CN108852184A - A kind of non-blind area sweeping robot and its cleaning control method based on deep learning algorithm - Google Patents
A kind of non-blind area sweeping robot and its cleaning control method based on deep learning algorithm Download PDFInfo
- Publication number
- CN108852184A CN108852184A CN201811071935.3A CN201811071935A CN108852184A CN 108852184 A CN108852184 A CN 108852184A CN 201811071935 A CN201811071935 A CN 201811071935A CN 108852184 A CN108852184 A CN 108852184A
- Authority
- CN
- China
- Prior art keywords
- blind area
- sweeping robot
- deep learning
- cleaning
- learning algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000010408 sweeping Methods 0.000 title claims abstract description 100
- 238000004140 cleaning Methods 0.000 title claims abstract description 53
- 238000013135 deep learning Methods 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 title claims description 14
- 230000007246 mechanism Effects 0.000 claims abstract description 40
- 238000010926 purge Methods 0.000 claims abstract description 27
- 238000001514 detection method Methods 0.000 claims abstract description 24
- 238000012545 processing Methods 0.000 claims abstract description 13
- 238000003384 imaging method Methods 0.000 claims abstract description 9
- 230000004927 fusion Effects 0.000 claims abstract description 4
- 230000004888 barrier function Effects 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 5
- 230000001133 acceleration Effects 0.000 claims description 4
- 238000003708 edge detection Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 claims description 2
- 239000003638 chemical reducing agent Substances 0.000 claims 2
- 230000007613 environmental effect Effects 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 4
- 239000000428 dust Substances 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000033001 locomotion Effects 0.000 description 3
- 238000005406 washing Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 230000002463 transducing effect Effects 0.000 description 2
- 241001417527 Pempheridae Species 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000009194 climbing Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000011065 in-situ storage Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000013011 mating Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000005086 pumping Methods 0.000 description 1
- 230000002000 scavenging effect Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
- A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
- A47L11/00—Machines for cleaning floors, carpets, furniture, walls, or wall coverings
- A47L11/24—Floor-sweeping machines, motor-driven
-
- A—HUMAN NECESSITIES
- A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
- A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
- A47L11/00—Machines for cleaning floors, carpets, furniture, walls, or wall coverings
- A47L11/40—Parts or details of machines not provided for in groups A47L11/02 - A47L11/38, or not restricted to one of these groups, e.g. handles, arrangements of switches, skirts, buffers, levers
-
- A—HUMAN NECESSITIES
- A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
- A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
- A47L11/00—Machines for cleaning floors, carpets, furniture, walls, or wall coverings
- A47L11/40—Parts or details of machines not provided for in groups A47L11/02 - A47L11/38, or not restricted to one of these groups, e.g. handles, arrangements of switches, skirts, buffers, levers
- A47L11/4002—Installations of electric equipment
-
- A—HUMAN NECESSITIES
- A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
- A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
- A47L11/00—Machines for cleaning floors, carpets, furniture, walls, or wall coverings
- A47L11/40—Parts or details of machines not provided for in groups A47L11/02 - A47L11/38, or not restricted to one of these groups, e.g. handles, arrangements of switches, skirts, buffers, levers
- A47L11/4011—Regulation of the cleaning machine by electric means; Control systems and remote control systems therefor
-
- A—HUMAN NECESSITIES
- A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
- A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
- A47L11/00—Machines for cleaning floors, carpets, furniture, walls, or wall coverings
- A47L11/40—Parts or details of machines not provided for in groups A47L11/02 - A47L11/38, or not restricted to one of these groups, e.g. handles, arrangements of switches, skirts, buffers, levers
- A47L11/4036—Parts or details of the surface treating tools
-
- A—HUMAN NECESSITIES
- A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
- A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
- A47L11/00—Machines for cleaning floors, carpets, furniture, walls, or wall coverings
- A47L11/40—Parts or details of machines not provided for in groups A47L11/02 - A47L11/38, or not restricted to one of these groups, e.g. handles, arrangements of switches, skirts, buffers, levers
- A47L11/4063—Driving means; Transmission means therefor
- A47L11/4069—Driving or transmission means for the cleaning tools
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0219—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory ensuring the processing of the whole working surface
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0257—Control of position or course in two dimensions specially adapted to land vehicles using a radar
-
- A—HUMAN NECESSITIES
- A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
- A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
- A47L2201/00—Robotic cleaning machines, i.e. with automatic control of the travelling movement or the cleaning operation
- A47L2201/04—Automatic control of the travelling movement; Automatic obstacle detection
-
- A—HUMAN NECESSITIES
- A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
- A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
- A47L2201/00—Robotic cleaning machines, i.e. with automatic control of the travelling movement or the cleaning operation
- A47L2201/06—Control of the cleaning action for autonomous devices; Automatic detection of the surface condition before, during or after cleaning
Abstract
The invention discloses a kind of non-blind area sweeping robots based on deep learning algorithm, including body shell, laser radar, controller, power supply, walking mechanism, sensing mechanism and cleaning agency；The laser radar is arranged in the top of body shell, for carrying out horizontal 360-degree spacescan to purging zone, and the point cloud data of the strength information of the echo-signal scanned and generation is transferred to controller；Controller, including Context awareness module, the Context awareness module is used to carry out the strength information of echo-signal and the data of point cloud data two-dimensional intensity picture and Range Profile that processing operation is cleaned region, utilize the blind area in algorithm of target detection detection intensity picture, and utilize three-dimensional imaging algorithm fusion intensity image and Range Profile, three-dimensional reconstruction is carried out to purging zone, is cleaned blind area type and orientation present in region；Cleaning agency, including round brush motor, round brush, micro-cleaner and scalable Bian Shua mechanism, scalable Bian Shua mechanism include sub-shell, while brush, while brush motor and connecting rod pushing meanss for realizing the scalable cleaning of side brush eliminate and clean blind area.
