Autonomous traveling type robot cleaner

Abstract

To provide a robot cleaner that can respond to both climbing over a step and descending a step.SOLUTION: A robot cleaner comprises: a robot main body 2 which travels on a floor by traveling means 4; reflection intensity measurement means for measuring a reflection intensity of light from the floor in a traveling direction of the robot main body 2; attitude measurement means for measuring an attitude of the robot main body 2; and determination means 8. The determination means 8 estimates an elevation difference that the robot cleaner can travel along the floor in the traveling direction, and determines whether or not the robot cleaner can travel along the floor by employing first time-series data for a fixed period of a measured value obtained by the reflection intensity measurement means and second time-series data for a fixed period of a measured value obtained by the attitude measurement means.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] The disclosed technology relates to a self-driving robot vacuum cleaner for general households, and particularly to a technology for detecting steps.

Background Art

[0002] In recent years, robots that travel while avoiding obstacles and automatically clean the floor surface (self-driving robot vacuum cleaners, hereinafter simply referred to as "robot vacuum cleaners") have attracted attention. Due to the progress of computer technology, robot vacuum cleaners can detect obstacles usually installed indoors such as furniture, household appliances, and interiors with relatively high accuracy and can travel on the floor surface while avoiding them.

[0003] For example, Patent Document 1 discloses a robot vacuum cleaner configured to determine whether it is possible to cross a step on the floor surface based on the clearance detected by a step sensor and the amount of inclination detected by an inclination detection means. That is, in Patent Document 1, the clearance under the case in front of the main body case and the amount of inclination in the traveling direction are measured, and the amount of inclination is reflected in the measured clearance to determine whether it is possible to cross the step.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the technology of Patent Document 1, on the lower surface of the main body case, a step sensor and an inclination detection means are configured using optical sensors respectively provided before and after the center in the width direction. Specifically, the clearance (the distance between the lower surface of the case main body and the floor surface) is measured by the front sensor, and the front and rear inclinations are measured based on the detection values of the front and rear sensors and their change amounts.

[0006] Generally, when attempting to measure a distance such as clearance with a certain degree of accuracy, it is necessary to use a relatively expensive position detection device such as a PSD (Position Sensitive Detector), for example.

[0007] In addition to position detection devices such as PSDs, relatively inexpensive light reflection intensity sensors that measure distance using the reflection intensity of light are known. However, when using a light reflection intensity sensor, there is a problem that it is easily affected by the material and surface shape of the target floor surface, making it difficult to accurately measure the distance. For example, when there is a floor surface with different reflectivities before and after the traveling direction of a robot vacuum cleaner, there is a risk that the robot vacuum cleaner will misjudge that there is a step on the floor surface at that boundary. Also, when there are different reflectivities on the floor surface and a step lower than the floor surface, there is a risk of misjudging the depth of the step.

[0008] Also, from another perspective, a general robot vacuum cleaner is configured to be able to cross a step of about 20 mm. On the other hand, if it falls from a step with a depth greater than a predetermined level with respect to the floor surface, it becomes difficult to return to the original height or there is a risk of damage, so it is configured to function as a fall prevention function to stop or retreat.

[0009] Here, when using a light reflection intensity sensor to measure the clearance between the lower surface of the robot vacuum cleaner and the floor surface, during the process of climbing over a step, if the distance between the lower surface of the robot vacuum cleaner and the floor surface is measured to be separated by a predetermined threshold value or more, that is, if there is a step with a depth greater than a predetermined level in the traveling direction, it may be misjudged and the fall prevention function may be activated. Then, there is a problem that the robot vacuum cleaner stops or retreats even though it can cross the step. That is, there is a problem that the robot vacuum cleaner cannot sufficiently determine whether it can cross a step or descend a step.

[0010] Therefore, the technology disclosed herein aims to provide a robotic vacuum cleaner that can determine whether it can overcome a step or descend a step even when using a light reflection intensity sensor.

