CN108958263A - A kind of Obstacle Avoidance and robot - Google Patents
A kind of Obstacle Avoidance and robot Download PDFInfo
- Publication number
- CN108958263A CN108958263A CN201810878544.6A CN201810878544A CN108958263A CN 108958263 A CN108958263 A CN 108958263A CN 201810878544 A CN201810878544 A CN 201810878544A CN 108958263 A CN108958263 A CN 108958263A
- Authority
- CN
- China
- Prior art keywords
- avoidance
- barrier
- robot
- human body
- obstacle
- 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.)
- Pending
Links
- 230000004888 barrier function Effects 0.000 claims abstract description 304
- 230000003068 static effect Effects 0.000 claims abstract description 50
- 208000031872 Body Remains Diseases 0.000 claims abstract description 22
- 238000001514 detection method Methods 0.000 claims description 18
- 238000000034 method Methods 0.000 abstract description 13
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 208000010877 cognitive disease Diseases 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 238000001802 infusion Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0238—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using obstacle or wall sensors
- G05D1/024—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using obstacle or wall sensors in combination with a laser
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0242—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using non-visible light signals, e.g. IR or UV signals
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0255—Control of position or course in two dimensions specially adapted to land vehicles using acoustic signals, e.g. ultra-sonic singals
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Aviation & Aerospace Engineering (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Electromagnetism (AREA)
- Optics & Photonics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Acoustics & Sound (AREA)
- Manipulator (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The present invention provides a kind of Obstacle Avoidance and robot, when method includes: that robot has detected barrier, barrier is gone out according to collected frame picture recognition, and detect the motion conditions of the barrier；When identifying the barrier is that human body and the human body are kept in motion, the robot carries out avoidance according to the first avoidance distance；When identifying the barrier is that human body and the human body remain static, the robot carries out avoidance according to the second avoidance distance；When identifying the barrier is that object and the object are kept in motion, the robot carries out avoidance according to third avoidance distance；When identifying the barrier is that object and the object remain static, the robot carries out avoidance according to the 4th avoidance distance.Different avoidance distances can be configured when meeting different barriers by realizing, so that robot is more intelligent in avoidance.
Description
Technical field
The present invention relates to robot obstacle-avoiding field, espespecially a kind of Obstacle Avoidance and robot.
Background technique
With the continuous progress of science and technology, field in intelligent robotics is developed rapidly, and detection of obstacles and evacuation are them
The important embodiment of intelligent level.Good barrier avoiding function is the important leverage that mobile robot is walked safely.
The barrier avoiding function of service robot is mostly infrared sensing avoidance, ultrasonic radar avoidance at present, is set in avoidance
Fixed certain avoidance distance, when having detected barrier, position and avoidance distance of the server according to barrier are cooked up
The avoidance route of robot.
But in practical applications, if avoidance distance is arranged bigger, enter gateway or more crowded in robot
Channel when usually because avoidance lead to not walk apart from excessive；If avoidance distance is arranged smaller, in hospital, supermarket etc.
Under the bigger environment of mobility of people, and it is easy to touch with raw wipe of human hair.
For the avoidance distance more suitable for robot configuration, the present invention provides a kind of Obstacle Avoidance and machines
Device people.
Summary of the invention
The object of the present invention is to provide a kind of Obstacle Avoidance and robot, realizes and meeting different barriers
When can configure different avoidance distances so that robot is more intelligent in avoidance.
Technical solution provided by the invention is as follows:
The present invention provides a kind of Obstacle Avoidances, comprising steps of when robot has detected barrier, according to adopting
The frame picture recognition collected goes out barrier, and detects the motion conditions of the barrier；When identify the barrier be human body
And the human body, when being kept in motion, the robot carries out avoidance according to the first avoidance distance；When identifying the obstacle
For object for human body and when the human body remains static, the robot carries out avoidance according to the second avoidance distance；When identifying
For the barrier for object and when the object is kept in motion, the robot carries out avoidance according to third avoidance distance；
When identifying the barrier is that object and the object remain static, the robot according to the 4th avoidance distance into
Row avoidance.
Preferably, the object figure of robot acquisition human body image data, different objects is further comprised the steps of:
As data；The robot carries out depth image according to the human body image data, the object image data of different objects
It practises, allows the robot to identify human body and object.
Preferably, when robot identifies that the barrier is that human body and the human body are kept in motion, server
According to the human body in the position of present frame and the human body in the position of historical frames, the movement for predicting the human body becomes
Gesture track；Movement tendency track and first avoidance distance of the server according to the human body generate the first barrier
Domain；The server generates the first avoidance route according to first barrier zone and navigation map；The robot according to
The first avoidance route carries out avoidance；
Or；
When robot identifies that the barrier is that human body and the human body remain static, the server according to
The second avoidance distance and the human body generate the second barrier zone in the position of present frame；The server is according to institute
The second barrier zone and navigation map are stated, the second avoidance route is generated；The robot according to the second avoidance route into
Row avoidance；
Or；
When the robot identifies that the barrier is that object and the object are kept in motion, the server
According to the object in the position of present frame and the object in the position of historical frames, the movement for predicting the object becomes
Gesture track；Movement tendency track and third avoidance distance of the server according to the object generate third barrier
Domain；The server generates third avoidance route according to the third barrier zone and navigation map；The robot according to
The third avoidance route carries out avoidance；
Or；
When robot identifies that the barrier is that object and the object remain static, the server according to
The 4th avoidance distance and the object generate the 4th barrier zone in the position of present frame；The server is according to institute
The 4th barrier zone and navigation map are stated, the 4th avoidance route is generated；The robot according to the 4th avoidance route into
Row avoidance.
Preferably, when robot has detected barrier, barrier is identified by collected picture frame, this step
It specifically includes: when robot has detected barrier, obtaining the barrier in the position of present frame；Judge the barrier
In the position of present frame whether within the scope of default avoidance；If the barrier is in the position of present frame in default avoidance range
It is interior, then the motion conditions of barrier and the barrier are identified by collected picture frame.
Preferably, it further comprises the steps of: when identifying the barrier is that human body and the human body are kept in motion, institute
It states robot and avoidance is carried out with first movement speed；When identifying that the barrier is that human body and the human body remain static
When, the robot carries out avoidance with the second movement speed；When identifying that the barrier is that object and the object are in fortune
When dynamic state, the robot carries out avoidance with third movement speed；When identify the barrier be object and the object
When remaining static, the robot carries out avoidance with the 4th movement speed.
Preferably, it further comprises the steps of: when robot recognizes multiple barriers, the robot is according to the multiple barrier
The motion conditions for hindering object are kept away according to the corresponding first avoidance distance of the multiple barrier or the second avoidance distance or third
Maximum avoidance distance in barrier distance or the 4th avoidance distance carries out avoidance.
The present invention also provides a kind of robots, comprising: detection of obstacles module, for detecting barrier and obstacle
The situation of movement of object；Image capture module is electrically connected with the obstacle detection module, when for having detected barrier, acquisition
The present frame picture of the barrier；Obstacle recognition module is electrically connected with described image acquisition module, is collected for basis
Present frame picture recognition go out barrier；Obstacle avoidance module, respectively with the detection of obstacles module, the obstacle recognition module
Electrical connection, for when identifying the barrier is that human body and the human body are kept in motion, the robot is according to the
One avoidance distance carries out avoidance；When identifying the barrier is that human body and the human body remain static, the machine
People carries out avoidance according to the second avoidance distance；When identifying the barrier is that object and the object are kept in motion,
The robot carries out avoidance according to third avoidance distance；When identifying that the barrier is that object and the object are in static
When state, the robot carries out avoidance according to the 4th avoidance distance.
Preferably, described image acquisition module is also used to acquire the subject image number of human body image data, different objects
According to；The obstacle recognition module, is also electrically connected with described image acquisition module, is used for according to the human body image data, no
With the object image data of object, depth image study is carried out, allows the robot to identify human body and object.
