WO2020059292A1

WO2020059292A1 - Autonomous traveling cleaner

Info

Publication number: WO2020059292A1
Application number: PCT/JP2019/029120
Authority: WO
Inventors: 浅井　幸治; 杉本　博子
Original assignee: パナソニックＩｐマネジメント株式会社
Priority date: 2018-09-21
Filing date: 2019-07-25
Publication date: 2020-03-26
Also published as: JP2020047188A

Abstract

An autonomous traveling cleaner (100) comprises an imaging unit (160) for imaging an object present in the vicinity, and a drive mechanism (170) capable of changing the orientation of the imaging unit (160) about at least one axis along a horizontal plane. Furthermore, the autonomous traveling cleaner (100) comprises: an estimation unit for estimating the position of the autonomous traveling cleaner (100) on the basis of image information obtained from the imaging unit (160); and a drive control unit for controlling the drive mechanism (170) so that the accuracy of estimation of the position of the autonomous traveling cleaner (100) by the estimation unit rises, and changing the orientation of the imaging unit (160). There is thereby provided an autonomous traveling cleaner (100) that is capable of detecting obstacles and estimating the position of the autonomous traveling cleaner (100) using the imaging unit (160), e.g., at least one camera having a narrow angle of view.

Description

Autonomous traveling vacuum cleaner

The present invention relates to an autonomous traveling cleaner for autonomously traveling and cleaning a floor.

In recent years, based on the information from the camera provided by the vacuum cleaner, based on its own movement and its positional relationship with the surroundings, it performs self-position estimation to grasp which position in the room, and based on that information, Autonomous traveling vacuum cleaners that determine the operation content have been proposed.

Also, there is disclosed an autonomous traveling vacuum cleaner that calculates a distance from a shooting point of a camera to an object included in the image from an image obtained by the camera (for example, see Patent Document 1).

自律 Specifically, the autonomous traveling vacuum cleaner disclosed in Patent Literature 1 estimates a self-position by one camera and detects an obstacle. At this time, for example, when an obstacle is detected by imaging a floor surface and the own position is estimated from the obtained feature points, if the user turns, the self position is easily lost. On the other hand, if a camera with a wide angle of view is used to image obstacles existing on the ceiling and floor at one time, the self-position can be estimated using the feature points existing on the ceiling. Can be avoided. However, in this case, since the distortion of the obtained image becomes too large, a large error is included in the value of the distance to the obstacle, and the error of the self-position estimation becomes large.

Japanese Patent No. 3215616

The present invention provides an autonomous traveling vacuum cleaner capable of estimating its own position and imaging an obstacle on the floor with a single imaging unit.

One embodiment of the present invention is an autonomous traveling cleaner that autonomously travels and performs cleaning. The autonomous traveling cleaner has an image capturing unit that captures an image of an object present in the surroundings, a driving mechanism that drives the attitude of the image capturing unit to be changeable at least around one axis along a horizontal plane, and image information obtained from the image capturing unit. And an estimating unit for performing self-position estimation. Furthermore, a drive control unit that controls the drive mechanism and changes the attitude of the imaging unit is provided.

Note that each process by the autonomous traveling cleaner may be realized by a system, a method, an integrated circuit, a computer program, or a computer-readable recording medium such as a CD-ROM. Further, each processing by the autonomous traveling cleaner may be realized by an arbitrary combination of a system, a method, an integrated circuit, a computer program, and a recording medium.

According to the present invention, it is possible to provide an autonomous traveling cleaner capable of accurately detecting an obstacle existing on a floor surface and accurately estimating its own position with one imaging unit.

FIG. 1 is a plan view showing the autonomous traveling cleaner according to the embodiment from above. FIG. 2 is a plan view showing the autonomous traveling vacuum cleaner from below. FIG. 3 is a side view showing the autonomous traveling cleaner from a side. FIG. 4 is a block diagram showing a functional configuration of the autonomous traveling vacuum cleaner together with a mechanical configuration and the like. FIG. 5 is a flowchart showing the operation of the autonomous traveling vacuum cleaner.