Means for Solving the Problems

[0011] The disclosed technology relates to an autonomous mobile robotic vacuum cleaner.

[0012] The robotic vacuum cleaner according to one aspect of the present disclosure includes a robot body that travels on the floor by traveling means, a reflection intensity measuring means that measures the reflection intensity of light from the floor in the traveling direction of the robot body, an attitude measuring means that measures the attitude of the robot body, and a first time-series data that is time-series data of measurement values measured by the reflection intensity measuring means for a certain period and a second time-series data that is time-series data of measurement values measured by the attitude detection means for a certain period. Using these data, it is estimated whether there is a height difference on the floor in the traveling direction that allows travel, and a determination means for determining whether travel is possible is provided.

[0013] The robotic vacuum cleaner of the present disclosure estimates whether there is a height difference on the floor in the traveling direction that allows travel using the first time-series data of the measurement values measured by the reflection intensity measuring means for a certain period and the second time-series data of the measurement values measured by the attitude detection means for a certain period. That is, a combination of measurement means, namely the measurement value of the reflection intensity measuring means and the measurement value of the attitude detection means, is performed, and the possibility of travel in the traveling direction is comprehensively determined using the time-series data of these measurement values for a certain period.

[0014] Thereby, the accuracy of the travel determination corresponding to the presence or absence of steps and changes in the floor material can be improved. Specifically, in the conventional configuration, when there is a step on the floor surface that the robotic vacuum cleaner can overcome, when the robotic vacuum cleaner climbs over the step, the distance between the lower surface of the robotic vacuum cleaner and the floor surface increases, resulting in a determination that the step cannot be overcome, and the fall prevention function may malfunction erroneously. However, by using the technology of the present disclosure, such malfunction can be suppressed. That is, in the robotic vacuum cleaner of the present disclosure, the accuracy of the travel determination corresponding to the presence or absence of steps and changes in the floor material can be improved.

[0015] In the above aspect, speed measurement means for measuring the traveling speed of the robot cleaner or inertial measurement means for measuring the inertia of the robot body is provided, and the determination means uses third time-series data which is time-series data of measurement values by the speed measurement means for a certain period or fourth time-series data which is time-series data of measurement values by the inertial measurement means for a certain period to estimate whether the floor on the traveling direction side has a height difference that allows travel. This may be the case.

[0016] In this way, by using the third time-series data to the fourth time-series data in the determination process of the determination means, the determination accuracy of the determination means can be improved. For example, as the speed measurement means, there is a method of measuring the rotational speed of the drive wheel using a wheel speed sensor that measures the wheel speed of the robot cleaner. In addition to the measurement result of this speed measurement means, an acceleration command to the drive wheel may be used.

[0017] In the above aspect, the traveling means includes a drive wheel driven by a motor and a driven wheel rotatably provided on the rear side in the traveling direction than the drive wheel, and the reflection intensity measurement means includes an optical reflection intensity sensor provided on the front side in the traveling direction than the drive wheel. This may be the case.

[0018] As a result, information in the traveling direction can be obtained earlier, so when there is a height difference on the floor surface, the determination timing of whether travel is possible due to the height difference can be advanced.

[0019] In the above aspect, the reflection intensity measurement means may include a plurality of optical reflection intensity sensors arranged with their positions shifted in the traveling direction.

[0020] As a result, it becomes easier to grasp the state of change of the floor material during the traveling process of the robot cleaner.

Effect of the Invention

[0021] According to the disclosed technology, it is possible to provide a robot cleaner capable of coping with both step-over and step-down.

Brief Description of the Drawings

[0022]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Modes for Carrying Out the Invention

[0023] Hereinafter, the disclosed technology will be described. However, the description is merely illustrative in nature and is not intended to limit the present invention, its applications, or its uses. Also, in the following description, the description will focus on configurations, operations, etc. highly related to the present invention.