Description
Technical field
The present invention relates to wipe clean field of mechanical technique more particularly to a kind of non-blind area based on deep learning algorithm to sweep
Floor-washing robot and its cleaning control method.
Background technique
Existing domestic intelligent sweeping robot also known as intellective dust collector, frequently referred to precious and dog of sweeping the floor.It is by moving machine
Device people technology and cleaner technology are organically fused together, and realize that autonomous cleans the function of ground refuse.Domestic intelligent
Largely manufacture is independently manipulated by computer chip in remote controler and machine at flat circular, can also reserve cleaning sweeping robot.When
After machine startup, fuselage presses preset path, direction Mobile cleaning, and meeting automatic turning is hidden when on the way encountering barrier,
Until completing cleaning works；When not enough power supply machine can Automatic-searching cradle charge, it is fully charged after automatically into standby
State.
When sweeping robot works, machine makes the movements such as advance, turning, stop and requires to be predicted by multiple sensors certainly
It is determined after the relationship of body and ambient enviroment, such as differentiates that there is clear in front, if needs are avoided；There is un-grooved below fuselage
Class or step class may cause the state of ground of fuselage collision overturning etc..Since cost is relatively low for infrared sensor, Household floor-sweeping machine
The device National People's Congress all uses multiple groups infrared sensor.Since the range accuracy of infrared sensor is lower and can not detect the object of closer distance
Body, sweeping robot can be separated with certain distance when cleaning along edges such as wall sides with edge, and marginal position, which exists, cleans blind area.
Sweeping robot is mostly disk structure, since inner space needs to install the portions such as battery pack, dust-collecting box, motor
Part, in addition robot needs to have certain obstacle climbing ability, the height of sweeping robot is generally greater than 6 to 8 centimetres.Therefore it sweeps the floor
Robot cannot be introduced into the low place of spatial altitudes such as sofa bottom, bed bottom and be cleaned, can not to the shapes such as corner by
The place of limit is cleaned, and there are biggish cleaning blind areas.
Sweeping robot currently on the market is essentially all to realize pathfinding using random collision formula or path planning formula
It cleans.Compared to random collision pathfinding mode, path planning pathfinding mode has sweeping efficiency higher, and it is preferably excellent to clean effect
Point, but still can not accomplish to differentiate and clean blind area and still remained the problem of cleaning blind area, clean blind area.
Thus existing sweeping robot technology can't thoroughly be met the needs of users, it is also necessary to be improved, one
Kind can be realized detection positioning clean blind area and non-blind area sweeping robot product that blind area is cleaned become there is an urgent need to.
Summary of the invention
In order to solve the above technical problems, the object of the present invention is to provide a kind of non-blind areas based on deep learning algorithm to sweep the floor
Robot and its cleaning control method, the robot use algorithm of target detection and extension type scavenging machine based on deep learning
Structure, sweeping robot start blind area existing for detectable purging zone before cleaning and know its corresponding type and orientation, sweep the floor
Extension type cleaning agency can clean blind area in robot cleaning process, eliminate and clean blind area.
The purpose of the present invention is realized by technical solution below：
A kind of non-blind area sweeping robot based on deep learning algorithm, sweeps including the non-blind area based on deep learning algorithm
Floor-washing robot, which is characterized in that including body shell, laser radar, controller, power supply, walking mechanism, sensing mechanism and cleaning
Mechanism；It is described
Laser radar is arranged in the top of body shell, for carrying out horizontal 360-degree spacescan to purging zone, and
The point cloud data of the strength information of the echo-signal scanned and generation is transferred to controller；
Controller, including Context awareness module, the Context awareness module are used for strength information and point cloud to echo-signal
The data of data carry out the two-dimensional intensity picture and Range Profile that processing operation is cleaned region, are detected using algorithm of target detection strong
Blind area of the degree as in, and three-dimensional imaging algorithm fusion intensity image and Range Profile are utilized, three-dimensional reconstruction is carried out to purging zone, is obtained
Blind area type and orientation present in purging zone；
Cleaning agency, including round brush motor, round brush, micro-cleaner and scalable Bian Shua mechanism, scalable Bian Shua mechanism
Including sub-shell, while brush, while brush motor and connecting rod pushing meanss for realizing the scalable cleaning of side brush eliminate and clean blind area.
A kind of cleaning control method of the non-blind area sweeping robot based on deep learning algorithm, including：
Before starting cleaning works, laser radar carries out horizontal 360-degree spacescan to purging zone, and scan data is passed
It is defeated to arrive controller；The scan data includes the intensity data of point cloud data and echo-signal；
Scan data is handled and calculated by Context awareness module, synthesizing has dead zone information and spatial information
3-D image is formulated according to blind area and orientation is cleaned and cleans path；
Cleaning works is carried out according to the cleaning path of setting, and sweeping robot is carried out in fact by reckoning sensing unit
Shi Dingwei；When anticollision sensing unit detects barrier or prevents that falling sensing unit detects ground level drop, controller control
Robot processed is retreated or is turned to；
When being swept into corner or unapproachable hanging bottom, sweeping robot stops walking, is controlled and is rotated by controller
Motor work, torque by retarder drive two link mechanism slow rotations, push sub-shell forward extend out, side brush to blind area into
Row cleans；
It when cleaning edge, works along side sensing unit, photoelectric sensor is by calculating sweeping robot and barrier edge
Distance, clean the abundant welt of sweeping robot, remove edge blind area.