Preferably, the obstacle avoidance module is also used to identify that the barrier is human body and the people when the robot
When body is kept in motion, navigate according to the first avoidance route that the server generates；The first avoidance route by
The server generates to obtain according to first barrier zone and navigation map；First obstacle-avoidance area is by the service
Device is generated and is obtained according to the movement tendency track of the human body and the first avoidance distance；The movement tendency track of the human body
It is obtained in the position of present frame and the human body in the position prediction of historical frames by the server according to the human body；
Or；
The obstacle avoidance module is also used to identify that the barrier is human body and the human body is in quiet when the robot
Only when state, avoidance is carried out according to the second avoidance route that the server generates；The second avoidance route is by the service
Device generates to obtain according to second barrier zone and navigation map；Second obstacle-avoidance area is by the server according to institute
The second avoidance distance and the human body is stated to generate to obtain in the position of present frame；
Or；
The obstacle avoidance module is also used to identify that the barrier is object and the object is in fortune when the robot
When dynamic state, navigate according to the third avoidance route that the server generates；The third avoidance route is by the service
Device generates to obtain according to the third barrier zone and navigation map；The third obstacle-avoidance area is by the server according to institute
The movement tendency track and third avoidance distance for stating object generate and obtain；The movement tendency track of the object is by the clothes
Business device is obtained in the position of present frame and the object in the position prediction of historical frames according to the object；
Or；
The obstacle avoidance module is also used to identify that the barrier is object and the object is in quiet when the robot
Only when state, avoidance is carried out according to the 4th avoidance route that the server generates；The 4th avoidance route is by the service
Device generates to obtain according to the 4th barrier zone and navigation map, and the 4th barrier zone is by the server according to institute
The 4th avoidance distance and the object is stated to generate to obtain in the position of present frame.
Preferably, position acquisition module, for when robot has detected barrier, obtaining the barrier current
The position of frame；Judgment module is electrically connected with the position acquisition module, for judging that the barrier is in the position of present frame
It is no to preset within the scope of avoidance；Obstacle recognition module, if being also used to the barrier in the position of present frame in default avoidance
In range, then barrier is identified by collected picture frame, and the obstacle is detected by the detection of obstacles module
The motion conditions of object.
Preferably, the obstacle avoidance module is also used to work as and identifies that the barrier is human body and the human body is in movement
When state, the robot carries out avoidance with first movement speed；The obstacle avoidance module is also used to work as and identifies the barrier
For human body and when the human body remains static, the robot carries out avoidance with the second movement speed；The obstacle avoidance module,
It is also used to when identifying the barrier is that object and the object are kept in motion, the robot is with the mobile speed of third
Degree carries out avoidance；The obstacle avoidance module is also used to work as and identifies that the barrier is object and the object remains static
When, the robot carries out avoidance with the 4th movement speed.
Preferably, the obstacle avoidance module is also used to when robot recognizes multiple barriers, and the robot is according to institute
The motion conditions for stating multiple barriers, according to the corresponding first avoidance distance of the multiple barrier or the second avoidance distance
Or the maximum avoidance distance in third avoidance distance or the 4th avoidance distance carries out avoidance.
A kind of Obstacle Avoidance and robot provided through the invention can be brought following at least one beneficial to effect
Fruit:
1, barrier-avoiding method provided by the invention can be machine according to the difference of barrier and the motion conditions of barrier
Corresponding avoidance distance is arranged in device people, so that the avoidance distance of robot is more in line with the characteristic of barrier, enhances robot
Avoidance intelligence degree.
2, by the deep learning to human body image data, object image data, robot can be made effectively to go out respectively
Barrier allows the robot to the avoidance distance setting for realizing differentiation.
3, robot can be according to barrier in the position of present frame and the position of historical frames, and active predicting goes out barrier
The availability of avoidance is improved so that the avoidance of robot has foresight in movement tendency track.
Detailed description of the invention
Below by clearly understandable mode, preferred embodiment is described with reference to the drawings, to a kind of Obstacle Avoidance
And above-mentioned characteristic, technical characteristic, advantage and its implementation of robot are further described.
Fig. 1 is a kind of flow chart of one embodiment of Obstacle Avoidance of the present invention；
Fig. 2 is a kind of flow chart of another embodiment of Obstacle Avoidance of the present invention；
Fig. 3 is a kind of flow chart of another embodiment of Obstacle Avoidance of the present invention；
Fig. 4 is a kind of structural schematic diagram of one embodiment of robot of the present invention.
Drawing reference numeral explanation:
11- detection of obstacles module, 12- image capture module, 13- obstacle recognition module, 14- obstacle avoidance module, 15-
It sets and obtains module, 16- judgment module, 2- server.
Specific embodiment
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, Detailed description of the invention will be compareed below
A specific embodiment of the invention.It should be evident that drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing, and obtain other embodiments.
To make simplified form, part related to the present invention is only schematically shown in each figure, they are not represented
Its practical structures as product.In addition, there is identical structure or function in some figures so that simplified form is easy to understand
Component only symbolically depicts one of those, or has only marked one of those.Herein, "one" is not only indicated
" only this ", can also indicate the situation of " more than one ".
The present invention provides a kind of one embodiment of Obstacle Avoidance, as shown in Figure 1, comprising:
When S1 robot has detected barrier, barrier is gone out according to collected frame picture recognition, and detect the barrier
Hinder the motion conditions of object；
For S21 when identifying the barrier is that human body and the human body are kept in motion, the robot is according to the
One avoidance distance carries out avoidance；
For S22 when identifying the barrier is that human body and the human body remain static, the robot is according to the
Two avoidances distance carries out avoidance；
For S23 when identifying the barrier is that object and the object are kept in motion, the robot is according to the
Three avoidances distance carries out avoidance；
For S24 when identifying the barrier is that object and the object remain static, the robot is according to the
Four avoidances distance carries out avoidance.
Specifically, barrier-avoiding method provided by the invention can be applied on service humanoid robot, the present embodiment is in hospital
Service robot for, the barrier in this scene of hospital is rather complicated, for instance that patient, trolley for hospital use, mobile disease
Bed, infusion support, dustbin, fixed hospital bed, seat, workbench etc., and the barrier such as patient, trolley for hospital use, Mobile sickbed
Also every now and then in movement.In the prior art, certain avoidance distance would generally be arranged in service humanoid robot at work,
So that robot is unlikely to encounter barrier during the motion.If but the avoidance of setting will lead to robot apart from excessive
By narrow door ways, or when flow of the people is more can not normal walking, reduce the practicability of robot.If but setting
For avoidance apart from too small, robot very likely collides barriers of pedestrian or other movements when walking, brings very big wind
Danger.Be arranged moderate avoidance apart from when, the defect of above two scheme can not be solved.Thus one kind is devised in the present invention
The barrier-avoiding method of robot:
Robot, can be or super by being mounted on the laser radar or infrared sensor of robot surrounding when mobile
Whether there is barrier on the obstacle detecting apparatus such as sound radar detection robot moving direction, if detected barrier,
Current picture can be acquired by the camera of robot front, and barrier is identified from picture with image recognition technology
Type, while the picture of different frame that acquires also according to camera of robot and laser radar or ultrasonic radar detect
The motion conditions of barrier.