[0024] <Robot Vacuum Cleaner> In FIGS. 1 and 2, an autonomous traveling type robot vacuum cleaner 1 to which the disclosed technology is applied is illustrated. FIG. 1 is a side view showing a schematic configuration of the robot vacuum cleaner 1. FIG. 2 is a schematic view showing the bottom surface of the robot vacuum cleaner 1.

[0025] 〔Robot Main Body〕 The robot vacuum cleaner 1 includes a robot main body 2 having a flat box-shaped appearance. On the lower surface of the front end portion of the robot main body 2, a horizontally long suction port 3a that is long in the left-right direction is provided. Inside the suction port 3a, a brush (not shown) is pivotally supported so as to extend along the suction port 3a.

[0026] Although not shown, the robot main body 2 incorporates a suction pump, a dust box, etc. When the robot vacuum cleaner 1 is in operation, it is configured to suck dust etc. present on the floor surface into the dust box through the suction port 3a while scraping it up with a brush. That is, the suction port 3a and the brush constitute a cleaning unit 3 for cleaning the floor surface. Note that the cleaning unit 3 may have a wiping function.

[0027] A pair of drive wheels 4a are installed on both the left and right sides of the lower surface of the robot main body 2. Also, driven wheels 4b are installed near each corner (outer side) of the suction port 3a on the lower surface of the robot main body 2, and a driven wheel 4c is installed on the rear side at the center of the lateral width of the lower surface of the robot main body 2. Each drive wheel 4a is configured to be able to independently rotate forward and backward freely by driving control of a motor. That is, the pair of drive wheels 4a and the driven wheels 4b, 4c constitute a traveling means 4 for traveling on the floor surface.

[0028] Thereby, the robot vacuum cleaner 1 can perform operations such as forward movement, backward movement, and turning left or right, and can travel freely on the floor surface. However, the robot main body 2 has a forward movement as the basic operating state and normally travels forward. Backward movement is performed when avoiding obstacles etc. In the present disclosure, the forward movement direction of the robot vacuum cleaner 1 is referred to as the "traveling direction" or "front", and the backward (rearward) movement direction is referred to as the "backward direction" or "rear". Furthermore, up, down, left, and right are defined based on the line of sight of the robot vacuum cleaner 1 facing the traveling direction.

[0029] As shown in FIG. 3, the robot main body 2 includes a plurality of measurement means, a determination means 8, and a control means 9. A plurality of optical reflection intensity sensors 5, wheel speed sensors 6, and an IMU 7 are installed in the robot main body 2 as the plurality of measurement means.

[0030] -Light reflection intensity sensor (reflection intensity measurement means)- Returning to FIG. 2, the light reflection intensity sensor 5 is attached to the lower surface of the robot body 2, irradiates the floor surface with infrared light of a predetermined wavelength, and measures the reflection intensity thereof. The measured value of the light reflection intensity sensor 5 is transmitted to the determination means 8. It is used for the purpose of acquiring information related to the clearance between the lower surface of the robot body 2 and the floor surface (hereinafter referred to as "clearance information").

[0031] When the material of the floor surface, etc. is uniform, that is, when the light reflectivity of the floor surface is the same, the light reflection intensity sensor 5 can measure the clearance between the lower surface of the robot body and the floor surface. On the other hand, in an environment where the material of the floor surface changes midway, that is, in an environment where the reflectivity of the floor surface changes midway and the change in the reflectivity is not grasped in advance, there is a characteristic that the clearance between the lower surface of the robot body 2 and the floor surface cannot be accurately measured. That is, although the above-mentioned "clearance information" can be obtained, there is a characteristic that the clearance between the lower surface of the robot body 2 and the floor surface cannot be accurately determined only by the clearance information in some cases.