Compared with prior art, one or more embodiments of the invention can have following advantage：
1, it using the algorithm of target detection based on deep learning, can automatically detect sweeping robot in purging zone
Blind area, and scan data Processing Algorithm and three-dimensional imaging algorithm are combined, sweeping robot can be known before cleaning works starts
Blind area type and its orientation existing for purging zone.
2, using extension type Bian Shua mechanism, when sweeping robot is swept into corner or unapproachable hanging bottom,
Controller controls rotary electric machine rotation, and torque is acted on by the deceleration increment of retarder, and drive link mechanism slow rotation pushes
Sub-shell forward extends out, and the side brush of sub-shell bottom can clean blind area, eliminates and cleans blind area.
3, calculate along side using photoelectrical position sensor, when sweeping robot is cleaned along wall side, photoelectric sensor
By calculating sweeping robot at a distance from edge, judges whether robot can execute cleaning along side for closer distance, make machine
People can sufficiently welt clean, and eliminate and clean blind area.
Detailed description of the invention
Fig. 1 is the non-blind area sweeping robot internal structure chart based on deep learning algorithm；
Fig. 2 is intensity data of the non-blind area sweeping robot based on deep learning algorithm to point cloud data and echo-signal
Processing Algorithm flow chart；
Fig. 3 is the non-blind area sweeping robot algorithm of target detection schematic diagram based on deep learning algorithm；
Fig. 4 is the non-blind area sweeping robot three-dimensional imaging algorithm principle figure based on deep learning algorithm；
Fig. 5 is the non-blind area sweeping robot controller principle schematic diagram based on deep learning algorithm；
Fig. 6 is the non-blind area sweeping robot walking mechanism working principle diagram based on deep learning algorithm；
Fig. 7 is the non-blind area sweeping robot anticollision sensing unit working principle diagram based on deep learning algorithm；
Fig. 8, which is that the non-blind area sweeping robot based on deep learning algorithm is anti-, falls sensing unit working principle diagram；
Fig. 9 is the non-blind area sweeping robot based on deep learning algorithm along side sensing unit working principle diagram；
Figure 10 is the non-blind area sweeping robot control cleaning method flow chart based on deep learning algorithm；
Figure 11 is the non-blind area sweeping robot main view based on deep learning algorithm；
Figure 12 is the non-blind area sweeping robot top view based on deep learning algorithm；
Figure 13 is the non-blind area sweeping robot bottom view based on deep learning algorithm；
Figure 14 is the non-blind area sweeping robot side view based on deep learning algorithm；
Figure 15 a, 15b, 15c and 15d are that the non-blind area sweeping robot based on deep learning algorithm eliminates corner blind area effect
Fruit figure.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with examples and drawings to this hair
It is bright to be described in further detail.
As shown in Figure 1, be the non-blind area sweeping robot internal structure based on deep learning algorithm, including：Body shell
1, laser radar 6, controller 2, power supply 3, walking mechanism, sensing mechanism and cleaning agency；It is described
Laser radar is arranged in the top of body shell, for carrying out horizontal 360-degree spacescan to purging zone, and
The point cloud data of the strength information of the echo-signal scanned and generation is transferred to controller；
Controller, including Context awareness module, the Context awareness module are used for strength information and point cloud to echo-signal
The data of data carry out the two-dimensional intensity picture and Range Profile that processing operation is cleaned region, are detected using algorithm of target detection strong
Blind area of the degree as in, and three-dimensional imaging algorithm fusion intensity image and Range Profile are utilized, three-dimensional reconstruction is carried out to purging zone, is obtained
Blind area type and orientation present in purging zone；
Cleaning agency, including round brush 16, round brush motor 17, micro-cleaner and scalable Bian Shua mechanism, scalable side brush
Mechanism include sub-shell 20, while brush 21, while brush motor 22 and connecting rod pushing meanss disappear for realizing the scalable cleaning of side brush
Except cleaning blind area.
As shown in Fig. 2, the data processing algorithm of scan data, that is, strength information and point cloud data in Context awareness module
Include the following steps：
Purging zone is calculated according to the x of point cloud data, the intensity value of y-axis coordinate value and echo-signal in step A1
Intensity image；
Step A2 carries out Wavelet Denoising Method processing to intensity image, obtains intensity image a, cleans blind area for detecting；
Step A3 carries out edge detection to intensity image a, obtains intensity image b, is used for three-dimensional imaging；
The Range Profile of purging zone is calculated according to the z-axis coordinate value of point cloud data in step A4.
The algorithm of target detection being illustrated in figure 3 in Context awareness module, includes the following steps：
Step B1：Offline acquisition training sample image.
The scan data of different purging zones is acquired offline (including echo-signal using the laser radar of sweeping robot
Strength information and point cloud data), and corresponding intensity image a is generated using the data processing algorithm；
Step B2：Image is marked offline.
Utilize target candidate frame and the matched sample set of target labels in marking software production intensity image a, the target of detection
It specifically includes edge, corner and 3 class of hanging bottom and cleans blind area；
Step B3：Off-line training target detection network.
Off-line training is carried out to target detection network using sample set, trained target detection network after training successfully
It is embedded into the Context awareness module of controller；
Step B4：Detection cleans blind area.Sweeping robot is before starting cleaning, the first intensity that data processing algorithm is obtained
As a is input to trained target detection network, blind area type existing for purging zone is detected.