Hereinafter, embodiments of the autonomous traveling vacuum cleaner according to the present invention will be described in detail with reference to the drawings. Each of the embodiments described below shows a specific example of the present invention. Therefore, the numerical values, shapes, materials, constituent elements, arrangement and connection forms of the constituent elements, the steps and the order of the steps, and the like shown in the following embodiments are merely examples, and do not limit the present invention. Therefore, among the components in the following embodiments, components that are not described in the independent claims that represent the highest concept of the present invention are described as arbitrary components.

図 Moreover, each drawing is a schematic diagram and is not necessarily strictly illustrated. In each drawing, the same components are denoted by the same reference numerals.

In the following embodiments, directions indicating directions are used. For example, the orthogonal direction means not only completely orthogonal, but also substantially orthogonal, that is, including an angle difference of, for example, about several percent. The same applies to expressions using other directions.

(Embodiment)
Hereinafter, an autonomous traveling cleaner 100 according to an embodiment of the present invention will be described with reference to FIGS. 1 to 3.

FIG. 1 is a plan view showing the autonomous traveling vacuum cleaner 100 according to the present embodiment from above. FIG. 2 is a plan view showing the autonomous traveling cleaner 100 from below. FIG. 3 is a side view showing the autonomous traveling cleaner 100 from the side. The autonomous traveling vacuum cleaner 100 according to the embodiment is a robot-type vacuum cleaner that autonomously travels on a floor of a house having a ceiling and sucks dust on the floor.

Specifically, as shown in FIGS. 1 to 3, the autonomous traveling cleaner 100 of the present embodiment includes a main body 110, a driving unit 120, a cleaning unit 130, a suction unit 140, and a control unit 150. , An imaging unit 160, a driving mechanism 170, a battery 180 (see FIG. 4), a storage unit 190 (see FIG. 4), and the like. The drive unit 120 moves the main body 110 on the floor. The cleaning unit 130 collects refuse existing on the floor. The suction unit 140 sucks dust collected by the cleaning unit 130 into the main body 110. The control unit 150 controls operations of the drive unit 120, the cleaning unit 130, the suction unit 140, and the like.

(4) The main body 110 forms a housing that houses the drive unit 120, the control unit 150, and the like. The main body 110 includes a lower part and an upper part, and the upper part is configured to be removable with respect to the lower part. The main body 110 includes a bumper (not shown) provided on the outer peripheral portion so as to be displaceable with respect to the main body 110. Further, the main body 110 has a suction port 111 at a lower portion for sucking dust into the main body 110.

Drive unit 120 causes autonomous traveling cleaner 100 to travel on the floor based on an instruction from control unit 150. In the present embodiment, one drive unit 120 is arranged on each of the left and right sides with respect to the center in the width direction of main body 110 in plan view. The number of drive units 120 is not limited to two, but may be one, or three or more.

The drive unit 120 includes a wheel (not shown) that travels on the floor, a travel motor that applies torque to the wheel, a housing that houses the travel motor, and the like. The wheel is housed in a recess (not shown) formed on the lower surface of the lower part of the main body 110 and is rotatably attached to the main body 110.

Note that the drive unit 120 may include an encoder 121 (see FIG. 4) for detecting odometry information used for calculating the traveling direction of the autonomous traveling cleaner 100. In addition, hereinafter, the traveling direction of the autonomous traveling vacuum cleaner 100 (the direction in which the autonomous traveling vacuum cleaner 100 travels) is described not only in the direction in which the vehicle is actually traveling but also in the direction including the direction in which the vehicle travels.

The autonomous traveling cleaner 100 is configured by a two-wheel opposed drive system including the casters 129 as auxiliary wheels. By independently controlling the rotation of the two wheels, the autonomous traveling cleaner 100 can travel freely, such as going straight, retreating, turning left, turning right, and so on.

The cleaning unit 130 is a unit for sucking dust from the suction port 111. The cleaning unit 130 includes a main brush (not shown) disposed in the suction port 111, a brush drive motor for rotating the main brush, and the like.

The suction unit 140 is arranged inside the main body 110. The suction unit 140 includes a fan case (not shown), an electric fan disposed inside the fan case, and the like. The electric fan sucks air inside the trash box unit 141 and discharges the air outside the main body 110. Thus, the refuse sucked from the suction port 111 is stored in the trash box unit 141.

The control unit 150 is arranged inside the main body 110. The control unit 150 includes a memory, a CPU (Central Processing Unit) that executes a control program stored in the memory, and the like. Each processing unit realized by the control unit 150 will be described later.