[0032] In the present embodiment, the light reflection intensity sensor 5 is provided in the traveling direction with respect to the driven wheel 4b provided on the front side of the suction port 3a, and includes a light reflection intensity sensor 5a arranged side by side with a distance therebetween on the left and right, a light reflection intensity sensor 5b provided at the right front of the right drive wheel 4a, and a light reflection intensity sensor 5c provided at the left front of the left drive wheel 4b. In other words, the light reflection intensity sensor 5 (5a, 5b, 5c) is provided in the traveling direction with respect to the drive wheel 4a. Further, the light reflection intensity sensor 5a and the light reflection intensity sensors 5b and 5c are arranged with their positions shifted in the front-rear direction. In this way, by shifting the positions of the plurality of light reflection intensity sensors 5 in the traveling direction (front-rear direction), when the measured value of the light reflection intensity sensor 5 changes during the traveling of the robot cleaner 1, it becomes easier to grasp the change in the traveling environment in time series in combination with the measured values of other sensors.

[0033] The light reflection intensity sensor 5 is an example of a reflection intensity measurement means for measuring the light reflection intensity from the floor in the traveling direction of the robot main body. Note that, as the reflection intensity measurement means, means different from the light reflection intensity sensor 5 may be used.

[0034] Also, the number and installation positions of the light reflection intensity sensors 5 (5a, 5b, 5c) are not limited to the configuration shown in FIG. 2. For example, the two light reflection intensity sensors 5a provided on the left and right in the front may be grouped together and regarded as one light reflection intensity sensor 5a.

[0035] - Wheel speed sensor (speed measurement means)- The wheel speed sensor 6 measures the rotational speed of the left and right drive wheels 4a and the direction in which the robot cleaner 1 travels. The wheel speed sensor 6 is used for the purpose of obtaining the traveling speed of the robot cleaner 1. The measured value of the wheel rotational speed measured by the wheel speed sensor 6 is transmitted to the determination means 8. For example, when measuring the rotational speed of the drive wheel 4a with a hall sensor, for each drive wheel 4a, two pieces of data can be obtained, and a total of four pieces of data can be used as input data (refer to the column of the wheel speed sensor in FIGS. 5 to 7).

[0036] The wheel speed sensor 6 is an example of a speed measurement means for measuring the traveling speed of the robot cleaner 1. Note that the speed measurement means is not limited to the wheel speed sensor 6, and other means may be used to measure the traveling speed of the robot cleaner 1. Also, with the wheel speed sensor 6, only the rotational speed of the drive wheel 4a may be measured. In this case, in the determination means 8 described later, for example, an acceleration command of the drive wheel 4a may be additionally used for the measurement result of the rotational speed.

[0037] - IMU (attitude measurement means, inertial measurement means)- The IMU (Inertial Measurement Unit) 7 is configured to be able to measure the inertia and attitude of the robot main body 2 based on changes and inclinations in the traveling direction of the robot main body 2.

[0038] Specifically, the IMU 7 is configured to measure the angles or angular velocities of the three axes of the robot body 2, that is, the inclination in the left-right direction (Yaw), the inclination in the front-rear direction (Pitch), and the degree of rotation (Roll). By using the measured values of the IMU 7, it is possible to measure changes in the traveling direction and inclination of the robot body 2.

[0039] The IMU 7 is an example of attitude measurement means for measuring the attitude of the robot body 2. Further, the IMU 7 is an example of inertial measurement means for measuring the inertia of the robot body 2. Note that the attitude measurement means or inertial measurement means is not limited to the IMU 7. For example, attitude measurement and inertial measurement may be measured using separate sensors or devices. Further, means other than the IMU 7 may be used to measure the attitude or inertia of the robot body 2.

[0040] - Judgment means - The judgment means 8 estimates whether the floor in the traveling direction has a height difference that allows travel using the time-series data DT of the measured values respectively measured by a plurality of measurement means (for example, the light reflection intensity sensor 5, the wheel speed sensor 6, the IMU 7), and judges whether travel is possible. The judgment means 8 may use the acceleration command to the drive wheels in addition to the measured value measured by the wheel speed sensor 6.

[0041] The time-series data DT includes first time-series data DT1 that is time-series data of the reflection intensity measured by the light reflection intensity sensor 5 for a certain period, and second time-series data DT2 that is time-series data of the attitude information of the robot body 2 measured by the IMU 7 for a certain period.