As shown in figure 4, the three-dimensional imaging algorithm in Context awareness module carries out model construction using space lattice partitioning,
On the basis of the body outline of intensity image b, intensity image a and Range Profile through target detection are merged, synthesis has dead zone information
With the 3-D image of spatial information.
As shown in figure 5, controller further include signal analysis module, control walking module and control clean module, and with electricity
Source 3, laser radar 6, walking mechanism, sensing mechanism are connected with cleaning agency, and 2 fixed and arranged of controller is at 1 rear of body shell.
The outside portion of above-mentioned body shell 1 is equipped with rubber ring 4 of the circle for crusherbull zone protection；The body shell
A plenum chamber 5 is equipped among 1 bottom.
Walking mechanism includes driving wheel, universal wheel 15 and driving motor, before executing sweeping robot on the ground
Into, retreat, go to action.Left driving wheel 11 and right driving wheel 12 are arranged symmetrically in the left and right sides of 1 bottom of body shell, and divide
Not with the output axis connection of left driving motor 13 and right driving motor 14, universal wheel 15 is arranged in the front end of 1 bottom of body shell,
And be located on the axis of symmetry of two driving wheels, controller 2 is connect with driving motor；Controller 2 is produced by two driving motors of control
Raw different revolving speed and steering, make sweeping robot realize different walking motions, and control walking principle is as shown in Figure 6.
Sensing mechanism includes reckoning sensing unit 10, anticollision sensing unit, dropproof sensing unit and senses along side
Information for obtaining the information of sweeping robot ambient enviroment, then is transferred to controller 2 as electronic signals and divided by unit
Analysis, controller 2 issue other mechanisms according to present circumstances and instruct.The reckoning sensing unit 10 includes being used for
The gyroscope of angular speed and the acceleration transducer for measuring acceleration are measured, package arrangement is before the bottom of body shell 1
End, for calculating the travel track of sweeping robot, positions robot in real time.
Above-mentioned anticollision sensing unit includes the infrared sensor 7 of nine front side portions for being evenly arranged in shell, phase
Angle between adjacent two sensors is 15 degree, and anticollision sensing unit can be to 120 degree in front of sweeping robot within sweep of the eye
Barrier detected.Transducing signal analysis module sets two distance threshold L1 and L2 (L1>L2), work as infrared sensor
7 detect front obstacle apart from when being less than L1, and controller 2 is issued to driving motor and instructed, before control sweeping robot deceleration
Into；When infrared sensor 7 detects that front obstacle distance is less than L2, controller 2 is issued to driving motor and is instructed, control
Robot is moved to the left or right, avoiding obstacles, and the workflow of anticollision sensing unit is as shown in Figure 7.
The dropproof sensing unit include three be evenly arranged in it is infrared in the same horizontal line of housing bottom front end
Line sensor 8；The working principle of dropproof sensing unit is：Infrared sensor 8 is by a transmitter and a receiver group
At transmitter emits infrared-ray to the ground at regular intervals, and transducing signal analysis module sets a time threshold T1, such as
The time that fruit receiver receives infrared-ray is greater than T1, then judges in front of sweeping robot for the biggish ground of height fall.When
When dropproof sensing unit detects that front is the height falls such as step big ground, controller 2 is issued to driving motor and is instructed,
Control sweeping robot first moves backward, and prevents sweeping robot from falling, the workflow of dropproof sensing unit such as Fig. 8 institute
Show.
Above-mentioned along side sensing unit includes two photoelectric sensors for being arranged symmetrically in the left and right side middle of shell
9, photoelectric sensor 9 can carry out ranging to the barrier of closer distance, and signal analysis module sets a distance threshold L3 (L3<
L2).When sweeping robot is cleaned along side, photoelectric sensor 9 understands the distance for calculating sweeping robot in real time to barrier edge,
If distance is greater than L3, controller 2 can calculate the safe distance that sweeping robot can be promoted to edge, control machine of sweeping the floor
People is close to edge, sufficiently cleans along side, eliminates edge blind area, and the workflow along side sensing unit is as shown in Figure 9.
Round brush 16 is arranged in the plenum chamber of housing bottom, and the output axis connection with round brush motor 17；Micro-cleaner
Including the dust-collecting box 18 being arranged in above plenum chamber and the true suction pump 19 for being arranged in 18 rear of dust-collecting box；Work as sweeping robot
When work, round brush motor 17 drives the rotation of round brush 16 to raise the dust on ground and sundries, while vacuum suction and pumping device 19 passes through sky
Chamber 5 is in dust debris inspiration dust-collecting box 18.
Side brush motor 22 includes two, is arranged in inside sub-shell 20, and side brush 21 is arranged in 20 bottom of sub-shell, and and side
Brush motor 22 exports axis connection；The fan-shaped body of the shape of sub-shell 20, the height of sub-shell 20 are less than the height of body shell 1,
And it is separable with body shell 1, it is cleaned for side brush 21 to be put in the unapproachable blind area of sweeping robot.Sub-shell 20
There are two settings, is arranged symmetrically in the left and right sides in 1 front of body shell, and body shell 1 and sub-shell 20 are equipped with mating
25 guide rail of guide wheel, 26 device used only moves in the front-back direction for limiting when sub-shell 20 moves.The guide wheel 25 wraps
Six are included, guide rail 26 includes two.