The image capturing section 160 is configured by a so-called digital camera or the like that captures an image of an object such as an obstacle existing around the main body 110 and outputs image information composed of digital signals. The imaging unit 160 according to the present embodiment includes a single-lens optical system having a single optical axis. Therefore, the image information obtained from the imaging unit 160 does not include information indicating the distance from the imaging unit 160 to the object.

画 The angle of view of the optical system included in the imaging unit 160 is selected from, for example, 90 degrees or less. Therefore, the imaging unit 160 employs an optical system that cannot capture an image of the entire circumference of the upper space of the main body 110. Accordingly, distortion of an image obtained from the imaging unit 160, a difference in resolution depending on a position in the image, and the like are suppressed.

The drive mechanism 170 includes a mechanism for moving the posture of the imaging unit 160 at least around one axis along a horizontal plane, as shown in FIG. In addition, the drive mechanism 170 of the present embodiment surrounds the first rotation axis extending in the width direction (the X-axis direction in FIG. 3) of the autonomous traveling cleaner 100 as indicated by a broken arrow in FIG. 3. , The posture of the imaging unit 160 is changed. Further, the drive mechanism 170 is configured to change the attitude of the imaging unit 160 around a second rotation axis extending in a vertical direction indicated by a chain line in FIG. Note that the driving mechanism 170 may be configured so that the attitude of the imaging unit 160 can be changed around a rotation axis extending in other directions other than the X-axis direction and the vertical direction. FIG. 3 illustrates and illustrates an example in which the imaging unit 160 and the driving mechanism 170 are provided on the top surface of the main body 110 so as to protrude. For example, a configuration may be adopted in which a concave portion is provided on the top surface of the main body 110, and the imaging section 160 and the driving mechanism 170 are installed in the concave portion. Thus, since the orientation of the imaging unit 160 can be known, it is possible to acquire image information from which self-position estimation can be performed more accurately.

Battery 180 (see FIG. 4) is electrically connected to the electronic device included in autonomous traveling vacuum cleaner 100. The battery 180 is a battery, such as a secondary battery, for supplying power to the connected electronic device.

The storage unit 190 is a memory such as a ROM (Read Only Memory) or a RAM (Random Access Memory) in which control programs executed by various processing units are stored. The storage unit 190 is realized by, for example, a hard disk drive (HDD), a flash memory, or the like.

(4) The storage unit 190 stores, for example, correlation information, size information, a floor map, and the like. Specifically, the correlation information is information on the estimated size of the pixel in the real space with respect to the pixel position of the image generated by the imaging unit 160. The size information is information on the size of the autonomous traveling vacuum cleaner 100 (specifically, the width of the main body 110, etc.). The floor map is information indicating a floor plan of a house or the like. Note that the lateral width of the main body 110 is the width of the autonomous traveling cleaner 100 in a direction orthogonal to the traveling direction of the autonomous traveling cleaner 100 when viewed from above.

Further, the autonomous traveling vacuum cleaner 100 includes various sensors such as an obstacle sensor 273, a distance measurement sensor 274, a collision sensor (not shown), a floor sensor 276, and a dust amount sensor 277, which are exemplified below. .

The obstacle sensor 273 is a sensor that detects an obstacle such as a peripheral wall or furniture existing in front of the main body 110, which is an obstacle to traveling. In the present embodiment, as the obstacle sensor 273, for example, an ultrasonic sensor is used. The obstacle sensor 273 includes a transmitting unit 271, a receiving unit 272, and the like. Transmitting section 271 is disposed at the front center of main body 110 and transmits ultrasonic waves forward. The receiving units 272 are arranged on both sides of the transmitting unit 271 in the X direction, and receive the ultrasonic waves transmitted from the transmitting unit 271. That is, the obstacle sensor 273 causes the receiving unit 272 to receive the reflected ultrasonic wave transmitted from the transmitting unit 271 and reflected by the obstacle and returned. Thereby, the obstacle sensor 273 detects the distance to the obstacle, the position of the obstacle, and the like.

The distance measuring sensor 274 is a sensor that detects a distance between the main body 110 and an object such as an obstacle existing around the main body 110. In the present embodiment, the distance measuring sensor 274 is configured by, for example, an infrared sensor having a light emitting unit and a light receiving unit. That is, the distance measurement sensor 274 measures the distance to the obstacle based on the elapsed time from when the infrared light emitted from the light emitting unit and reflected by the obstacle returns and is received by the light receiving unit.