[0042] Note that the time-series data DT is not limited to the first time-series data DT1 and the second time-series data DT2. For example, as the time-series data DT, third time-series data DT3 that is time-series data of the traveling speed of the robot cleaner 1 measured by the wheel speed sensor 6 for a certain period may be included. Further, fourth time-series data DT4 that is time-series data of the inertia value of the robot body 2 measured by the IMU 7 for a certain period may be included. By increasing the types of the time-series data DT, the estimation accuracy of whether the floor in the traveling direction has a height difference that allows travel can be improved.

[0043] Specifically, the determination means 8 includes a trained machine learning model that estimates whether the floor in the traveling direction has a height difference that allows travel using a plurality of time-series data DT (for example, the aforementioned DT1 to DT4). As the trained machine learning model, for example, a deep learning model can be used.

[0044] An example of a method for generating the trained machine learning model is shown. For example, for the height difference within the travelable limit set for the robot vacuum cleaner 1 that is the operation target (for example, 20 mm), learning travel data with the floor material changed before and after a step having a height difference up to twice that value is created. As the floor material, the materials of the main floors and carpets in the room are set. Then, a trained machine learning model is generated based on the created learning travel data.

[0045] When the aforementioned time-series data DT is input to this trained machine learning model, an estimation result as to whether the floor in the traveling direction has a height difference that allows travel is output. Based on the estimation result obtained from the trained machine learning model, it is determined whether the robot vacuum cleaner 1 can travel in the traveling direction. A specific example of the determination process by the determination means 8 will be described later.

[0046] - Control means - Inside the robot main body 2, a control means 9 for controlling the operation of the robot vacuum cleaner 1 is installed. The control means 9 is a so-called computer. That is, the control means 9 is composed of hardware such as a processor, a memory, and an interface, and software such as a control program and data installed in the memory.

[0047] More specifically, each function of the control means 9 is realized by a processor such as a CPU executing a program stored in the memory. Note that the control means 9 may be realized by hardware such as an FPGA having a function equivalent to that of a processor executing a program, or may be realized by the cooperation of software and hardware.

[0048] Although the description is omitted because it is different from the subject of the disclosed technology, an advanced control program such as AI that controls driving is implemented in the control means 9. Thereby, the robot body 2 is configured to control the traveling means 4 to travel autonomously and to travel while avoiding obstacles such as furniture and interiors. Further, during the traveling operation by the traveling means 4, the cleaning unit 3 is controlled so that the floor surface can be cleaned.

[0049] Furthermore, the control means 9 has a function of changing the operation of the robot cleaner 1 based on the determination result of the determination means 8. Specifically, for example, when it is determined in the determination means 8 that the robot cleaner 1 cannot proceed in the traveling direction, the control means 9 stops, reverses, or changes the traveling route so that the drive wheel 4a does not proceed further.

[0050] <Control Example of Robot Cleaner> FIG. 4 shows an example (flowchart) of the traveling control during the cleaning operation by the robot cleaner 1. That is, in FIG. 4, it is assumed that the robot cleaner 1 is performing a cleaning operation under the control of the control means 9 at the time of "START".

[0051] - Operation Example (Part 1)- First, the traveling operation (cleaning operation) shown in FIG. 5 will be described with reference to the flowchart of FIG. 4.

[0052] As shown in the upper part of FIG. 5, in this operation example, it is assumed that the robot cleaner 1 travels on the floor F using the floor material FA in the traveling direction (from left to right in the drawing). And it is assumed that there is a step (for example, H = 15 mm) that the robot cleaner 1 can overcome in front of the traveling direction.

[0053] The lower part of FIG. 5 shows an example of the time-series data DT output from a plurality of measurement means to the determination means 8. Note that FIGS. 6 and 7 show the measurement waveforms of the same measurement target although the traveling scenes are different.

[0054] Specifically, the time changes during running of the output waveforms of the light reflection intensity sensor 5, the output waveform of the wheel speed sensor 6, and the output waveform of the IMU 7 are illustrated by arranging them vertically with their time axes aligned with each other.