Above-mentioned connecting rod pushing meanss include two link mechanisms, rotary electric machine 27 and retarder 28；Two link mechanisms
Including master connecting-rod 23 and slave connecting rod 24, for pushing sub-shell 20 to make, front-rear direction is mobile, and the fulcrum of master connecting-rod 23 is hinged on machine
1 bottom of body shell, the fulcrum of slave connecting rod 24 are hinged on 20 bottom of sub-shell；Master connecting-rod 23 is driven by rotary electric machine 27, for reality
Existing link mechanism can start smooth and slow rotation, and retarder 28 is equipped between rotary electric machine 27 and master connecting-rod 23 and plays deceleration increasing
Away from effect.
The maximum distance that above-mentioned sub-shell 20 forward extends out depends on length, initial position and the angle of rotation of master connecting-rod 23
Degree considers the height of body shell 1, the master connecting-rod of this example since the link mechanism rotational plane of this example is perpendicular to ground
23 length are 3 centimetres；Consider the dead-centre position of link mechanism, in order to avoid dead point, the master connecting-rod 23 of this example sub-shell not
The initial angle and bottom shell (direction retreated with sweeping robot) forward extended out is in 30 degree of angles；Consider sub-shell 20 to
The maximum distance of preceding stretching, the 23 maximum rotation angle of master connecting-rod of this example are 150 degree；It can be calculated by geometric formula, this reality
The maximum distance that the sub-shell 20 of example forward extends out is 5.1 centimetres.When sweeping robot is cleaned to corner or bottom flying height
When low furniture edge, controller 2 first controls driving motor and stops operating, and sweeping robot is made to halt, and then control turns
Dynamic motor 27 rotates, and torque acts on drive link mechanism slow rotation by the deceleration increment of retarder 28, pushes sub-shell 20
It forward extends out, the side of 20 bottom of sub-shell is brushed 21 pairs of blind areas and cleaned.
The cleaning control method of non-blind area sweeping robot based on deep learning algorithm shown in Figure 10 includes as follows：
Before sweeping robot starts cleaning works, laser radar 6 carries out horizontal 360-degree spacescan to purging zone, and
Scan data is transferred to controller；The scan data includes the intensity data of point cloud data and echo-signal.
Context awareness module is handled and is calculated to scan data, and the three-dimensional with dead zone information and spatial information is synthesized
Image, sweeping robot has been detected by cleaning blind area and its orientation present in purging zone at this time.
Sweeping robot starts cleaning works according to the cleaning path of setting, and reckoning sensing unit, anticollision sensing are single
Member and the anti-sensing unit that falls start work.Reckoning sensing unit positions sweeping robot in real time；When anticollision senses
When unit detects that there is barrier in front, controller controls sweeping robot cut-through object；Fall sensing unit when anti-and detect
When front ground level drop is big, controller controls sweeping robot and retreats or turn to.
When sweeping robot is swept into corner or unapproachable hanging bottom, sweeping robot stops walking, control
Device controls rotary electric machine work, and torque drives two link mechanism slow rotations by retarder, pushes sub-shell to forward extend out, side
Brush cleans blind area；
When sweeping robot is cleaned along edge, work along side sensing unit, photoelectrical position sensor is swept the floor by calculating
Robot cleans the abundant welt of sweeping robot, removes edge blind area at a distance from barrier edge.
Figure 11-Figure 14 is respectively non-blind area sweeping robot main view, top view, bottom view based on deep learning algorithm
And side view.
Sweeping robot eliminates the corner blind area that can not clean effect such as Figure 15 a of corner blind area, 15b is cleaned in X direction
Shown in the corner blind area of corner blind area, 15c behind Y-direction cleaning corner blind area and 15d cleaning corner, the radius of sweeping robot
Having a size of 175mm；The length of side brush 21 is 77mm；Side brush 21 be mounted on sweeping robot front end distance be 77mm, with sweep
The position that floor-washing robot most left (right side) end distance is 67mm；Circle of dotted line is the cleaning range of side brush in figure；Due to the infrared biography of anticollision
The detection accuracy of sensor is not high, and sweeping robot executes avoidance movement when detecting has barrier at the 20mm of front；In figure
Dash area is that can not clean corner blind area existing for the sweeping robot in corner, and area is denoted as S1.Such as Figure 15 b) and 15c) institute
Show, sweeping robot can carry out the cleaning of X and Y both direction to the same corner, due to being mounted with along side sensor, sweeper
Device people can be cleaned with the safe distance of 10mm along edge.When being swept into corner, in situ, then sweeping robot stops
Side brush is forward extended out to clean corner blind area.Figure 15 d) in dash area be can clean existing for the sweeping robot in corner
Corner blind area, is denoted as S2.Area can be acquired by CAD mapping：S1=73.71cm2, S2=6.48cm2, sweeping robot is to corner
Cleaning rate be 91.21%, relative to entire purging zone, the corner area that can not clean is negligible.Therefore, of the invention
Non-blind area sweeping robot can thoroughly remove corner blind area.
Although disclosed herein embodiment it is as above, the content is only to facilitate understanding the present invention and adopting
Embodiment is not intended to limit the invention.Any those skilled in the art to which this invention pertains are not departing from this
Under the premise of the disclosed spirit and scope of invention, any modification and change can be made in the implementing form and in details,
But scope of patent protection of the invention, still should be subject to the scope of the claims as defined in the appended claims.
