text{, R}\right)\sim \text{R * exp}\left(-{\left(\text{z}-\text{zmax}\right)}^2/\text{sig}\right)$$

where R is the desired reflection coefficient of the surface, z the desired distance to the ground, zmax the known distance at maximum intensity, i.e. the crossing point of the transmitting and receiving aperture, and sig describes a known property of the arrangement which is dominant of the angle between the transmitter and Recipient depends.

The DC component of the signal I can be determined, for example, by averaging and corresponds to the function value I (z, R). The AC component can be determined by filtering or synchronous demodulation with the excitation frequency of the drive axle and, after division by the amplitude of the oscillation, corresponds to the functional value of the derivative I '(z, R).

Functionally, the derivation can be described according to equation 3: $$\text{I.}\text{'}\left(\text{z}\text{, R}\right)\sim -2\text{R.}\left(\text{z}-\text{zmax}\right)/\text{sig}*\text{exp}\left(-{\left(\text{zmax}\right)}^2/\text{sig}\right)$$

The required distance z can be calculated according to equation 4: $$\text{z}=-\text{I '}\left(\text{z}\text{, R}\right)/\text{I.}\left(\text{z}\text{, R}\right)*\text{sig}/2+\text{zmax}$$

The reflection coefficient R can then be determined by substituting it in equation 2 or 3.

The values zmax and sig from equation 2 are dependent on the alignment of the optical components of the sensor and can be subject to fluctuations due to the manufacturing process.

These values can be determined using the calibration method described above a background with a known reflection coefficient R. This calibration background can, for example, be integrated into the charging station of the arrangement or of the robot. After the calibration, the same surface can be used in further operation of the robot in order to determine the degree of soiling of the optical sensor. The information about the degree of contamination of the sensor can support the identification of the floor and can be used to inform the user that the sensor needs to be cleaned.

The sensor system or the method ultimately provides information on the distance z, the reflection coefficient R and the flexibility G of the soil. The distance is used for the detection of abysses and door sills or generally for the identification of the floor topology. The reflection coefficient and the compliance become accordingly 5 fed to a classifier.

5 shows such a classifier 300 , with the reflectance R as the input variable 302 and the compliance G 304 are fed. The classifier 300 then gives the soil type T 310 the end.

The classifier 300 was previously trained with training data from known soils and known methods of machine learning (machine learning, e.g. support vector machines, decision tree, etc.). In a further embodiment, the device (reference number 100 in 3 ) provided the option of using the classifier 300 adapt and refine. This can be done, for example, on the basis of the data collected in the company. An exemplary implementation provides that the device is operated in a network or has an Internet connection. These can be used to transmit raw data and / or, in order to reduce the power consumption by radio, preprocessed sensor data to a data service and to use the classifier 300 to refine and thus enable reliable soil detection.

QUOTES INCLUDED IN THE DESCRIPTION

This list of the documents listed 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 1967115 A2 [0014]
• DE 69120176 T2 [0015]

Claims (14)

A method for determining at least one item of information about a floor (112) with an arrangement (100) which can be moved relative to the floor (112) by means of a drive (104), the drive (104) being controlled with a drive signal which has a curve with a periodic component which causes a torque (106) with a periodic curve to be impressed on at least one drive wheel (102) of the arrangement (100), this impressing causing the arrangement (100) to oscillate, with the oscillation characterizing signal is detected and evaluated in order to determine the at least one piece of information. Procedure according to Claim 1 , in which the signal characterizing the oscillation carries information on an amplitude of the oscillation. Procedure according to Claim 2 , in which the amplitude of the oscillation is related to the course of the periodic component of the drive signal in order to determine the at least one item of information. Procedure according to Claim 2 , in which the amplitude of the oscillation is related to a signal characterizing the applied torque (106) in order to determine the at least one item of information. Method according to one of the Claims 1 until 4th , in which a torque (106) is impressed with a sinusoidal curve. Method according to one of the Claims 1 until 5 , in which a floor distance z is measured with an optical distance sensor (120), a reflection coefficient R (302) also being determined. Procedure according to Claim 6 , where the ground clearance z about $$\text{z}=-\text{I '}\left(\text{z}\text{, R}\right)/\text{I.}\left(\text{z}\text{, R}\right)*\text{sig}/\text{2}+\text{zmax}$$

is determined, where I is the intensity and sig is a property of an arrangement (100) with which the method is carried out, which depends on an angle between a transmitter and a receiver of the distance sensor (120). Method according to one of the Claims 1 until 7th , in which the flexibility G (304) of the soil (112) is determined as information. Procedure according to Claim 8 , in which the compliance G (304) is determined via

whereby

is the amplitude of the vibration and M is the applied torque (306). Procedure according to Claim 6 and Claim 8 or 9 , in which a classification of the soil (112) is carried out with the determined flexibility G (304) and the determined reflection coefficient R (302). Method according to one of the Claims 1 until 10 , in which a calibration is carried out first. Arrangement for detecting at least one item of information about a floor (112), the arrangement (100) for carrying out a method according to one of the Claims 1 until 11 is set up. Arrangement according to Claim 10 , which is designed as a household robot. Arrangement according to Claim 13 which is set up to clean the floor (112).

Images