[0055] The output waveforms of the light reflection intensity sensor 5 are, in order from the top, (1) the output waveform of the light reflection intensity sensor 5a provided at the left front of the robot body 2, (2) the output waveform of the light reflection intensity sensor 5a provided at the right front of the robot body 2, (3) the output waveform of the light reflection intensity sensor 5c provided in front of the left of the left drive wheel 4a, and (4) the output waveform of the light reflection intensity sensor 5b provided in front of the right of the right drive wheel 4a.

[0056] The output waveform of the wheel speed sensor 6 shows an example when a hall sensor is used as the wheel speed sensor. In the examples of FIGS. 5 to 7, an example is shown in which, from the measurement results of the hall sensor, not only the speeds of the drive wheels 4a (right wheel and left wheel) but also the 2 values of the a layer and the b layer are used as inputs to determine forward and backward movement. More specifically, in order from the top, (1) the measured waveform of the rotational speed (a layer) of the left drive wheel 4a, (2) the measured waveform of the rotational speed (b layer) of the left drive wheel 4a, (3) the measured waveform of the rotational speed (a layer) of the right drive wheel 4a, and (4) the measured waveform of the rotational speed (b layer) of the right drive wheel 4a.

[0057] The output waveforms of the IMU 7 are, in order from the top, (1) the Yaw waveform (waveform of the inclination in the left - right direction), (2) the Pitch waveform (waveform of the inclination in the front - rear direction), and (3) the Roll waveform (waveform indicating the rotational state).

[0058] Also, regarding the period surrounded by the dashed line, period T1 is the stop state of the robot vacuum cleaner 1, period T2 is the start - up period of the robot vacuum cleaner 1, period T3 is the normal running period without steps, period T4 is the period around reaching a step, and period T5 is the period of overcoming the step.

[0059] In step S1, the control means 9 determines whether to continue the cleaning operation. In a normal driving state where no step or the like is detected, such as from period T1 to period T3 (hereinafter referred to as "normal driving state"), a YES determination is made in step S1, and the cleaning operation is continued (step S2).

[0060] During the cleaning operation, measurements are performed by the light reflection intensity sensor 5, the wheel speed sensor 6, and the IMU 7, and the respective measurement values are transmitted to the determination means 8 (step S3).

[0061] In step S4, a process is executed to estimate whether the floor in the traveling direction has a height difference that allows travel and to determine whether travel is possible. Specifically, the determination means 8 uses the time-series data DT (DT1 to DT3), which are the measurement values of the light reflection intensity sensor 5, the wheel speed sensor 6, and the IMU 7 measured for a certain period in the past starting from the determination time point, to estimate whether the floor in the traveling direction has a height difference that allows travel and to determine whether travel is possible.

[0062] Note that the length of the certain period can be arbitrarily set and is not particularly limited. Specifically, the certain period is set based on resources available for the estimation process of whether there is a passable height difference, the sampling rate of each measurement means, the required accuracy of the estimation process, and the like.

[0063] In FIG. 5, when the robot cleaner 1 reaches a step, the clearance decreases by the amount of the step, so the detection value of the light reflection intensity sensor 5 rises slightly upward (see period T4). Also, it can be seen that there is some disturbance in the output waveforms of the wheel speed sensor 6 and the IMU 7 as the robot cleaner passes over the step.

[0064] Even during this period, the determination means 8 uses the time-series data DT (DT1 to DT3) to comprehensively consider the change and determine whether it can be overcome. In this example, it is determined that it can be overcome (NO in step S4), and the robot cleaner 1 continues the cleaning operation. That is, the loop operation from step S1 to S4 is continued.

[0065] During the step-over period (refer to period T5), the output waveform of the light reflection intensity sensor 5 detects that the distance between the lower surface of the robot vacuum cleaner 1 and the floor surface has significantly increased. However, the determination means 8 comprehensively considers the time-series data DT (DT1 to DT3) and determines that the step can be stepped over, and a process of stepping over the step is executed. That is, the loop operation from steps S1 to S4 continues.