Claims (10)
1. a kind of non-blind area sweeping robot based on deep learning algorithm, which is characterized in that including body shell, laser thunder
It reaches, controller, power supply, walking mechanism, sensing mechanism and cleaning agency；It is described
Laser radar is arranged in the top of body shell, for carrying out horizontal 360-degree spacescan to purging zone, and will sweep
The strength information for the echo-signal retouched and the point cloud data of generation are transferred to controller；
Controller, including Context awareness module, the Context awareness module are used for strength information and point cloud data to echo-signal
Data carry out processing operation and be cleaned the two-dimensional intensity picture and Range Profile in region, utilize algorithm of target detection detection intensity picture
In blind area, and utilize three-dimensional imaging algorithm fusion intensity image and Range Profile, to purging zone carry out three-dimensional reconstruction, be cleaned
Blind area type and orientation present in region；
Cleaning agency, including round brush motor, round brush, micro-cleaner and scalable Bian Shua mechanism, scalable Bian Shua mechanism include
Sub-shell, while brush, while brush motor and connecting rod pushing meanss for realizing the scalable cleaning of side brush eliminate and clean blind area.
2. as described in claim 1 based on the non-blind area sweeping robot of deep learning algorithm, which is characterized in that the intensity
The data processing algorithm of information and point cloud data includes the following steps：
The intensity of purging zone is calculated according to the x of point cloud data, the intensity value of y-axis coordinate value and echo-signal in step A1
Picture；
Step A2 carries out Wavelet Denoising Method processing to intensity image, obtains intensity image a, cleans blind area for detecting；
Step A3 carries out edge detection to intensity image a, obtains intensity image b, is used for three-dimensional imaging；
The Range Profile of purging zone is calculated according to the z-axis coordinate value of point cloud data in step A4.
3. as described in claim 1 based on the non-blind area sweeping robot of deep learning algorithm, which is characterized in that the target
Detection algorithm includes the following steps：
Step B1 acquires the strength information and point cloud data of the echo-signal of different purging zones offline, and utilizes echo-signal
Strength information and Point Cloud Processing algorithm generate corresponding intensity image a；
Step B2 is detected and is marked target candidate frame and the matched sample set of target labels in intensity image a；
Step B3 carries out off-line training to target detection network using sample set, and trained target detection internet startup disk is arrived
The Context awareness module of controller；
Intensity image a is input to trained target detection network by step B4, detects blind area type existing for purging zone.
4. as described in claim 1 based on the non-blind area sweeping robot of deep learning algorithm, which is characterized in that the mesh of detection
Mark includes that edge, corner and hanging bottom three classes clean blind area.
5. as described in claim 1 based on the non-blind area sweeping robot of deep learning algorithm, which is characterized in that the control
Device further includes that signal analysis module, control walking module and control clean module；The signal analysis module, control walking module
Module is cleaned with control to connect with laser radar, power supply, walking mechanism, sensing mechanism and cleaning agency respectively.
6. as described in claim 1 based on the non-blind area sweeping robot of deep learning algorithm, which is characterized in that the walking
Mechanism includes driving wheel, driving motor and universal wheel, for executing advance, retrogressing, stopping and the turning function of robot.
7. as described in claim 1 based on the non-blind area sweeping robot of deep learning algorithm, which is characterized in that the sensing
Mechanism is used to obtain the environmental information of purging zone, including reckoning sensing unit, anticollision sensing unit, dropproof sensing list
Member and along side sensing unit；It is described
Reckoning sensing unit, package arrangement body shell bottom front, for calculating the traveling rail of sweeping robot
Mark positions robot in real time；Acceleration including the gyroscope for measuring angular speed and for measuring acceleration passes
Sensor；
Anticollision sensing unit, including nine anticollision infrared sensors, the anticollision infrared sensor are evenly arranged in fuselage
The front side of shell, and the angle between adjacent anticollision infrared sensor is 15 degree；
Dropproof sensing unit, including three dropproof infrared sensors, and it is evenly arranged in the bottom front of body shell
Position；
Along side sensing unit, including two photoelectrical position sensors, and it is arranged symmetrically in the left and right side interposition of body shell
It sets.
8. as described in claim 1 based on the non-blind area sweeping robot of deep learning algorithm, which is characterized in that the pair
There are two shell includes, and it is set as segment, two sub-shells are arranged symmetrically in the arranged on left and right sides of body shell front end；
The body shell and sub-shell are equipped with matching used guide rail and guide wheel, move forward and backward for limiting sub-shell.
9. as described in claim 1 based on the non-blind area sweeping robot of deep learning algorithm, which is characterized in that the connecting rod
Pushing meanss include two link mechanisms, retarder and rotary electric machine；It is described
Two link mechanisms include that fulcrum is hinged on the intracorporal master connecting-rod of body housing and fulcrum is hinged on the intracorporal slave connecting rod of subshell；
Inside body shell, reducer input shaft is connect retarder fixed and arranged with rotary electric machine, reducer output shaft and institute
The master connecting-rod connection stated.