Specifically, the distance measurement sensor 274 is arranged, for example, near the right front top and near the left front top. The distance measuring sensor 274 on the right side outputs light (infrared rays) toward the diagonally right front of the main body 110. The distance measuring sensor 274 on the left side outputs light (infrared light) toward the front left side of the main body 110. With this configuration, when the autonomous traveling cleaner 100 is turning, the distance measurement sensor 274 detects the distance between the main body 110 and the surrounding object closest to the contour of the main body 110.

The floor sensor 276 is a sensor that is disposed at a plurality of locations on the bottom surface of the main body 110 and detects the state of the floor. In the present embodiment, floor sensor 276 is formed of, for example, an infrared sensor having a light emitting unit and a light receiving unit. That is, the floor surface sensor 276 detects the state of the floor surface, for example, the floor surface is wet, based on the amount of light (infrared light) returned from the light emitting unit and received by the light receiving unit.

The dust amount sensor 277 is, for example, a sensor including a light-emitting element and a light-receiving element, and the amount of light emitted from the light-emitting element is detected by the light-receiving element and output. At this time, the control unit 150 associates the amount of received light with the amount of dust based on the output information. Specifically, control unit 150 determines that the smaller the amount of received light, the greater the amount of dust. Then, the control unit 150 generates dust amount information indicating that.

The obstacle sensor 273, the distance measurement sensor 274, the floor sensor 276, and the dust amount sensor 277 described above are examples of sensors. Therefore, the autonomous traveling vacuum cleaner 100 does not need to include all the above sensors. In addition, the autonomous traveling cleaner 100 may include a sensor having a different form from the above.

For example, the autonomous traveling cleaner 100 may further include a collision sensor, an encoder 121 (see FIG. 4), an acceleration sensor, an angular velocity sensor, and the like, which are not shown.

The collision sensor is composed of, for example, a switch contact displacement sensor, and is disposed on a bumper attached around the main body 110. The switch contact displacement sensor is turned on when the obstacle comes into contact with the bumper and the bumper is pushed into the main body 110.

The encoder 121 is disposed in the drive unit 120. The encoder 121 detects each rotation angle of a pair of wheels rotated by the traveling motor. Based on the information from the encoder 121, for example, the traveling amount, the turning angle, the speed, the acceleration, the angular speed, and the like of the autonomous traveling cleaner 100 are calculated.

The acceleration sensor detects the acceleration when the autonomous traveling cleaner 100 travels.

The angular velocity sensor detects an angular velocity when the autonomous traveling cleaner 100 turns.

情報 The information detected by the acceleration sensor and the angular velocity sensor is used, for example, as information for correcting an error caused by idling of the wheel.

Next, each processing unit realized by the control unit 150 will be described with reference to FIG.

FIG. 4 is a block diagram showing a functional configuration of the autonomous traveling cleaner 100 according to the embodiment, together with a mechanical configuration and the like.

As shown in FIG. 4, the autonomous traveling cleaner 100 includes, as processing units executed by the control unit 150, an estimation unit 151, a drive control unit 152, a type identification unit 153, a traveling control unit 154, and a cleaning control unit. 155 and the like. In the present embodiment, the control unit 150 is described as one unit in one functional block. However, each processing unit is divided into a plurality of functional blocks, and is realized by a plurality of CPUs corresponding to the respective functional blocks. You may.

The estimation unit 151 is a processing unit that performs self-position estimation based on image information obtained from the imaging unit 160. The method of self-position estimation is not particularly limited, but in the case of the present embodiment, estimation section 151 performs self-position estimation using SLAM (Simultaneous Localization and Mapping) technology. Specifically, the estimation unit 151 includes a plurality of feature points (landmarks) included in the image information acquired from the imaging unit 160, posture information indicating the posture of the imaging unit 160 obtained from the drive control unit 152, and a drive unit. The self-position estimation is executed by integrating signals from the encoder 121 incorporated in the sub-station 120. Then, based on the estimated self-position, the estimating unit 151 generates self-position information.

{Circle around (5)} The estimating unit 151 performs the self-position estimation while traveling. As a result, the estimating unit 151 also generates a map using the traveling results that are sets of self-position information generated at a plurality of locations. At this time, the position (area) of an obstacle such as a wall may be added as information to the map using information detected by various sensors.