[0066] After that, for example, when it is determined in the control means 9 that the cleaning operation is not to be continued, such as when the cleaning is completed, it becomes NO in step S1, and an end process such as returning to the dust box or the charging facility is executed (step S6).

[0067] -Operation example (Part 2)- Next, the traveling operation (cleaning operation) shown in FIG. 6 will be described with reference to the flowchart of FIG. 4. Here, the description will focus on the differences from the operation example of FIG. 5, and duplicate descriptions may be omitted.

[0068] As shown in the upper part of FIG. 6, in this operation example, the point that there is a step (for example, H = 15 mm) that the robot vacuum cleaner 1 can step over in the forward direction of travel is common with FIG. 5, but the point that the material of the floor covering is different before and after the step is different from FIG. 5. Specifically, it is assumed that the infrared reflectivity of the floor covering FA before the step is higher than the infrared reflectivity of the floor covering FB after the step. Note that in the machine learning model, learning based on the change from the floor covering FA to the floor covering FB has been performed in advance.

[0069] First, in the normal traveling state of periods T1 to T3, it operates in the same manner as in the case of FIG. 5. That is, a process of looping steps S1 to S4 is executed.

[0070] In the example of FIG. 6, when the light reflection intensity sensor 5a of the robot vacuum cleaner 1 passes over the step, the material of the floor covering has changed to one with a low reflectivity.

[0071] Similar to the case of FIG. 5, in step S4, the determination means 8 determines whether it is possible to overcome the change amount by comprehensively considering the time-series data DT (DT1 to DT3). As shown in FIG. 6, since the material of the floor covering has a low reflectance, the output waveform of the light reflection intensity sensor 5a changes greatly.

[0072] In this example, in the determination means 8, it is determined that the step can be overcome (NO in step S4), and the robot cleaner 1 continues the cleaning operation. That is, the loop operation from steps S1 to S4 is continued.

[0073] Thereafter, for example, when it is determined by the control means 9 that the cleaning operation is not to be continued, such as when the cleaning is completed, it becomes NO in step S1, and an end process such as returning to the dust box or the charging facility is executed (step S6).

[0074] -Operation example (No. 3)- Next, the traveling operation (cleaning operation) shown in FIG. 7 will be described with reference to the flowchart of FIG. 4. Here, the description will focus on the differences from the operation example of FIG. 6, and duplicate descriptions may be omitted.

[0075] As shown in the upper part of FIG. 7, this operation example is different from FIG. 6 in that there is a step (for example, H = 30 mm) in front of the traveling direction where the robot cleaner 1 cannot descend.

[0076] First, in the normal traveling state during periods T1 to T3, the operation is the same as in the case of FIG. 6. That is, the process of looping steps S1 to S4 is executed.

[0077] In the example of FIG. 7, when the light reflection intensity sensor 5a of the robot cleaner 1 passes through the step, the material of the floor covering changes to one with a low reflectance, and there is a downward step.

[0078] Similar to the case of FIG. 6, in step S4, the determination means 8 determines whether it is possible to overcome the change amount by comprehensively considering the time-series data DT (DT1 to DT3). In the example of FIG. 7, the determination means 8 determines that there is a step where the robot cleaner 1 cannot descend, that is, it is determined that progress is impossible, considering the difference between the floor material FA and the floor material FB.

[0079] Then, it becomes NO in step S4, and the control means 9 executes travel control according to the travel location. Specifically, the control means 9 executes control to stop, reverse, or change the direction of travel of the robot cleaner 1, for example.

[0080] Thereafter, for example, in the control means 9, it is determined whether to continue the cleaning operation. If it is determined not to continue the cleaning operation, it becomes NO in step S1, and end processing such as returning to the dust box or the charging facility is executed (step S6).