The present embodiment be arranged avoidance apart from when can be from passage speed and current safety two from the aspect of.Hinder when identifying
When to hinder object be human body, needing more to consider the particularity of current safety, especially hospital, pedestrian is mostly patient, in order to
It prevents robot from colliding pedestrian when walking, therefore the weight of current safety can be increased, reduce the weight of passage speed,
The avoidance distance setting of robot is a little big.It, will when barrier is human body and human body remains static in the present embodiment
The avoidance distance of robot is set as the second avoidance distance, such as 0.4m~0.5m.And when human body is kept in motion, movement
Controllability it is poor, avoidance distance can be arranged bigger, robot is avoided to collide pedestrian, in the present embodiment, work as barrier
For human body and when the human body is kept in motion, the avoidance distance of robot is set as the first avoidance distance, such as 0.5m~
0.6m。
And when barrier is object, human body can be lower than in the weight of current safety, while in order to ensure robot
Passage speed, the avoidance distance of robot can be set below when barrier be human body when avoidance distance.The present embodiment
By barrier for object and when object remains static, the avoidance distance of robot is set as the 4th avoidance distance, such as
0.2m~0.3m.Controllability when being kept in motion in view of object is poor, can be that object and object are in fortune by barrier
When dynamic state, the avoidance distance of robot is set as the third avoidance distance greater than the 4th avoidance distance, for example, be set as 0.3m~
0.4m。
According to avoidance mode provided in this embodiment, so that robot takes into account barrier safety and current speed in avoidance
Degree, improves intelligence degree of the robot in avoidance.
As shown in Fig. 2, the present invention provides a kind of one embodiment of Obstacle Avoidance, comprising:
S01 robot acquires the object image data of human body image data, different objects；
Robot described in S02 carries out depth image according to the human body image data, the object image data of different objects
Study, allows the robot to identify human body and object.
When S1 robot has detected barrier, barrier is gone out according to collected frame picture recognition, and detect the barrier
Hinder the motion conditions of object；
S211 when robot the identifies barrier is human body and the human body is kept in motion, server according to
The human body, in the position of historical frames, predicts the movement tendency rail of the human body in the position of present frame and the human body
Mark；
Movement tendency track and first avoidance distance of the server described in S212 according to the human body generate the first barrier
Hinder region；
Server described in S213 generates the first avoidance route according to first barrier zone and navigation map；
Robot described in S214 carries out avoidance according to the first avoidance route；
S221 is when robot the identifies barrier is human body and the human body remains static, the server
According to the second avoidance distance and the human body in the position of present frame, the second barrier zone is generated；
Server described in S222 generates the second avoidance route according to second barrier zone and navigation map；
Robot described in S223 carries out avoidance according to the second avoidance route；
S231 is when the robot identifies that the barrier is that object and the object are kept in motion, the clothes
Be engaged in device according to the object in the position of present frame and the object in the position of historical frames, predict the fortune of the object
Dynamic trend track；
Movement tendency track and third avoidance distance of the server described in S232 according to the object generate third barrier
Hinder region；
Server described in S233 generates third avoidance route according to the third barrier zone and navigation map；
Robot described in S234 carries out avoidance according to the third avoidance route；
S241 is when robot the identifies barrier is object and the object remains static, the server
According to the 4th avoidance distance and the object in the position of present frame, the 4th barrier zone is generated；
Server described in S242 generates the 4th avoidance route according to the 4th barrier zone and navigation map；
Robot described in S243 carries out avoidance according to the 4th avoidance route.
Specifically, in order to enable there is the ability of cognitive disorders object to need in the process of implementation to robot for robot
Carry out image recognition training.Firstly, the object image data of the human body image data of big quantity, a variety of objects can be acquired, then
Deep learning is carried out to these data using by CNN neural network algorithm or faster R-CNN neural network algorithm, so that
Robot has the ability of cognitive disorders object.
Robot can acquire in real time the image data of each frame on track route when walking by camera,
When there is barrier from image data, then barrier is identified, while being measured with laser radar or ultrasonic radar
Distance of the robot to barrier out.
Above-mentioned collected data are uploaded to server by robot, and server is according to the current seat on map of robot
Cursor position and barrier are to the relative position of robot, so that the position to barrier positions.When human body is in movement
When state, server can pass through the position of human body in the image of adjacent multiframe, and the laser thunder in each frame, in robot
It reaches or distance relative to robot of human body that ultrasonic radar detects, simulates movement tendency track (such as the people of human body
If the straight line walked, then the movement tendency track of near linear can be simulated).Then in being with the movement tendency track of human body
The heart generates the first barrier zone using the first avoidance distance as diameter.Server is further according to the first obstacle-avoidance area and navigation map
On channel case (channel width, length), cook up the first avoidance route again for robot, and will be under the first avoidance route
It is sent to robot, robot is made to carry out avoidance according to the first avoidance route.When human body remains static, server can be with people
Centered on body current location, using the second avoidance distance as diameter, the second obstacle-avoidance area is generated, then further according to the second obstacle-avoidance area,
Channel case on navigation map, be the second avoidance of robot planning route, make robot according to the second avoidance route into
Row avoidance.
When detecting barrier is object, server can pass through the objects in images of the collected adjacent multiframe of camera
Position, and in each frame, the object that laser radar or ultrasonic radar in robot monitor is relative to robot
Distance, simulate the movement tendency track of object.Then centered on the movement tendency track of object, with third avoidance distance
For diameter, third barrier zone is generated.Server is machine further according to the channel case on third obstacle-avoidance area and navigation map
Device people cooks up third avoidance route again, and third avoidance route is issued to robot, makes robot according to third avoidance
Route carries out avoidance.
When detecting that object remains static, then directly centered on the current location of object, with the 4th avoidance away from
From for diameter, the 4th barrier zone is generated.Server is further according to the channel case on the 4th obstacle-avoidance area and navigation map
Robot cooks up the 4th avoidance route again, and third avoidance route is issued to robot, keeps away robot according to the 4th
Hinder route and carries out avoidance.
As shown in figure 3, the present invention provides a kind of one embodiment of Obstacle Avoidance, comprising:
S01 robot acquires the object image data of human body image data, different objects；
Robot described in S02 carries out depth image according to the human body image data, the object image data of different objects
Study, allows the robot to identify human body and object.
S11 obtains the barrier in the position of present frame when robot has detected barrier；
Whether S12 judges the barrier in the position of present frame within the scope of default avoidance；
If the S13 barrier within the scope of default avoidance, is known in the position of present frame by collected picture frame
Not Chu barrier and the barrier motion conditions；
If the S14 barrier not within the scope of default avoidance, does not identify barrier in the position of present frame；
S211 when robot the identifies barrier is human body and the human body is kept in motion, server according to
The human body, in the position of historical frames, predicts the movement tendency rail of the human body in the position of present frame and the human body
Mark；
Movement tendency track and first avoidance distance of the server described in S212 according to the human body generate the first barrier
Hinder region；
Server described in S213 generates the first avoidance route according to first barrier zone and navigation map；
Robot described in S214 carries out avoidance according to the first avoidance route；
S221 is when robot the identifies barrier is human body and the human body remains static, the server
According to the second avoidance distance and the human body in the position of present frame, the second barrier zone is generated；
Server described in S222 generates the second avoidance route according to second barrier zone and navigation map；
Robot described in S223 carries out avoidance according to the second avoidance route；
S231 is when the robot identifies that the barrier is that object and the object are kept in motion, the clothes
Be engaged in device according to the object in the position of present frame and the object in the position of historical frames, predict the fortune of the object
Dynamic trend track；
Movement tendency track and third avoidance distance of the server described in S232 according to the object generate third barrier
Hinder region；
Server described in S233 generates third avoidance route according to the third barrier zone and navigation map；
Robot described in S234 carries out avoidance according to the third avoidance route；
S241 is when robot the identifies barrier is object and the object remains static, the server
According to the 4th avoidance distance and the object in the position of present frame, the 4th barrier zone is generated；
Server described in S242 generates the 4th avoidance route according to the 4th barrier zone and navigation map；
Robot described in S243 carries out avoidance according to the 4th avoidance route.