The drive control unit 152 is a processing unit that controls the drive mechanism 170 so as to increase the accuracy of the self-position estimation performed by the estimation unit 151 and changes the attitude of the imaging unit 160. When detecting an obstacle existing on the floor based on the image information obtained from the imaging unit 160, the drive control unit 152 sets the orientation of the imaging unit 160 so that the floor is included in the angle of view. To change.

In the following, the orientation of the imaging unit 160 for self-position estimation may be described as “estimated orientation”, and the orientation of the imaging unit 160 for obstacle detection may be described as “detected orientation”.

Here, the estimated posture of the imaging unit 160 at which the accuracy of the self-position estimation becomes high changes depending on the surrounding environment and the situation of the autonomous traveling cleaner 100. In this case, for example, the posture in which the number of feature points included in one piece of image information obtained from the imaging unit 160 is large is set as the estimated posture of the imaging unit 160. This tends to increase the accuracy of the self-position estimation. Thus, the drive control unit 152 controls the drive mechanism 170 so that the angle of view includes many feature points, and changes the estimated attitude of the imaging unit 160.

In the present embodiment, the method for determining the estimated posture is not particularly limited. As a specific method of determining the estimated posture, first, for example, the drive control unit 152 controls the drive mechanism 170 so that the posture of the imaging unit 160 changes in a predetermined order or randomly. Next, image information is sequentially acquired from the imaging unit 160. Then, when the feature point in the image information exceeds the first threshold value, for example, the posture in which the image information is acquired such that the feature point exceeds 1000 may be determined as the estimated posture of the imaging unit 160.

In addition, when the type specifying unit 153 obtains information indicating that ceiling illumination is included in the image information, the type in which the image information is obtained may be determined as the estimated attitude. Furthermore, the posture at which the image information of the ceiling illumination image at the center is obtained may be determined as the estimated posture.

(5) The drive control unit 152 controls the drive mechanism 170 so that the imaging unit 160 can acquire a plurality of pieces of image information covering the whole sphere or a range close to the whole sphere. Then, a posture in which the number of feature points within the angle of view of the imaging unit 160 exceeds the first threshold may be determined as the estimated posture of the imaging unit 160. Further, the posture having the most feature points within the angle of view of the imaging unit 160 may be determined as the estimated posture of the imaging unit 160. That is, as described above, the estimated attitude of the imaging unit 160 is determined using the number of feature points included in the angle of view. Thereby, for example, an estimation result obtained from processing such as self-position estimation performed by the estimation unit 151 can be obtained with high accuracy.

In addition, the drive control unit 152 controls the drive mechanism 170 so that the optical axis of the imaging unit 160 is directed upward so that the floor surface does not enter the angle of view in order to determine an estimated posture including many feature points. Alternatively, the image capturing section 160 may acquire a plurality of pieces of image information having different areas. Thus, the number of pieces of image information acquired to increase the number of feature points included in the image information can be suppressed. As a result, the processing required for determining the estimated attitude can be further speeded up.

In addition, the drive control unit 152 may control the drive mechanism 170 so that the type specifying unit 153 described below specifies, for example, ceiling illumination, and the image capturing unit 160 may acquire image information. Thus, the number of pieces of image information acquired to increase the number of feature points included in the image information can be suppressed. As a result, the processing required for determining the estimated attitude can be further speeded up.

On the other hand, the estimated posture of the imaging unit 160 with which the accuracy of the self-position estimation is high is selected as the estimated posture having a large number of feature points that can be matched based on, for example, a plurality of pieces of image information obtained from the imaging unit 160. Is also good. As a result, the accuracy of the self-position estimation increases. That is, the drive control unit 152 controls the drive mechanism 170 so that the angle of view includes many feature points that can be matched, and changes the imaging unit 160 to the selected estimated posture.

Specifically, the drive control unit 152 determines that the area overlaps in, for example, a celestial sphere, a range close to the celestial sphere, a range where the floor surface does not fall within the angle of view, or a range where ceiling illumination is included within the angle of view. The driving mechanism 170 is controlled so that the image capturing unit 160 can acquire a plurality of pieces of image information. Then, the drive control unit 152 determines the posture in which the number of feature points that can be matched within the angle of view of the imaging unit 160 exceeds the second threshold, for example, the case where the number of feature points that can be matched exceeds 100, as the estimated posture. May be. Furthermore, the drive control unit 152 may determine, as the estimated posture, the posture having the most matching feature points within the angle of view of the imaging unit 160.

The type specifying unit 153 described above is a processing unit that specifies the type of a captured object from image information obtained from the imaging unit 160. In addition, the method of specifying the type is not particularly limited. As a method for specifying the type, for example, an AI (Artificial @ Intelligence) technique can be exemplified. Specifically, the image capturing unit 160 first acquires a plurality of pieces of image information in advance in an environment (including a pseudo environment) in which the autonomous traveling cleaner 100 performs cleaning. At this time, the type specifying unit 153 prepares the image of the object in the acquired image information and the type of the object corresponding to the image as data in the storage unit 190 or the like. Next, the type specifying unit 153 learns a situation such as an object in an environment (including a pseudo environment) by using a technique such as deep learning. Then, when the acquired image includes the image of the learned object, the type identification unit 153 constructs a type identification model so as to output the type of the object. At this time, the type specifying unit 153 may prepare, in addition to the image of the object in the image information and the type of the object corresponding to the image, the image position where the object is displayed as data. Accordingly, when the acquired image includes the image of the learned object, the type specifying unit 153 can also construct a model that can output the displayed image position along with the type of the object. Further, the actual distance between the object and the main body 110 can be estimated to some extent based on the displayed image position. That is, the type specifying unit 153 can exemplify a case in which the type of an object is specified using each model constructed as described above.

特定 Specifying the type of the object includes not only specifying the detailed type of the object but also specifying whether the object can be an obstacle to the traveling of the autonomous traveling vacuum cleaner 100 or not. Specifically, for example, a window image appearing in the image information is not specified as an obstacle. On the other hand, the case where the image of the small carpet on the floor is specified as an obstacle is exemplified.

In other words, when detecting an obstacle that hinders the travel of the autonomous traveling vacuum cleaner 100, the type specifying unit 153 specifies the type of the object based on the image information obtained from the imaging unit 160 in the detected posture. As a result, the type identification unit 153 detects a moving obstacle. At this time, the type identification unit 153 is based on information on the angle of the imaging unit 160 in the detected attitude acquired from the drive control unit 152, odometry information obtained from the encoder 121 and the like, the position of the identified obstacle in the image information, and the like. Then, the distance to the obstacle may be calculated. Thus, the operation of the autonomous traveling vacuum cleaner 100 can be changed depending on the type of the obstacle and the obstacle to be cleaned up and the obstacle to be avoided without approaching.

走行 Furthermore, the traveling control unit 154 is a processing unit that controls the traveling operation of the autonomous traveling vacuum cleaner 100. Specifically, for example, the self-position of the autonomous traveling cleaner 100 is estimated from information detected by various sensors including the imaging unit 160 included in the autonomous traveling cleaner 100. Then, the traveling control unit 154 controls the drive unit 120 based on the estimated self-position of the autonomous traveling cleaner 100. Thereby, the traveling control unit 154 controls traveling and steering of the autonomous traveling cleaner 100 to move on the floor.

(4) The cleaning control unit 155 controls the suction unit 140 to suck dust from the suction port 111 of the main body 110. As a result, the floor is cleaned.

Hereinafter, the operation of the autonomous traveling vacuum cleaner 100 will be described with reference to FIG. FIG. 5 is a flowchart showing the operation of the autonomous traveling vacuum cleaner 100.

First, as illustrated in FIG. 5, the drive control unit 152 of the autonomous traveling cleaner 100 controls the drive mechanism 170 to change the attitude of the imaging unit 160 so that the floor to be cleaned is included in the angle of view. The detection posture is set (step S101).

Next, the type identification unit 153 detects whether there is an obstacle in the image information acquired from the imaging unit 160 (Step S102). At this time, when there is an obstacle (Yes in step S102), the type identification unit 153 estimates the distance to the obstacle (step S103). On the other hand, when no obstacle is detected (No in step S102), the type identification unit 153 sets the estimated distance to the maximum detection distance (for example, about 1 meter) (step S104), and proceeds to step S105. Thereby, the traveling operation can be performed while being ticked for each detection maximum distance. In addition, the detection position accuracy for a distant obstacle tends to be low. Therefore, by setting the detection maximum distance, it is possible to accurately catch an obstacle and travel while correcting a decrease in the detection position accuracy.