[0081] As described above, the robot cleaner 1 according to the present embodiment is configured to include a determination means 8 that estimates whether the floor in the traveling direction has a height difference that allows progress and determines whether progress is possible using the time-series data of the measurement values of the light reflection intensity sensor 5 for a certain period, the time-series data of the measurement values of the wheel speed sensor 6 for a certain period, and the time-series data of the measurement values of the IMU 7 for a certain period.

[0082] That is, in the present embodiment, the possibility of progress is comprehensively determined using the time-series data for a certain period regarding the light reflection intensity of the floor surface in the traveling direction, the traveling speed information of the robot cleaner 1, and the attitude and inertial information of the robot body 2.

[0083] Thereby, it is possible to improve the accuracy of progress determination corresponding to the presence or absence of steps and changes in floor materials.

[0084] As described above, embodiments have been described as examples of the technologies disclosed in the present application. However, the technologies in the present disclosure are not limited to this, and are also applicable to embodiments in which appropriate changes, replacements, additions, omissions, etc. are made. It is also possible to combine the respective components described in each of the above embodiments to form a new embodiment.

[0085] For example, in the operation example of the above embodiment, an example has been described in which the measured values of the light reflection intensity sensor 5, the wheel speed sensor 6, and the IMU 7 are used as time-series data, but it is not limited to this. For example, the determination means 8 may estimate whether the floor in the traveling direction has a height difference that allows travel based on the measured value of the light reflection intensity sensor 5 and the attitude information of the robot body 2 measured by the IMU 7 or the like, and determine whether travel is possible.

[0086] Also, in the above embodiment, an example has been described in which a learned machine learning model is used in the determination process of the determination processing unit, but it is not limited to this. For example, instead of the learned machine learning model, a program for statistical processing may be used, or a formula-based algorithm other than the machine learning model may be used.

Description of Reference Numerals

[0087] 1 Robot vacuum cleaner 2 Robot body 4 Traveling means 4a Driving wheel 5 Light reflection intensity sensor (reflection intensity measurement means) 6 Wheel speed sensor (speed measurement means) 7 IMU (attitude measurement means, inertial measurement means) 8 Determination means

Claims

1. A self-propelled robotic vacuum cleaner, comprising a robot body that travels on the floor in the traveling direction by traveling means, reflection intensity measuring means for measuring the reflection intensity of light from the floor in the traveling direction, posture measuring means for measuring the posture of the robot body, and determination means for estimating whether the floor in the traveling direction has a height difference that allows travel and determining whether travel is possible, using first time-series data that is time-series data of measurement values by the reflection intensity measuring means for a certain period and second time-series data that is time-series data of measurement values by the posture measuring means for a certain period. A robotic vacuum cleaner characterized by the above.

2. In the robotic vacuum cleaner according to Claim 1, it is provided with speed measuring means for measuring the traveling speed of the robotic vacuum cleaner or inertia measuring means for measuring the inertia of the robot body, and the determination means uses third time-series data that is time-series data of measurement values by the speed measuring means for a certain period or fourth time-series data that is time-series data of measurement values by the inertia measuring means for a certain period for estimating whether the floor in the traveling direction has a height difference that allows travel. A robotic vacuum cleaner characterized by the above.

3. In the robotic vacuum cleaner according to Claim 1, the traveling means includes a driving wheel driven by a motor and a driven wheel rotatably provided behind the driving wheel in the traveling direction, and the reflection intensity measuring means includes a light reflection intensity sensor provided on the traveling direction side of the driving wheel. A robotic vacuum cleaner characterized by the above.

4. In the robotic vacuum cleaner according to Claim 1, the reflection intensity measuring means includes a plurality of light reflection intensity sensors arranged with a shift in the position in the traveling direction. A robotic vacuum cleaner characterized by the above.

5. In the robotic vacuum cleaner according to any one of Claims 1 to 4, the determination means includes a trained machine learning model for estimating whether the floor in the traveling direction has a height difference that allows travel, using the first time-series data and the second time-series data. A robotic vacuum cleaner characterized by the above.

Images