Specifically, laser can be passed through when robot has detected barrier in the collected image data of camera
Radar obtains the barrier in the position of present frame；It is contemplated that the problem of memory size, robot can't immediately by
Location information preserves, but judges whether the barrier (can be set as in default avoidance range in the position of present frame
In 5M).If within a preset range, then it represents that barrier has had reached the range for needing to carry out avoidance, otherwise, illustrates barrier
It also farther out from robot, can be without avoidance.If the barrier is being preset within the scope of avoidance in the position of present frame, machine
Barrier is then stored and is sent to server in the position of present frame by people, and identifies barrier by collected picture frame
Hinder object.
Server can refer to a upper embodiment in the method for the avoidance route of planning robot, and the present embodiment is not repeating.
When barrier is human body and human body remains static, in order to ensure that robot does not collide human body in the process of traveling, |
The travel speed of robot can also be adjusted to the second movement speed (such as 2km/h~3km/h)；And when human body is in movement
It,, can in order to avoid robot is not collide human body in the opposite movement of human body since the controllability of movement is poor when state
The travel speed of robot is set as first movement speed (such as 1.5km/h~2km/h) more lower than the second movement speed.And
When barrier is object, and object is in static, the controllability school of object is high at this time, in order to improve the travel speed of robot
And working efficiency, the travel speed of robot can be adjusted to the 4th movement speed (such as 6km/h~8km/h).And when barrier
Hindering object is object, and when object is kept in motion, robot and barrier collide in order to prevent, can be by the row of robot
It sails speed and is adjusted to third movement speed (such as 3km/h~5km/h).
In addition, the robot is respectively corresponded according to the multiple barrier when robot recognizes multiple barriers
The first avoidance distance or the second avoidance distance or third avoidance distance in maximum avoidance distance carry out avoidance.
Such as first avoidance distance be set as 0.5m, the second avoidance distance can be set as 0.4m, and third avoidance distance can be set as
0.3m, the 4th avoidance distance are set as 0.2m.In robot ambulation, multiclass barrier may be encountered simultaneously, such as encounter fortune simultaneously
Moving body, static human body and moving object, it is static when, in order to ensure safety, avoidance traveling can be carried out according to the first avoidance distance.
Identical, when robot recognizes multiple barriers, the robot is right respectively according to the multiple barrier
Minimum translating velocity in the first movement speed or the second movement speed or third movement speed answered or the 4th movement speed into
Row avoidance.
The present invention also provides a kind of one embodiment of robot, comprising:
Detection of obstacles module 11, for detecting the situation of movement of barrier and barrier
Image capture module 12 is electrically connected with the obstacle detection module 11, when for having detected barrier, acquires institute
State the present frame picture of barrier；
Obstacle recognition module 13 is electrically connected with described image acquisition module 12, for according to collected present frame figure
Piece identifies barrier；
Obstacle avoidance module 14 is electrically connected with the detection of obstacles module 11, the obstacle recognition module 13 respectively, is used for
When identifying the barrier is that human body and the human body are kept in motion, the robot according to the first avoidance distance into
Row avoidance；When identifying the barrier is that human body and the human body remain static, the robot is kept away according to second
Barrier distance carries out avoidance；When identifying the barrier is that object and the object are kept in motion, the robot is pressed
Avoidance is carried out according to third avoidance distance；It is described when identifying the barrier is that object and the object remain static
Robot carries out avoidance according to the 4th avoidance distance.
Specifically, barrier-avoiding method provided by the invention can be applied on service humanoid robot, the present embodiment is in hospital
Service robot for, the barrier in this scene of hospital is rather complicated, for instance that patient, trolley for hospital use, mobile disease
Bed, infusion support, dustbin, fixed hospital bed, seat, workbench etc., and the barrier such as patient, trolley for hospital use, Mobile sickbed
Also every now and then in movement.In the prior art, certain avoidance distance would generally be arranged in service humanoid robot at work,
So that robot is unlikely to encounter barrier during the motion.If but the avoidance of setting will lead to robot apart from excessive
By narrow door ways, or when flow of the people is more can not normal walking, reduce the practicability of robot.If but setting
For avoidance apart from too small, robot very likely collides barriers of pedestrian or other movements when walking, brings very big wind
Danger.Be arranged moderate avoidance apart from when, the defect of above two scheme can not be solved.Thus one kind is devised in the present invention
The barrier-avoiding method of robot:
Robot, can be or super by being mounted on the laser radar or infrared sensor of robot surrounding when mobile
Whether there is barrier on the obstacle detecting apparatus such as sound radar detection robot moving direction, if detected barrier,
Current picture can be acquired by the camera of robot front, and barrier is identified from picture with image recognition technology
Type, while the picture of different frame that acquires also according to camera of robot and laser radar or ultrasonic radar detect
The motion conditions of barrier.
The present embodiment be arranged avoidance apart from when can be from passage speed and current safety two from the aspect of.Hinder when identifying
When to hinder object be human body, needing more to consider the particularity of current safety, especially hospital, pedestrian is mostly patient, in order to
It prevents robot from colliding pedestrian when walking, therefore the weight of current safety can be increased, reduce the weight of passage speed,
The avoidance distance setting of robot is a little big.It, will when barrier is human body and human body remains static in the present embodiment
The avoidance distance of robot is set as the second avoidance distance, such as 0.4m~0.5m.And when human body is kept in motion, movement
Controllability it is poor, avoidance distance can be arranged bigger, therefore in the present embodiment, when barrier is at human body and the human body
When motion state, the avoidance distance of robot is set as the first avoidance distance, such as 0.5m~0.6m
And when barrier is object, human body can be lower than in the weight of current safety, while in order to ensure robot
Passage speed, the avoidance distance of robot can be set below when barrier be human body when avoidance distance.The present embodiment
By barrier for object and when object remains static, the avoidance distance of robot is set as the 4th avoidance distance, such as
0.2m~0.3m.Controllability when being kept in motion in view of object is poor, can be that object and object are in fortune by barrier
When dynamic state, the avoidance distance of robot is set as third avoidance distance, can be set as 0.3m~0.4m.
According to avoidance mode provided in this embodiment, so that robot takes into account barrier safety and current speed in avoidance
Degree, improves intelligence degree of the robot in avoidance.
The present invention provides a kind of another embodiments of robot, as shown in Figure 4, comprising:
Described image acquisition module 12 is also used to acquire the object image data of human body image data, different objects；
The obstacle recognition module 13, is also electrically connected with described image acquisition module 12, for according to the human figure
As data, the object image data of different objects, depth image study is carried out, allows the robot to identify human body and object.
Detection of obstacles module 11, for detecting the situation of movement of barrier and barrier；
Image capture module 12 is electrically connected with the obstacle detection module 11, when for having detected barrier, acquires institute
State the present frame picture of barrier；
Position acquisition module 15 is electrically connected with the detection of obstacles module, for having detected barrier when robot
When, the barrier is obtained in the position of present frame；
Judgment module 16 is electrically connected with the position acquisition module 15, for judging the barrier in the position of present frame
It sets whether within the scope of default avoidance；
Obstacle recognition module 13, if being also used to the barrier in the position of present frame within the scope of default avoidance,
Barrier is identified by collected picture frame.
The obstacle avoidance module 14 is also used to identify that the barrier is human body and the human body is in when the robot
When motion state, navigate according to the first avoidance route that the server generates；
The first avoidance route is generated to obtain by the server according to first barrier zone and navigation map；
First obstacle-avoidance area is generated by the server according to the movement tendency track and the first avoidance distance of the human body
It arrives；Position and the human body of the movement tendency track of the human body by the server according to the human body in present frame
It is obtained in the position prediction of historical frames；
Or；
The obstacle avoidance module 14 is also used to identify that the barrier is human body and the human body is in when the robot
When stationary state, avoidance is carried out according to the second avoidance route that the server generates；
The second avoidance route is generated to obtain by the server according to second barrier zone and navigation map；
Second obstacle-avoidance area is raw in the position of present frame according to the second avoidance distance and the human body by the server
At obtaining；
Or；
The obstacle avoidance module 14 is also used to identify that the barrier is object and the object is in when the robot
When motion state, navigate according to the third avoidance route that the server generates；
The third avoidance route is generated to obtain by the server according to the third barrier zone and navigation map；
The third obstacle-avoidance area is generated by the server according to the movement tendency track and third avoidance distance of the object
It arrives；Position and the object of the movement tendency track of the object by the server according to the object in present frame
It is obtained in the position prediction of historical frames；
Or；
The obstacle avoidance module 14 is also used to identify that the barrier is object and the object is in when the robot
When stationary state, avoidance is carried out according to the 4th avoidance route that the server generates；
The 4th avoidance route is generated to obtain by the server according to the 4th barrier zone and navigation map,
4th barrier zone is raw in the position of present frame according to the 4th avoidance distance and the object by the server
At obtaining.
The obstacle avoidance module 14 is also used to work as and identifies that the barrier is human body and the human body is kept in motion
When, the robot carries out avoidance with first movement speed；
The obstacle avoidance module 14 is also used to work as and identifies that the barrier is human body and the human body remains static
When, the robot carries out avoidance with the second movement speed；
The obstacle avoidance module 14 is also used to work as and identifies that the barrier is object and the object is kept in motion
When, the robot carries out avoidance with third movement speed；
The obstacle avoidance module 14 is also used to work as and identifies that the barrier is object and the object remains static
When, the robot carries out avoidance with the 4th movement speed.
The obstacle avoidance module 14 is also used to when robot recognizes multiple barriers, and the robot is according to described more
The motion conditions of a barrier, according to the multiple barrier corresponding first avoidance distance or the second avoidance distance or the
Maximum avoidance distance in three avoidances distance or the 4th avoidance distance carries out avoidance.
Specifically, in order to enable there is the ability of cognitive disorders object to need in the process of implementation to robot for robot
Carry out image recognition training.Firstly, the human body image data of big quantity, object image data can be acquired, then uses and pass through CNN
Neural network algorithm or faster R-CNN neural network algorithm carry out deep learning to these data, know so that robot has
The ability of other barrier.
Robot can acquire in real time the image data of each frame on track route when walking by camera,
When there is barrier from image data, then barrier is identified, while being measured with laser radar or ultrasonic radar
Distance of the robot to barrier out.
Above-mentioned collected data are uploaded to server by robot, and server is according to the current seat on map of robot
Cursor position and barrier are to the relative position of robot, so that the position to barrier positions.When human body is in movement
When state, server can pass through the position of human body in the image of adjacent multiframe, and the laser thunder in each frame, in robot
It reaches or distance relative to robot of human body that ultrasonic radar detects, simulates movement tendency track (such as the people of human body
If the straight line walked, then the movement tendency track of near linear can be simulated).Then in being with the movement tendency track of human body
The heart generates the first barrier zone using the first avoidance distance as diameter.Server is further according to the first obstacle-avoidance area and navigation map
On channel case (channel width, length), cook up the first avoidance route again for robot, and will be under the first avoidance route
It is sent to robot, robot is made to carry out avoidance according to the first avoidance route.When human body remains static, server can be with people
Centered on body current location, using the second avoidance distance as diameter, the second obstacle-avoidance area is generated, then further according to the second obstacle-avoidance area,
Channel case on navigation map, be the second avoidance of robot planning route, make robot according to the second avoidance route into
Row avoidance.
When detecting barrier is object, server can pass through the objects in images of the collected adjacent multiframe of camera
Position, and in each frame, the object that laser radar or ultrasonic radar in robot monitor is relative to robot
Distance, simulate the movement tendency track of object.Then centered on the movement tendency track of object, with third avoidance distance
For diameter, third barrier zone is generated.Server is machine further according to the channel case on third obstacle-avoidance area and navigation map
Device people cooks up third avoidance route again, and third avoidance route is issued to robot, makes robot according to third avoidance
Route carries out avoidance.
When detecting that object remains static, then directly centered on the current location of object, with the 4th avoidance away from
From for diameter, the 4th barrier zone is generated.Server is further according to the channel case on the 4th obstacle-avoidance area and navigation map
Robot cooks up the 4th avoidance route again, and third avoidance route is issued to robot, keeps away robot according to the 4th
Hinder route and carries out avoidance.
When robot has detected barrier in the collected image data of camera, can be obtained by laser radar
The barrier is in the position of present frame；It is contemplated that the problem of memory size, robot can't be immediately by location information
It preserves, but whether judge the barrier interior in default avoidance range (can be set as 5M) in the position of present frame.If
Within a preset range, then it represents that barrier has had reached the range for needing to carry out avoidance, otherwise, illustrates that barrier is also disembarked device
People farther out, can be without avoidance.If the barrier is in the position of present frame within the scope of default avoidance, robot will hinder
Hinder object to store and be sent to server in the position of present frame, and barrier is identified by collected picture frame.
Server can refer to a upper embodiment in the method for the avoidance route of planning robot, and the present embodiment is not repeating.
When barrier is human body and human body remains static, in order to ensure that robot does not collide human body in the process of traveling, |
The travel speed of robot can also be adjusted to the second movement speed (such as 2km/h~3km/h)；And when human body is in movement
It,, can in order to avoid robot is not collide human body in the opposite movement of human body since the controllability of movement is poor when state
The travel speed of robot is set as first movement speed (such as 1.5km/h~2km/h) more lower than the second movement speed.And
When barrier is object, and object is in static, the controllability school of object is high at this time, in order to improve the travel speed of robot
And working efficiency, the travel speed of robot can be adjusted to the 4th movement speed (such as 6km/h~8km/h).And when barrier
Hindering object is object, and when object is kept in motion, robot and barrier collide in order to prevent, can be by the row of robot
It sails speed and is adjusted to third movement speed (such as 3km/h~5km/h).
In addition, the robot is respectively corresponded according to the multiple barrier when robot recognizes multiple barriers
The first avoidance distance or the second avoidance distance or third avoidance distance in maximum avoidance distance carry out avoidance.
Such as first avoidance distance be set as 0.5m, the second avoidance distance can be set as 0.4m, and third avoidance distance can be set as
0.3m, the 4th avoidance distance are set as 0.2m.In robot ambulation, multiclass barrier may be encountered simultaneously, such as encounter fortune simultaneously
Moving body, static human body and moving object, it is static when, in order to ensure safety, avoidance traveling can be carried out according to the first avoidance distance.
Identical, when robot recognizes multiple barriers, the robot is right respectively according to the multiple barrier
Minimum translating velocity in the first movement speed or the second movement speed or third movement speed answered or the 4th movement speed into
Row avoidance.
It should be noted that above-described embodiment can be freely combined as needed.The above is only of the invention preferred
Embodiment, it is noted that for those skilled in the art, in the premise for not departing from the principle of the invention
Under, several improvements and modifications can also be made, these modifications and embellishments should also be considered as the scope of protection of the present invention.
Claims (12)
1. a kind of Obstacle Avoidance, which is characterized in that comprising steps of
When robot has detected barrier, barrier is gone out according to collected picture recognition, and detect the fortune of the barrier
Emotionally condition；
When identifying the barrier is that human body and the human body are kept in motion, the robot according to the first avoidance away from
From progress avoidance；When identifying the barrier is that human body and the human body remain static, the robot is according to the
Two avoidances distance carries out avoidance；
When identifying the barrier is that object and the object are kept in motion, the robot according to third avoidance away from
From progress avoidance；When identifying the barrier is that object and the object remain static, the robot is according to the
Four avoidances distance carries out avoidance.
2. a kind of Obstacle Avoidance according to claim 1, which is characterized in that further comprise the steps of:
Robot acquires the object image data of human body image data, different objects；
The robot carries out depth image study, makes according to the human body image data, the object image data of different objects
Robot can recognize that human body and object.
3. a kind of Obstacle Avoidance according to claim 1, it is characterised in that:
When robot identifies that the barrier is that human body and the human body are kept in motion, server is according to the human body
In the position of present frame and the human body in the position of historical frames, the movement tendency track of the human body is predicted；
Movement tendency track and first avoidance distance of the server according to the human body generate the first barrier zone；
The server generates the first avoidance route according to first barrier zone and navigation map；The robot is pressed
Avoidance is carried out according to the first avoidance route；
Or；
When robot identifies that the barrier is that human body and the human body remain static, the server is according to
Second avoidance distance and the human body generate the second barrier zone in the position of present frame；
The server generates the second avoidance route according to second barrier zone and navigation map；
The robot carries out avoidance according to the second avoidance route；
Or；
When the robot identifies that the barrier is that object and the object are kept in motion, the server according to
The object, in the position of historical frames, predicts the movement tendency rail of the object in the position of present frame and the object
Mark；
Movement tendency track and third avoidance distance of the server according to the object generate third barrier zone；
The server generates third avoidance route according to the third barrier zone and navigation map；
The robot carries out avoidance according to the third avoidance route；
Or；
When robot identifies that the barrier is that object and the object remain static, the server is according to
4th avoidance distance and the object generate the 4th barrier zone in the position of present frame；
The server generates the 4th avoidance route according to the 4th barrier zone and navigation map；
The robot carries out avoidance according to the 4th avoidance route.
4. a kind of Obstacle Avoidance according to claim 1, which is characterized in that when robot has detected barrier
When, barrier is identified by collected picture frame, this step specifically includes:
When robot has detected barrier, the barrier is obtained in the position of present frame；
Judge the barrier in the position of present frame whether within the scope of default avoidance；
If the barrier within the scope of default avoidance, identifies obstacle by collected picture frame in the position of present frame
Object, and detect the motion conditions of the barrier.
5. a kind of Obstacle Avoidance described in any one of -4 according to claim 1, which is characterized in that further comprise the steps of:
When identifying the barrier is that human body and the human body are kept in motion, the robot is with first movement speed
Carry out avoidance；
When identifying the barrier is that human body and the human body remain static, the robot is with the second movement speed
Carry out avoidance；
When identifying the barrier is that object and the object are kept in motion, the robot is with third movement speed
Carry out avoidance；
When identifying the barrier is that object and the object remain static, the robot is with the 4th movement speed
Carry out avoidance.
6. a kind of Obstacle Avoidance described in any one of -4 according to claim 1, which is characterized in that further comprise the steps of:
When robot recognizes multiple barriers, the robot is according to the motion conditions of the multiple barrier, according to institute
State the corresponding first avoidance distance of multiple barriers or the second avoidance distance or third avoidance distance or the 4th avoidance distance
In maximum avoidance distance carry out avoidance.
7. a kind of robot characterized by comprising
Detection of obstacles module, for detecting the situation of movement of barrier and barrier；
Image capture module is electrically connected with the obstacle detection module, when for having detected barrier, acquires the barrier
Present frame picture；
Obstacle recognition module is electrically connected with described image acquisition module, for being gone out according to collected present frame picture recognition
Barrier；
Obstacle avoidance module is electrically connected with the detection of obstacles module, the obstacle recognition module respectively, identifies institute for working as
Barrier is stated for human body and when the human body is kept in motion, the robot carries out avoidance according to the first avoidance distance；When
The barrier is identified for human body and when the human body remains static, the robot is carried out according to the second avoidance distance
Avoidance；When identifying the barrier is that object and the object are kept in motion, the robot is according to third avoidance
Distance carries out avoidance；When identifying the barrier is that object and the object remain static, the robot according to
4th avoidance distance carries out avoidance.
8. a kind of robot according to claim 7, it is characterised in that:
Described image acquisition module is also used to acquire the object image data of human body image data, different objects；
The obstacle recognition module, is also electrically connected with described image acquisition module, is used for according to the human body image data, no
With the object image data of object, depth image study is carried out, allows the robot to identify human body and object.
9. a kind of robot according to claim 7, it is characterised in that:
The obstacle avoidance module is also used to identify that the barrier is human body and the human body is in movement shape when the robot
When state, navigate according to the first avoidance route that the server generates；
The first avoidance route is generated to obtain by the server according to first barrier zone and navigation map；It is described
First obstacle-avoidance area is generated and is obtained according to the movement tendency track of the human body and the first avoidance distance by the server；
The movement tendency track of the human body is being gone through according to the human body in the position of present frame and the human body by the server
The position prediction of history frame obtains；
Or；
The obstacle avoidance module is also used to identify that the barrier is human body and the human body is in static shape when the robot
When state, avoidance is carried out according to the second avoidance route that the server generates；
The second avoidance route is generated to obtain by the server according to second barrier zone and navigation map；It is described
Second obstacle-avoidance area is generated according to the second avoidance distance and the human body in the position of present frame by the server
It arrives；
Or；
The obstacle avoidance module is also used to identify that the barrier is object and the object is in movement shape when the robot
When state, navigate according to the third avoidance route that the server generates；
The third avoidance route is generated to obtain by the server according to the third barrier zone and navigation map；It is described
Third obstacle-avoidance area is generated and is obtained according to the movement tendency track and third avoidance distance of the object by the server；
The movement tendency track of the object is being gone through according to the object in the position of present frame and the object by the server
The position prediction of history frame obtains；
Or；
The obstacle avoidance module is also used to identify that the barrier is object and the object is in static shape when the robot
When state, avoidance is carried out according to the 4th avoidance route that the server generates；
The 4th avoidance route is generated to obtain by the server according to the 4th barrier zone and navigation map, described
4th barrier zone is generated according to the 4th avoidance distance and the object in the position of present frame by the server
It arrives.
10. a kind of robot according to claim 7, it is characterised in that:
Position acquisition module, for when robot has detected barrier, obtaining the barrier in the position of present frame；
Judgment module is electrically connected with the position acquisition module, for judge the barrier the position of present frame whether
Within the scope of default avoidance；
Obstacle recognition module, if being also used to the barrier in the position of present frame within the scope of default avoidance, by adopting
The picture frame collected identifies barrier, and the motion conditions of the barrier are detected by the detection of obstacles module.
11. a kind of robot according to any one of claim 7-10, it is characterised in that:
The obstacle avoidance module is also used to when identifying the barrier is that human body and the human body are kept in motion, described
Robot carries out avoidance with first movement speed；
The obstacle avoidance module is also used to when identifying the barrier is that human body and the human body remain static, described
Robot carries out avoidance with the second movement speed；
The obstacle avoidance module is also used to when identifying the barrier is that object and the object are kept in motion, described
Robot carries out avoidance with third movement speed；
The obstacle avoidance module is also used to when identifying the barrier is that object and the object remain static, described
Robot carries out avoidance with the 4th movement speed.
12. a kind of robot according to any one of claim 7-10, it is characterised in that:
The obstacle avoidance module is also used to when robot recognizes multiple barriers, and the robot is according to the multiple obstacle
The motion conditions of object, according to the corresponding first avoidance distance of the multiple barrier or the second avoidance distance or third avoidance
Maximum avoidance distance in distance or the 4th avoidance distance carries out avoidance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810878544.6A CN108958263A (en) | 2018-08-03 | 2018-08-03 | A kind of Obstacle Avoidance and robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810878544.6A CN108958263A (en) | 2018-08-03 | 2018-08-03 | A kind of Obstacle Avoidance and robot |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108958263A true CN108958263A (en) | 2018-12-07 |
Family
ID=64467175
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810878544.6A Pending CN108958263A (en) | 2018-08-03 | 2018-08-03 | A kind of Obstacle Avoidance and robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108958263A (en) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109445442A (en) * | 2018-12-17 | 2019-03-08 | 中新智擎科技有限公司 | A kind of robot obstacle-avoiding control method, device, storage medium and robot |
| CN109523574A (en) * | 2018-12-27 | 2019-03-26 | 联想(北京)有限公司 | A kind of run trace prediction technique and electronic equipment |
| CN110175523A (en) * | 2019-04-26 | 2019-08-27 | 南京华捷艾米软件科技有限公司 | A kind of self-movement robot animal identification and hide method and its storage medium |
| CN110315538A (en) * | 2019-06-27 | 2019-10-11 | 深圳乐动机器人有限公司 | A kind of method, apparatus showing barrier on the electronic map and robot |
| CN110370284A (en) * | 2019-08-07 | 2019-10-25 | 北京凌天世纪控股股份有限公司 | A kind of automatic control system of explosion-proof autonomous crusing robot |
| CN110370273A (en) * | 2019-06-27 | 2019-10-25 | 华为技术有限公司 | A kind of Obstacle Avoidance, device and system |
| CN110658819A (en) * | 2019-09-30 | 2020-01-07 | 北京猎户星空科技有限公司 | Obstacle avoidance method and device, electronic equipment and storage medium |
| CN110653837A (en) * | 2019-10-31 | 2020-01-07 | 灵动科技（北京）有限公司 | Autonomous moving device and warehouse logistics system |
| CN111026115A (en) * | 2019-12-13 | 2020-04-17 | 华南智能机器人创新研究院 | Robot obstacle avoidance control method and device based on deep learning |
| CN111390909A (en) * | 2020-03-30 | 2020-07-10 | 上海应用技术大学 | Obstacle avoidance method and system for industrial robot |
| CN111427357A (en) * | 2020-04-14 | 2020-07-17 | 北京石头世纪科技股份有限公司 | Robot obstacle avoidance method and device and storage medium |
| WO2020151664A1 (en) * | 2019-01-25 | 2020-07-30 | 长城汽车股份有限公司 | Vehicle-mounted control device, field end positioning device, and vehicle |
| WO2020218464A1 (en) * | 2019-04-25 | 2020-10-29 | 株式会社クボタ | Harvester, obstacle determination program, recording medium on which obstacle determination program is recorded, obstacle determination method, agricultural work machine, control program, recording medium on which control program is recorded, and control method |
| JP2020178618A (en) * | 2019-04-25 | 2020-11-05 | 株式会社クボタ | Agricultural implement |
| CN112034856A (en) * | 2020-09-08 | 2020-12-04 | 深圳优地科技有限公司 | Robot traveling method, device, equipment and storage medium |
| CN112405532A (en) * | 2020-11-06 | 2021-02-26 | 深圳优地科技有限公司 | Movement control method and device and robot |
| CN112449106A (en) * | 2019-09-03 | 2021-03-05 | 东芝泰格有限公司 | Frame plate shooting device and information processing device |
| CN112975989A (en) * | 2021-04-21 | 2021-06-18 | 炬星科技（深圳）有限公司 | Robot safe operation method, equipment and storage medium |
| CN113219969A (en) * | 2021-04-22 | 2021-08-06 | 深圳拓邦股份有限公司 | Obstacle avoidance control method and device for floor washing robot and floor washing robot |
| CN113848871A (en) * | 2020-06-28 | 2021-12-28 | 北京极智嘉科技股份有限公司 | Method, device and system for controlling robot to avoid obstacle |
| CN114363826A (en) * | 2021-12-22 | 2022-04-15 | 长沙智能驾驶研究院有限公司 | Information prompting method, device, equipment and readable storage medium |
| CN114348022A (en) * | 2022-01-21 | 2022-04-15 | 北京三快在线科技有限公司 | Control method and device for unmanned vehicle, storage medium and electronic equipment |
| CN114415726A (en) * | 2022-01-18 | 2022-04-29 | 江苏锐天智能科技股份有限公司 | Unmanned aerial vehicle obstacle avoidance control system and method based on image analysis |
| WO2022127439A1 (en) * | 2020-12-17 | 2022-06-23 | 深圳市普渡科技有限公司 | Robot obstacle avoidance processing method and apparatus, device, and computer readable storage medium |
| CN115185285A (en) * | 2022-09-06 | 2022-10-14 | 深圳市信诚创新技术有限公司 | Automatic obstacle avoidance method, device and equipment for dust collection robot and storage medium |
| CN115472039A (en) * | 2021-06-10 | 2022-12-13 | 上海博泰悦臻网络技术服务有限公司 | Information processing method and related product |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103823466A (en) * | 2013-05-23 | 2014-05-28 | 电子科技大学 | Path planning method for mobile robot in dynamic environment |
| CN107065880A (en) * | 2017-05-16 | 2017-08-18 | 南京农业大学 | Remote operating agricultural vehicle man-computer cooperation paths planning method based on dynamic guiding point |
| CN107092252A (en) * | 2017-04-11 | 2017-08-25 | 杭州光珀智能科技有限公司 | A kind of robot automatic obstacle avoidance method and its device based on machine vision |
| CN107894773A (en) * | 2017-12-15 | 2018-04-10 | 广东工业大学 | A kind of air navigation aid of mobile robot, system and relevant apparatus |
-
2018
- 2018-08-03 CN CN201810878544.6A patent/CN108958263A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103823466A (en) * | 2013-05-23 | 2014-05-28 | 电子科技大学 | Path planning method for mobile robot in dynamic environment |
| CN107092252A (en) * | 2017-04-11 | 2017-08-25 | 杭州光珀智能科技有限公司 | A kind of robot automatic obstacle avoidance method and its device based on machine vision |
| CN107065880A (en) * | 2017-05-16 | 2017-08-18 | 南京农业大学 | Remote operating agricultural vehicle man-computer cooperation paths planning method based on dynamic guiding point |
| CN107894773A (en) * | 2017-12-15 | 2018-04-10 | 广东工业大学 | A kind of air navigation aid of mobile robot, system and relevant apparatus |
Non-Patent Citations (2)
| Title |
|---|
| 石琳: "《智能虚拟环境中的人工情感研究》", 31 July 2015, 武汉大学出版社 * |
| 陈孟元: "《移动机器人SLAM、目标跟踪及路径规划》", 31 May 2018, 北京航空航天大学出版社 * |
Cited By (43)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109445442A (en) * | 2018-12-17 | 2019-03-08 | 中新智擎科技有限公司 | A kind of robot obstacle-avoiding control method, device, storage medium and robot |
| CN109523574A (en) * | 2018-12-27 | 2019-03-26 | 联想(北京)有限公司 | A kind of run trace prediction technique and electronic equipment |
| WO2020151664A1 (en) * | 2019-01-25 | 2020-07-30 | 长城汽车股份有限公司 | Vehicle-mounted control device, field end positioning device, and vehicle |
| JP7174487B2 (en) | 2019-04-25 | 2022-11-17 | 株式会社クボタ | agricultural machine |
| CN113766824A (en) * | 2019-04-25 | 2021-12-07 | 株式会社久保田 | Harvester, obstacle determination program, recording medium having obstacle determination program recorded thereon, obstacle determination method, agricultural machine, control program, recording medium having control program recorded thereon, and control method |
| CN113766824B (en) * | 2019-04-25 | 2023-07-14 | 株式会社久保田 | Harvester, obstacle determination program, recording medium, and obstacle determination method |
| JP2020178618A (en) * | 2019-04-25 | 2020-11-05 | 株式会社クボタ | Agricultural implement |
| WO2020218464A1 (en) * | 2019-04-25 | 2020-10-29 | 株式会社クボタ | Harvester, obstacle determination program, recording medium on which obstacle determination program is recorded, obstacle determination method, agricultural work machine, control program, recording medium on which control program is recorded, and control method |
| CN110175523A (en) * | 2019-04-26 | 2019-08-27 | 南京华捷艾米软件科技有限公司 | A kind of self-movement robot animal identification and hide method and its storage medium |
| CN110175523B (en) * | 2019-04-26 | 2021-05-14 | 南京华捷艾米软件科技有限公司 | Self-moving robot animal identification and avoidance method and storage medium thereof |
| CN110370273B (en) * | 2019-06-27 | 2021-04-09 | 华为技术有限公司 | Robot obstacle avoidance method, device and system |
| CN110315538A (en) * | 2019-06-27 | 2019-10-11 | 深圳乐动机器人有限公司 | A kind of method, apparatus showing barrier on the electronic map and robot |
| CN110370273A (en) * | 2019-06-27 | 2019-10-25 | 华为技术有限公司 | A kind of Obstacle Avoidance, device and system |
| CN110370284A (en) * | 2019-08-07 | 2019-10-25 | 北京凌天世纪控股股份有限公司 | A kind of automatic control system of explosion-proof autonomous crusing robot |
| CN112449106A (en) * | 2019-09-03 | 2021-03-05 | 东芝泰格有限公司 | Frame plate shooting device and information processing device |
| CN112449106B (en) * | 2019-09-03 | 2022-05-31 | 东芝泰格有限公司 | Frame plate shooting device and information processing device |
| CN110658819A (en) * | 2019-09-30 | 2020-01-07 | 北京猎户星空科技有限公司 | Obstacle avoidance method and device, electronic equipment and storage medium |
| WO2021063247A1 (en) * | 2019-09-30 | 2021-04-08 | 北京猎户星空科技有限公司 | Obstacle avoidance method and apparatus, electronic device and storage medium |
| EP4026666A4 (en) * | 2019-10-31 | 2023-02-08 | Lingdong Technology (Beijing) Co. Ltd | Autonomous moving device and warehouse logistics system |
| WO2021082964A1 (en) * | 2019-10-31 | 2021-05-06 | 灵动科技（北京）有限公司 | Autonomous moving device and warehouse logistics system |
| CN110653837A (en) * | 2019-10-31 | 2020-01-07 | 灵动科技（北京）有限公司 | Autonomous moving device and warehouse logistics system |
| CN110653837B (en) * | 2019-10-31 | 2021-08-13 | 灵动科技（北京）有限公司 | Autonomous moving device and warehouse logistics system |
| CN113580155A (en) * | 2019-10-31 | 2021-11-02 | 灵动科技（北京）有限公司 | Autonomous moving device and warehouse logistics system |
| CN111026115A (en) * | 2019-12-13 | 2020-04-17 | 华南智能机器人创新研究院 | Robot obstacle avoidance control method and device based on deep learning |
| CN111390909A (en) * | 2020-03-30 | 2020-07-10 | 上海应用技术大学 | Obstacle avoidance method and system for industrial robot |
| CN111390909B (en) * | 2020-03-30 | 2022-06-28 | 上海应用技术大学 | Obstacle avoidance method and system for industrial robot |
| CN111427357A (en) * | 2020-04-14 | 2020-07-17 | 北京石头世纪科技股份有限公司 | Robot obstacle avoidance method and device and storage medium |
| CN113848871A (en) * | 2020-06-28 | 2021-12-28 | 北京极智嘉科技股份有限公司 | Method, device and system for controlling robot to avoid obstacle |
| CN112034856B (en) * | 2020-09-08 | 2024-05-17 | 深圳优地科技有限公司 | Robot travelling method, device, equipment and storage medium |
| CN112034856A (en) * | 2020-09-08 | 2020-12-04 | 深圳优地科技有限公司 | Robot traveling method, device, equipment and storage medium |
| CN112405532A (en) * | 2020-11-06 | 2021-02-26 | 深圳优地科技有限公司 | Movement control method and device and robot |
| CN112405532B (en) * | 2020-11-06 | 2021-12-28 | 深圳优地科技有限公司 | Movement control method and device and robot |
| WO2022127439A1 (en) * | 2020-12-17 | 2022-06-23 | 深圳市普渡科技有限公司 | Robot obstacle avoidance processing method and apparatus, device, and computer readable storage medium |
| CN112975989B (en) * | 2021-04-21 | 2021-08-03 | 炬星科技（深圳）有限公司 | Robot safe operation method, equipment and storage medium |
| CN112975989A (en) * | 2021-04-21 | 2021-06-18 | 炬星科技（深圳）有限公司 | Robot safe operation method, equipment and storage medium |
| CN113219969A (en) * | 2021-04-22 | 2021-08-06 | 深圳拓邦股份有限公司 | Obstacle avoidance control method and device for floor washing robot and floor washing robot |
| CN115472039B (en) * | 2021-06-10 | 2024-03-01 | 上海博泰悦臻网络技术服务有限公司 | Information processing method and related product |
| CN115472039A (en) * | 2021-06-10 | 2022-12-13 | 上海博泰悦臻网络技术服务有限公司 | Information processing method and related product |
| CN114363826A (en) * | 2021-12-22 | 2022-04-15 | 长沙智能驾驶研究院有限公司 | Information prompting method, device, equipment and readable storage medium |
| CN114415726A (en) * | 2022-01-18 | 2022-04-29 | 江苏锐天智能科技股份有限公司 | Unmanned aerial vehicle obstacle avoidance control system and method based on image analysis |
| CN114415726B (en) * | 2022-01-18 | 2023-01-03 | 江苏锐天智能科技股份有限公司 | Unmanned aerial vehicle obstacle avoidance control system and method based on image analysis |
| CN114348022A (en) * | 2022-01-21 | 2022-04-15 | 北京三快在线科技有限公司 | Control method and device for unmanned vehicle, storage medium and electronic equipment |
| CN115185285A (en) * | 2022-09-06 | 2022-10-14 | 深圳市信诚创新技术有限公司 | Automatic obstacle avoidance method, device and equipment for dust collection robot and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108958263A (en) | A kind of Obstacle Avoidance and robot | |
| US11673269B2 (en) | Method of identifying dynamic obstacle and robot implementing same | |
| Hirose et al. | Gonet: A semi-supervised deep learning approach for traversability estimation | |
| US8214082B2 (en) | Nursing system | |
| EP3510986B1 (en) | Mobility device and mobility system | |
| CN111367282A (en) | Robot navigation method and system based on multimode perception and reinforcement learning | |
| CN106808482A (en) | A kind of crusing robot multisensor syste and method for inspecting | |
| KR102615685B1 (en) | Method for estimating location by synchronizing multi-sensors and robot implementing it | |
| CN111906778B (en) | Robot safety control method and device based on multiple perceptions | |
| CN104718507A (en) | Autonomous traveling device traveling-information generation device, method, and program, and autonomous traveling device | |
| Shi et al. | Human-aware robot motion planning with velocity constraints | |
| KR101014531B1 (en) | Leg detecting method using range sensor, moving robot and human follow driving method for moving robot using the same | |
| JPH0830327A (en) | Active environment recognition system | |
| KR20190094130A (en) | Stroller robot and nethod for controlling the same based on user-recognition | |
| CN116038684A (en) | Robot collision early warning method based on vision | |
| CN112223273A (en) | Industrial robot visual detection and obstacle avoidance system | |
| CN115416047A (en) | Blind assisting system and method based on multi-sensor quadruped robot | |
| JP2019036192A (en) | Mobile robot which simulates walking of pedestrian | |
| Ghandour et al. | A hybrid collision avoidance system for indoor mobile robots based on human-robot interaction | |
| Chen et al. | A survey on deep-learning methods for pedestrian behavior prediction from the egocentric view | |
| Ali et al. | Smart wheelchair maneuvering among people | |
| JP7342664B2 (en) | Robots, control methods, and programs | |
| Skovfoged et al. | ” Is it There or Not?” Why Augmented White Canes Do Not Need to Provide Detailed Feedback about Obstacles | |
| Liang et al. | Application of human body posture recognition technology in robot platform for nursing empty-nesters | |
| Saegusa et al. | Development of a guide-dog robot: Leading and recognizing a visually-handicapped person using a LRF |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181207 |
|
| RJ01 | Rejection of invention patent application after publication |