10. a kind of cleaning control method of the non-blind area sweeping robot based on deep learning algorithm, which is characterized in that the side
Method includes：
Before starting cleaning works, laser radar carries out horizontal 360-degree spacescan to purging zone, and scan data is transferred to
Controller；The scan data includes the intensity data of point cloud data and echo-signal；
Scan data is handled and calculated by Context awareness module, synthesizes the three-dimensional with dead zone information and spatial information
Image is formulated according to blind area and orientation is cleaned and cleans path；
Cleaning works is carried out according to the cleaning path of setting, and sweeping robot determine in real time by reckoning sensing unit
Position；When anticollision sensing unit detects barrier or prevents that falling sensing unit detects ground level drop, controller controls machine
Device people retreats or turns to；
When being swept into corner or unapproachable hanging bottom, sweeping robot stops walking, controls rotary electric machine by controller
Work, torque drive two link mechanism slow rotations by retarder, and sub-shell is pushed to forward extend out, and side brush carries out blind area clear
It sweeps；
When cleaning edge, work along side sensing unit, photoelectric sensor by calculate sweeping robot and barrier edge away from
From cleaning the abundant welt of sweeping robot, remove edge blind area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811071935.3A CN108852184B (en) | 2018-09-14 | 2018-09-14 | Non-blind area sweeping robot based on deep learning algorithm and sweeping control method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811071935.3A CN108852184B (en) | 2018-09-14 | 2018-09-14 | Non-blind area sweeping robot based on deep learning algorithm and sweeping control method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108852184A true CN108852184A (en) | 2018-11-23 |
| CN108852184B CN108852184B (en) | 2023-12-26 |
Family
ID=64324183
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811071935.3A Active CN108852184B (en) | 2018-09-14 | 2018-09-14 | Non-blind area sweeping robot based on deep learning algorithm and sweeping control method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108852184B (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109605390A (en) * | 2018-12-28 | 2019-04-12 | 芜湖哈特机器人产业技术研究院有限公司 | A kind of automobile washing machine people's control system |
| CN109602344A (en) * | 2019-01-10 | 2019-04-12 | 珠海格力电器股份有限公司 | Sweeping robot control method and device, system based on microwave radar |
| CN110503040A (en) * | 2019-08-23 | 2019-11-26 | 斯坦德机器人（深圳）有限公司 | Obstacle detection method and device |
| CN110772164A (en) * | 2019-10-21 | 2020-02-11 | 马志强 | Glass cleaning device capable of cleaning corners of glass window |
| CN111513630A (en) * | 2020-04-27 | 2020-08-11 | 小狗电器互联网科技（北京）股份有限公司 | Edgewise sweeping method of sweeping robot |
| CN111789537A (en) * | 2020-06-23 | 2020-10-20 | 深圳市无限动力发展有限公司 | Method, device and computer device for sweeping a floor area of a furniture bottom |
| CN112180930A (en) * | 2020-09-30 | 2021-01-05 | 小狗电器互联网科技（北京）股份有限公司 | Sweeping robot and method and device for determining sweeping path planning area thereof |
| CN112426104A (en) * | 2020-11-19 | 2021-03-02 | 温州复弘机械设备有限公司 | Intelligent networking sweeping robot auxiliary dust removal mechanism |
| CN112471974A (en) * | 2020-11-25 | 2021-03-12 | 新化县涵壹科技开发有限公司 | Building waste collection device |
| CN112690709A (en) * | 2020-12-25 | 2021-04-23 | 苏州阿甘机器人有限公司 | Intelligent sweeping robot without dead angle and working method thereof |
| CN112716362A (en) * | 2020-12-25 | 2021-04-30 | 苏州阿甘机器人有限公司 | Intelligent floor sweeping robot and working method thereof |
| WO2021120997A1 (en) * | 2019-12-20 | 2021-06-24 | 深圳市杉川机器人有限公司 | Method for controlling autonomous robot, and autonomous robot |
| CN113017518A (en) * | 2021-03-09 | 2021-06-25 | 李侃 | Cleaning control method and device for sweeping and mopping integrated robot |
| CN113396881A (en) * | 2021-07-08 | 2021-09-17 | 邓华 | Self-walking sterilization and mite removal cleaning device and safe escape method |
| CN114510015A (en) * | 2020-10-29 | 2022-05-17 | 深圳市普森斯科技有限公司 | Sweeping robot moving method, electronic device and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011045694A (en) * | 2009-07-31 | 2011-03-10 | Fuji Heavy Ind Ltd | Self-propelled cleaning robot equipped with side brush device |
| US20160206165A1 (en) * | 2013-09-12 | 2016-07-21 | Lg Electronics Inc. | Automatic cleaner |
| WO2017024479A1 (en) * | 2015-08-10 | 2017-02-16 | 胡丹丽 | Floor sweeping robot having customised sweeping area, and control method therefor |
| CN107092254A (en) * | 2017-04-27 | 2017-08-25 | 北京航空航天大学 | A kind of design method for the Household floor-sweeping machine device people for strengthening study based on depth |
| CN107357297A (en) * | 2017-08-21 | 2017-11-17 | 深圳市镭神智能系统有限公司 | A kind of sweeping robot navigation system and its air navigation aid |
| JP2017228195A (en) * | 2016-06-24 | 2017-12-28 | 大成建設株式会社 | Cleaning robot |
| CN107788916A (en) * | 2017-11-08 | 2018-03-13 | 安嘉琦 | Smart home cleans all-in-one |
| CN108089586A (en) * | 2018-01-30 | 2018-05-29 | 北醒（北京）光子科技有限公司 | A kind of robot autonomous guider, method and robot |
| CN209564068U (en) * | 2018-09-14 | 2019-11-01 | 李子璐 | A kind of non-blind area sweeping robot |
-
2018