Next, the drive control unit 152 controls the drive mechanism 170 so that the optical axis of the imaging unit 160 is at least along the horizontal plane as described with reference to FIG. 3 so that the floor surface is not included in the angle of view. It moves (rotates) around one axis (for example, around the first rotation axis) and turns upward (step S105). At this time, the optical axis of the imaging unit 160 may be moved (rotated) around the second rotation axis.

Next, the estimating unit 151 acquires image information obtained by imaging a plurality of different areas from the imaging unit 160, and determines whether or not the imaging unit 160 is in the estimated posture (step S106). At this time, if the posture is not the estimated posture (No in step S106), the process returns to step S105, and the subsequent steps are similarly executed. On the other hand, when the estimated posture is reached (Yes in step S106), the estimating unit 151 determines the estimated posture of the imaging unit 160. Note that an arbitrary method such as the method described above can be adopted as a method for determining the estimated posture.

Next, the estimating unit 151 performs the self-position estimation based on the feature points in the image information obtained from the imaging unit 160 in the estimated posture (step S107).

Next, the estimating unit 151 continuously performs the self-position estimation while moving the autonomous traveling cleaner 100 on the floor by the drive unit 120, and performs creation and updating of the map as needed (step S108). Here, the drive control unit 152 causes the storage unit 190 to store a region imaged in the determined estimated posture of the imaging unit 160, the posture of the imaging unit 160, and the like. At this time, even if the autonomous traveling cleaner 100 moves on the floor, the estimated attitude of the imaging unit 160 may follow the traveling state of the autonomous traveling cleaner 100 so that the same area falls within the angle of view. Absent. As a result, even when the autonomous traveling cleaner 100 moves, the estimation unit 151 can maintain high accuracy of the self-position estimation. As a result, the estimation unit 151 can stably execute the self-position estimation with high accuracy.

Next, the estimating unit 151 determines whether or not the self-position has been lost during the cleaning (step S109). At this time, if the self-position has been lost (Yes in step S109), the process returns to step S105, and the subsequent steps are executed. Specifically, the drive control unit 152 controls the drive mechanism 170 again to change the attitude of the imaging unit 160 so as to increase the accuracy of the self-position estimation (step S105). Subsequently, the estimating unit 151 determines again the estimated posture of the imaging unit 160 (Step S106). Further, the respective steps of the above-described self-position estimation (step S107) and map creation (step S108) are executed.

On the other hand, if the self-position has not been lost (No in step S109), it is determined whether or not the cleaning of the predetermined area has been completed (step S110). At this time, if the cleaning has not been completed (step S110) (No in S110), the process returns to step S107, and the subsequent steps are executed.

On the other hand, if the cleaning has been completed (Yes in step S110), a series of flows ends.

自律 As described above, the autonomous traveling vacuum cleaner 100 operates.

As described above, the autonomous traveling cleaner 100 according to the present embodiment has at least one optical system that has a relatively narrow angle of view (for example, an angle of view of 90 degrees or less and 30 degrees or more) and a small distortion. The imaging unit 160 is provided. This makes it possible to detect a forward obstacle and calculate the distance to the obstacle with high accuracy. Further, the feature position included in the image information obtained from the imaging unit 160 having an appropriate estimated posture can perform the self-position estimation with high accuracy. As a result, the autonomous traveling vacuum cleaner 100 can move smoothly and completely without colliding with an obstacle, and can clean the obstacle and the corner of the floor cleanly.

The present invention is not limited to the above embodiment. For example, another embodiment that is realized by arbitrarily combining the components described in this specification and excluding some of the components may be an embodiment of the present invention. Further, the gist of the present invention with respect to the above-described embodiment, that is, modified examples obtained by performing various modifications conceivable by those skilled in the art without departing from the meaning indicated by the words described in the claims are also included in the present invention. It is.

Specifically, in the above-described embodiment, the flow in which the imaging unit 160 is set to the detection posture, the obstacle is detected, and then the cleaning is performed while executing the self-position estimation is described as an example. I can't. For example, first, the imaging unit 160 is set to the estimated posture, and the self-position estimation is performed. Thereafter, the imaging unit 160 is set to the detection posture, and the cleaning is performed while recognizing the obstacle and estimating the distance to the obstacle. Then, at a predetermined interval, such as after traveling a certain distance, the imaging unit 160 may be set to the estimated posture again and the cleaning may be performed while executing the self-position estimation.

In the above embodiment, when the self-position estimation is performed, first, the optical axis of the imaging unit 160 may be directed vertically upward. In this case, the probability that ceiling illumination is included in the angle of view of the imaging unit 160 increases. Therefore, the posture in which the ceiling illumination is located at the center may be used as the estimated posture of the imaging unit 160. Thereby, the feature point of the ceiling lighting can be always observed. Then, by performing self-position estimation based on the feature points of the ceiling lighting, the estimation accuracy can be improved.

Further, in the above embodiment, an example has been described in which the estimated attitude of the imaging unit 160 is determined based on the number of feature points included in one piece of image information and the number of feature points that can be matched. Not limited to For example, an image in which the total reliability of the feature amount is the maximum or the third threshold or more, for example, when the total reliability of the feature amount is obtained between 0 and 1, the total reliability is set to 100 or more. The position may be determined as the estimated posture.

Further, in the above embodiment, when it is desired to more accurately grasp the distance between the obstacle and the autonomous traveling cleaner 100, for example, when the user climbs on a carpet recognized as an obstacle and performs cleaning, the following autonomous operation is performed. The travel cleaner 100 may be controlled.

Specifically, first, the autonomous traveling cleaner 100 is moved to a position where the leading edge of the main body 110 of the autonomous traveling cleaner 100 and the obstacle fall within the angle of view of the imaging unit 160. Move closer to obstacles according to the distance information. Then, the drive control unit 152 controls the drive mechanism 170 such that the leading edge of the autonomous traveling cleaner 100 and the obstacle fall within the angle of view of the imaging unit 160. Thereby, the distance between the obstacle and the autonomous traveling cleaner 100 can be more accurately grasped, and the floor can be completely cleaned.

The present invention is applicable to an autonomous traveling vacuum cleaner that automatically cleans a floor surface in a house or the like.

Reference Signs List 100 autonomous traveling cleaner 110 main body 111 suction port 120 drive unit 121 encoder 129 caster 130 cleaning unit 140 suction unit 141 trash can unit 150 control unit 151 estimation unit 152 drive control unit 153 type identification unit 154 travel control unit 155 cleaning control unit 160 imaging Unit 170 Drive mechanism 180 Battery 190 Storage unit 271 Transmitting unit 272 Receiving unit 273 Obstacle sensor 274 Distance measuring sensor 276 Floor surface sensor 277 Dust amount sensor

Claims

An autonomous traveling vacuum cleaner that runs autonomously and performs cleaning,
An imaging unit for imaging an object present in the vicinity,
A drive mechanism that can change the attitude of the imaging unit at least around one axis along a horizontal plane;
An estimating unit that performs self-position estimation based on image information obtained from the imaging unit,
A drive control unit that controls the drive mechanism and changes the attitude of the imaging unit.
Autonomous traveling vacuum cleaner.
The drive control unit controls the drive mechanism, and changes the orientation of the imaging unit so that the angle of view has a large number of feature points included in the image information.
The autonomous traveling vacuum cleaner according to claim 1.
The drive control unit controls the drive mechanism, and based on the plurality of pieces of image information, changes the orientation of the imaging unit so that the number of feature points that can be matched has a large angle of view.
The autonomous traveling vacuum cleaner according to claim 1.
The drive control unit controls the drive mechanism so that the optical axis of the imaging unit is directed upward so that the floor surface does not enter the angle of view in order to increase the number of feature points included in the image information. ,
The autonomous traveling vacuum cleaner according to any one of claims 1 to 3.
A type identification unit that identifies the type of object imaged from the image information obtained from the imaging unit,
The drive control unit controls the drive mechanism, so that the type identification unit identifies the ceiling illumination, to change the attitude of the imaging unit,
The autonomous traveling vacuum cleaner according to any one of claims 1 to 4.
The drive control unit, when the estimating unit has lost its position, controls the drive mechanism to change the attitude of the imaging unit,
The autonomous traveling vacuum cleaner according to any one of claims 1 to 5.