- 2018-09-14 CN CN201811071935.3A patent/CN108852184B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011045694A (en) * | 2009-07-31 | 2011-03-10 | Fuji Heavy Ind Ltd | Self-propelled cleaning robot equipped with side brush device |
| US20160206165A1 (en) * | 2013-09-12 | 2016-07-21 | Lg Electronics Inc. | Automatic cleaner |
| WO2017024479A1 (en) * | 2015-08-10 | 2017-02-16 | 胡丹丽 | Floor sweeping robot having customised sweeping area, and control method therefor |
| JP2017228195A (en) * | 2016-06-24 | 2017-12-28 | 大成建設株式会社 | Cleaning robot |
| CN107092254A (en) * | 2017-04-27 | 2017-08-25 | 北京航空航天大学 | A kind of design method for the Household floor-sweeping machine device people for strengthening study based on depth |
| CN107357297A (en) * | 2017-08-21 | 2017-11-17 | 深圳市镭神智能系统有限公司 | A kind of sweeping robot navigation system and its air navigation aid |
| CN107788916A (en) * | 2017-11-08 | 2018-03-13 | 安嘉琦 | Smart home cleans all-in-one |
| CN108089586A (en) * | 2018-01-30 | 2018-05-29 | 北醒（北京）光子科技有限公司 | A kind of robot autonomous guider, method and robot |
| CN209564068U (en) * | 2018-09-14 | 2019-11-01 | 李子璐 | A kind of non-blind area sweeping robot |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109605390A (en) * | 2018-12-28 | 2019-04-12 | 芜湖哈特机器人产业技术研究院有限公司 | A kind of automobile washing machine people's control system |
| CN109602344A (en) * | 2019-01-10 | 2019-04-12 | 珠海格力电器股份有限公司 | Sweeping robot control method and device, system based on microwave radar |
| CN110503040A (en) * | 2019-08-23 | 2019-11-26 | 斯坦德机器人（深圳）有限公司 | Obstacle detection method and device |
| CN110772164B (en) * | 2019-10-21 | 2021-03-19 | 绍兴市晟途环保科技有限公司 | Glass cleaning device capable of cleaning corners of glass window |
| CN110772164A (en) * | 2019-10-21 | 2020-02-11 | 马志强 | Glass cleaning device capable of cleaning corners of glass window |
| WO2021120997A1 (en) * | 2019-12-20 | 2021-06-24 | 深圳市杉川机器人有限公司 | Method for controlling autonomous robot, and autonomous robot |
| CN111513630A (en) * | 2020-04-27 | 2020-08-11 | 小狗电器互联网科技（北京）股份有限公司 | Edgewise sweeping method of sweeping robot |
| CN111513630B (en) * | 2020-04-27 | 2021-10-08 | 小狗电器互联网科技（北京）股份有限公司 | Edgewise sweeping method of sweeping robot |
| CN111789537A (en) * | 2020-06-23 | 2020-10-20 | 深圳市无限动力发展有限公司 | Method, device and computer device for sweeping a floor area of a furniture bottom |
| CN112180930A (en) * | 2020-09-30 | 2021-01-05 | 小狗电器互联网科技（北京）股份有限公司 | Sweeping robot and method and device for determining sweeping path planning area thereof |
| CN114510015A (en) * | 2020-10-29 | 2022-05-17 | 深圳市普森斯科技有限公司 | Sweeping robot moving method, electronic device and storage medium |
| CN112426104A (en) * | 2020-11-19 | 2021-03-02 | 温州复弘机械设备有限公司 | Intelligent networking sweeping robot auxiliary dust removal mechanism |
| CN112471974A (en) * | 2020-11-25 | 2021-03-12 | 新化县涵壹科技开发有限公司 | Building waste collection device |
| CN112690709A (en) * | 2020-12-25 | 2021-04-23 | 苏州阿甘机器人有限公司 | Intelligent sweeping robot without dead angle and working method thereof |
| CN112716362A (en) * | 2020-12-25 | 2021-04-30 | 苏州阿甘机器人有限公司 | Intelligent floor sweeping robot and working method thereof |
| CN113017518A (en) * | 2021-03-09 | 2021-06-25 | 李侃 | Cleaning control method and device for sweeping and mopping integrated robot |
| CN113396881A (en) * | 2021-07-08 | 2021-09-17 | 邓华 | Self-walking sterilization and mite removal cleaning device and safe escape method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108852184B (en) | 2023-12-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108852184A (en) | A kind of non-blind area sweeping robot and its cleaning control method based on deep learning algorithm | |
| CN209564068U (en) | A kind of non-blind area sweeping robot | |
| US11712142B2 (en) | System of robotic cleaning devices | |
| CN111035327B (en) | Cleaning robot, carpet detection method, and computer-readable storage medium | |
| EP3230814B1 (en) | Using laser sensor for floor type detection | |
| Hasan et al. | Path planning algorithm development for autonomous vacuum cleaner robots | |
| US10678251B2 (en) | Cleaning method for a robotic cleaning device | |
| CN111166247B (en) | Garbage classification processing method and cleaning robot | |
| KR100922506B1 (en) | Autonomous canister vacuum cleaner, system thereof and method of vacuum cleaning using the same | |
| US11537135B2 (en) | Moving robot and controlling method for the moving robot | |
| EP3132732A1 (en) | Autonomous coverage robot | |
| CN109213137A (en) | sweeping robot, sweeping robot system and its working method | |
| CN110325938B (en) | Electric vacuum cleaner | |
| JP2017502371A (en) | Prioritizing cleaning areas | |
| JP7243967B2 (en) | Method for Detecting Level Differences on a Surface in Front of a Robotic Cleaning Device | |
| KR102070066B1 (en) | Robot cleaner and method for controlling the same | |
| WO2016005011A1 (en) | Method in a robotic cleaning device for facilitating detection of objects from captured images | |
| CN211933898U (en) | Cleaning robot | |
| CN113017492A (en) | Object recognition intelligent control system based on cleaning robot | |
| CN111225592B (en) | Autonomous traveling dust collector and extended area identification method | |
| CN110647152A (en) | Intelligent sweeping robot and control method | |
| KR20210146141A (en) | Moving robot and method for driving in corner areas thereof | |
| CN219609490U (en) | Self-moving equipment | |
| CN211375427U (en) | Floor sweeping robot | |
| TW202248674A (en) | Self-moving device and control method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |