CN104932494B - The build mechanism of distribution of obstacles figure in a kind of probabilistic type room - Google Patents

The build mechanism of distribution of obstacles figure in a kind of probabilistic type room Download PDF

Info

Publication number
CN104932494B
CN104932494B CN201510206078.3A CN201510206078A CN104932494B CN 104932494 B CN104932494 B CN 104932494B CN 201510206078 A CN201510206078 A CN 201510206078A CN 104932494 B CN104932494 B CN 104932494B
Authority
CN
China
Prior art keywords
robot
distribution
obstacles
barrier
grid member
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.)
Active
Application number
CN201510206078.3A
Other languages
Chinese (zh)
Other versions
CN104932494A (en
Inventor
刘外喜
吴颢
刘长红
高鹰
陈亮东
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangzhou University
Original Assignee
Guangzhou University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangzhou University filed Critical Guangzhou University
Priority to CN201510206078.3A priority Critical patent/CN104932494B/en
Publication of CN104932494A publication Critical patent/CN104932494A/en
Application granted granted Critical
Publication of CN104932494B publication Critical patent/CN104932494B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

The present invention proposes a kind of build mechanism of distribution of obstacles figure in probabilistic type room, and key step is as follows:1) room area is regarded as and is made of basic grid member, the grid member is square, its size is u;2) robot advance step-length is set to s, s=n*u, wherein n >=1, n are integer;3) center of maximum zone of ignorance in last time exploration result is set as starting point, and robot, which often takes a step forward, detects whether front has barrier;If so, the attribute of grid member barrier is set to;Otherwise, the attribute of grid member is set to clear.In detection process, the place for having obstacle in three faces to avoid robot turns not come out, and the direction turned to is selected by Probability p, if p>0.5.Meanwhile the step-length number preset value for exploring end can be adaptive to purging zone size.The distribution of obstacles figure build mechanism of the present invention can quickly verify indoor border, build irregular indoor distribution of obstacles figure, and the place that can have barrier from three faces, which is produced, to be come.

Description

The build mechanism of distribution of obstacles figure in a kind of probabilistic type room
Technical field
The present invention relates to the structure of distribution of obstacles figure, a kind of structure of distribution of obstacles figure in probabilistic type room is particularly related to Mechanism.
Background technology
More and more extensive with the application of robot, requirement of the people to robot is continuously improved, service robot conduct Important branch in robot application, its importance in production and living also gradually embody.Robot is complete indoors Into a certain task, first have to build indoor distribution of obstacles figure, the structure of indoor distribution of obstacles figure be in robot research most Basis, while be also sixty-four dollar question, how to cause robot quickly to verify indoor border, how to avoid robot from running into During barrier rapidly produce come, how to be established in irregular interior distribution of obstacles figure be those skilled in the art need after The problem of continuous research.
At present, mostly using straight trip cleaning method, can not keeping straight on just, a turn angle continues to keep straight on clean robot at random, should Algorithm is simple, and hardware configuration is simple, but efficiency is than relatively low.Related data shows:The usual first pass of stochastic programming can cover Clean the 65% of region, second time covering 85%, the 3rd time covering 92%, the 4th time covering 98%, not stinting if the time can be with Tend to 100%.But in fact, since clean robot is from charged pool, electricity is limited, with reference to energy expenditure and cleaning replacement rate etc. Parameter, it is difficult gratifying that the sweeping efficiency of the random cleaning method of this blindness, which is,.
So it is contemplated that the distribution map of barrier is explored by robot before cleaning, and cloud is stored that information in Platform.When robot exploration next time distribution of obstacles, pervious information can be utilized, and then improve distribution of obstacles figure Build speed and precision.
The content of the invention
The present invention proposes that a kind of build mechanism of distribution of obstacles figure in probabilistic type room solves robot in the prior art It is difficult to quick contouring, running into barrier cannot have the place of three face barriers to produce from one, it is impossible to irregular The problem of distribution of obstacles figure, is established in interior.
The technical proposal of the invention is realized in this way:The build mechanism of distribution of obstacles figure in a kind of probabilistic type room, bag Include following steps:
Step 1:Room area is regarded as and is made of many basic grid members, the grid member is square, grid member Size is u;Grid member is smaller, and distribution map is more accurate, and grid cell size is more than or equal to the error in robot direction, grid cell Size can be set to 5cm or 10cm.
Step 2:Robot advance step-length is arranged to s, s=n*u, wherein n >=1, n is integer;
Step 3:The starting point of robot is set, robot, which often takes a step forward, detects whether front has barrier;Robot Barrier is run into, the attribute status of grid member is provided with barrier;Robot does not run into barrier, by the category of grid member Character state is arranged to clear.
Where whole indoor edge can not being learned for the first time due to robot, so, the entirety of barrier in target chamber Distribution is including the indoor edge in constantly makeover process, i.e., indoor map can become complete in heuristic process.
Further, the build mechanism of distribution of obstacles figure further includes step 4 in the probabilistic type room, according to the last time Result of detection determines the starting point of robot next time:The last center for exploring the maximum zone of ignorance in result of selection As starting point, repeat step three.
Further, the build mechanism of distribution of obstacles figure further includes step 5 in the probabilistic type room:Robot returns to original Point or institute of robot walking length reach preset value, and exploration then terminates.
Further, the attribute status of grid member is arranged to 0 or 1, wherein 0 represents clear, wherein 1 indicates obstacle Thing.Robot is constantly explored, and the result explored every time and distribution map before are compared, and constantly refresh grid in distribution map The attribute of lattice member, or the grid member with attribute is added, it 0 is also likely to be 1 that its attribute, which is probably,.So, by exploring many times Cheng Hou, indoor edge, and distribution of obstacles situation can be confirmed, so as to build complete indoor distribution of obstacles figure.
Further, it is broken line that control system, which controls the route of travel of robot, and the starting point for setting robot is origin, will The position coordinates of robot is set as D=(x, y), and the unit of x, y is u, and robot advances a step-length toward front, then x 1, y is added to remain unchanged;Toward dead astern advance a step-length, then x subtract 1, y is remained unchanged;Advance a step-length toward front-left, then x Remain unchanged, y adds 1;Toward front-right advance a step-length, then x remain unchanged, y subtracts 1, what robot recorded during traveling Information is I={ ∪ of (x, y)=0 (x, y)=1 }, and I is stored in cloud platform, then calculates distribution of obstacles figure according to I;Meanwhile The distribution map is also stored in cloud platform, and follow-up exploration can be from learning to knowledge, and then improves distribution of obstacles figure structure Speed and precision.
Further, the robot is by four-wheel drive, by control system control robot advance towards four direction or Rotate.
Further, control system control robot, which often rotates, once rotates 90 °.
Further, n=1 is set, robot forward travel distance is more than 0.5u, and the attribute status of grid member is arranged to accessible Thing.
Further, the probability for setting control system to control robot to rotate clockwise 90 ° when running into barrier is p, wherein p >0.5。
Further, control system control robot advances in one direction, counterclockwise or clockwise until running into barrier It is rotated by 90 °.
Beneficial effects of the present invention are:The build mechanism of distribution of obstacles figure is suitable for actual in the probabilistic type room of the present invention Working environment, construction method simply preferably goes.Robot is controlled to advance or rotate towards four direction by control system, every time In one direction, untill it cannot walk.In this mode, after robot runs into barrier, 90 degree are changed, then, along This direction moves on.This process is carried out again and again, can rapidly verify indoor edge.In order to avoid robot Turn-take and can't get out in the place that there is barrier in three faces, we select the direction turned to by Probability p.In order to avoid original Ground is motionless, we set p>0.5.For irregular interior, using the thought similar to integration, robot forward travel distance exceedes 0.5u, you can think the grid member clear.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing There is attached drawing needed in technology description to be briefly described, it should be apparent that, drawings in the following description are only this Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, may be used also To obtain other attached drawings according to these attached drawings.
Fig. 1 is the method flow of build mechanism one embodiment of distribution of obstacles figure in a kind of probabilistic type room of the present invention Figure;
Fig. 2 is the interior of one embodiment of the build mechanism structure of distribution of obstacles figure in a kind of probabilistic type room of the present invention Distribution of obstacles figure;
Fig. 3 is that the state for inventing robot direction vector in the build mechanism of distribution of obstacles figure in a kind of probabilistic type room turns Move figure.
Embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.It is based on Embodiment in the present invention, those of ordinary skill in the art are obtained every other without creative efforts Embodiment, belongs to the scope of protection of the invention.
The build mechanism of distribution of obstacles figure, comprises the following steps in a kind of probabilistic type room:
Step 1:Room area is regarded as and is made of many basic grid members, the grid member is square, grid member Size is u;Grid member is smaller, and distribution map is more accurate, and grid cell size is more than or equal to the error in robot direction, grid cell Size can be set to 5cm or 10cm;
Step 2:Robot advance step-length is arranged to s, s=n*u, wherein n >=1, n is integer;
Step 3:Robot, which often takes a step forward, detects whether front has barrier;Robot runs into barrier, by grid member Attribute status be provided with barrier;Robot does not run into barrier, and the attribute status of grid member is arranged to accessible Thing.In the present embodiment, n=1 is set, robot forward travel distance is more than 0.5u, and the attribute status of grid member is arranged to clear.
Where whole indoor edge can not being learned for the first time due to robot, so, the entirety of barrier in target chamber Distribution is including the indoor edge in constantly makeover process, i.e., indoor map can become complete in heuristic process.
Step 4, the starting point of robot next time is determined according to last result of detection:Selection is last to explore knot The center of maximum zone of ignorance in fruit is as starting point, repeat step three.
Step 5:Robot returns to origin or institute of robot walking length reaches preset value exploration and terminates.
Determine that ith explores the step-length number L terminated as followsi, LiDetermined by equation below:
Li=(1- δ) × (Li-1)+δ×(Qi-1* 1.1) (i=2,3,4 ...)
As i=1, i.e., when exploring for the first time, L1Equal to A, A be with explore the relevant constant of area size, 100 >= A≥10。QiThe grid quantity summation of distribution map after expression ith is explored.δ is to update regulatory factor, 1 > δ > 0, this reality Apply a δ and be set to 0.7.
The attribute status of grid member is arranged to 0 or 1, wherein 0 represents clear, wherein 1 indicates barrier.Machine People constantly explores, and the result explored every time and distribution map before are compared, and constantly refresh grid member in distribution map Attribute, or the grid member with attribute is added, it 0 is also likely to be 1 that its attribute, which is probably,.So, after heuristic process many times, room Interior edge, and distribution of obstacles situation can be confirmed, so as to build complete indoor distribution of obstacles figure.The machine People controls robot to advance or rotate towards four direction by four-wheel drive by control system.
It is broken line that control system, which controls the route of travel of robot, and the starting point for setting robot is origin, by robot Position coordinates be set as D=(x, y), the unit of x, y is u, and robot advances a step-length, i.e. direction initialization toward front Vector (E=1, N=0), then x adds 1, y to remain unchanged;Toward one step-length of dead astern advance, i.e. direction initialization vector (E=-1, N =0), then x subtracts 1, and y is remained unchanged;Toward front-left one step-length of advance, i.e. direction initialization vector (E=0, N=1), then x holdings Constant, y adds 1;Toward front-right one step-length of advance, i.e. direction initialization vector (E=0, N=-1), then x is remained unchanged, and y subtracts 1, machine The information that device people records during traveling is I={ ∪ of (x, y)=0 (x, y)=1 }, and I is stored in cloud platform, is then counted according to I Calculate distribution of obstacles figure.Meanwhile the distribution map is also stored in cloud platform, follow-up exploration can from learning to knowledge, into And improve the speed and precision of distribution of obstacles figure structure.
As shown in Figure 1, in probabilistic type room distribution of obstacles figure build mechanism method flow diagram:
After beginning, the starting point for setting robot is origin, and robot advances in one direction, until running into barrier not It can walk, direction vector is initialized:(E=0, N=1), robot does not turn to, a step-length of advancing;Setting direction vector (E, N), as shown in figure 3, the state transition diagram of direction vector (E, N), runs into barrier, (x, y)=1, the attribute shape of grid member are set State is arranged to 1;Barrier is not run into, sets (x, y)=0, and the attribute status of grid member is arranged to 0;Judge when running into barrier Whether origin is reached, reach origin, i.e. x=0and y=0, end task, or during ith exploration, step-length number LSArrival is set Definite value Li, end task;Not up to origin and LSSetting value is not arrived, i.e. (x ≠ 0ory ≠ 0) and LS<Li, control system control Robot processed turns to 90 ° or turns to 90 ° counterclockwise clockwise, and wherein control system control robot turns to 90 ° clockwise Probability is more than 0.5, then proceedes to not turn to a step-length of advancing, setting direction vector simultaneously judges whether to run into barrier;Do not meet During to barrier, judge whether to reach origin, reach origin and then end task, or during ith exploration, step-length number LSArrival is set Definite value Li, reach setting value and end task, not up to origin and LSSetting value is not arrived, i.e. (x ≠ 0ory ≠ 0) and LS<Li, Do not turn to then, a step-length of advancing, repeats the above steps, finally obtain indoor distribution of obstacles figure as shown in Figure 2, black Square expression runs into barrier, and the attribute status of grid member is arranged to 1, and white square represents not run into barrier, grid The attribute status of lattice member is arranged to 0.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention With within principle, any modification, equivalent replacement, improvement and so on, should all be included in the protection scope of the present invention god.

Claims (6)

1. the build mechanism of distribution of obstacles figure in a kind of probabilistic type room, it is characterised in that comprise the following steps:
Step 1:Room area is regarded as and is made of many basic grid members, the grid member is square, the size of grid member For u;
Step 2:Robot advance step-length is arranged to s, s=n*u, wherein n >=1, n is integer;
Step 3:The starting point of robot is set, robot, which often takes a step forward, detects whether front has barrier;Robot is run into Barrier, barrier is provided with by the attribute status of grid member;Robot does not run into barrier, by the attribute shape of grid member State is arranged to clear;
It is broken line that control system, which controls the route of travel of robot, and robot, which often rotates, once rotates 90 °;Control system control is set Robot processed rotates clockwise 90 ° probability when running into barrier is p, wherein p>0.5;
The starting point for setting robot is origin, and the position coordinates of robot is set as D=(x, y), and the unit of x, y are u, Robot adds 1, y to remain unchanged toward one step-length of front advance, then x;Toward dead astern advance a step-length, then x subtract 1, y keep It is constant;Toward front-left advance a step-length, then x remain unchanged, y adds 1;Toward front-right advance a step-length, then x remain unchanged, y Subtract 1;
The information that robot records during traveling is I={ ∪ of (x, y)=0 (x, y)=1 }, and I is stored in cloud platform, then Distribution of obstacles figure is calculated according to I;Meanwhile the distribution map is also stored in cloud platform, follow-up exploration can from learning to Knowledge, and then improve the speed and precision of distribution of obstacles figure structure;
Step 4:The starting point of robot next time is determined according to last result of detection:Selection is last to be explored in result Maximum zone of ignorance center as starting point, repeat step three.
2. the build mechanism of distribution of obstacles figure in probabilistic type room as claimed in claim 1, it is characterised in that:The probabilistic type The build mechanism of indoor distribution of obstacles figure further includes step 5:Robot returns to origin or institute of robot walking length reaches pre- If value exploration terminates.
3. the build mechanism of distribution of obstacles figure in probabilistic type room as claimed in claim 1 or 2, it is characterised in that:By grid The attribute status of member is arranged to 0 or 1, wherein 0 represents clear, wherein 1 indicates barrier.
4. the build mechanism of distribution of obstacles figure in probabilistic type room as claimed in claim 3, it is characterised in that:The robot By four-wheel drive, robot is controlled to advance or rotate towards four direction by control system.
5. the build mechanism of distribution of obstacles figure in probabilistic type room as claimed in claim 4, it is characterised in that:
N=1 is set, robot forward travel distance is more than 0.5u, and the attribute status of grid member is arranged to clear.
6. the build mechanism of distribution of obstacles figure in probabilistic type room as claimed in claim 1, it is characterised in that:Control system control Robot processed advances in one direction, until running into barrier counterclockwise or rotating clockwise 90 °.
CN201510206078.3A 2015-04-27 2015-04-27 The build mechanism of distribution of obstacles figure in a kind of probabilistic type room Active CN104932494B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510206078.3A CN104932494B (en) 2015-04-27 2015-04-27 The build mechanism of distribution of obstacles figure in a kind of probabilistic type room

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510206078.3A CN104932494B (en) 2015-04-27 2015-04-27 The build mechanism of distribution of obstacles figure in a kind of probabilistic type room

Publications (2)

Publication Number Publication Date
CN104932494A CN104932494A (en) 2015-09-23
CN104932494B true CN104932494B (en) 2018-04-13

Family

ID=54119698

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510206078.3A Active CN104932494B (en) 2015-04-27 2015-04-27 The build mechanism of distribution of obstacles figure in a kind of probabilistic type room

Country Status (1)

Country Link
CN (1) CN104932494B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105136155B (en) * 2015-09-24 2018-12-14 联想(北京)有限公司 A kind of air navigation aid and electronic equipment
CN107463168A (en) * 2016-06-06 2017-12-12 苏州宝时得电动工具有限公司 Localization method and system, map constructing method and system, automatic running device
CN108481320B (en) * 2017-01-09 2020-03-27 广东宝乐机器人股份有限公司 Robot movement control method and robot
EP3633478B1 (en) * 2017-05-26 2023-10-18 Hangzhou Hikrobot Co., Ltd. Method and device for assessing probability of presence of obstacle in unknown position
CN109917791B (en) * 2019-03-26 2022-12-06 深圳市锐曼智能装备有限公司 Method for automatically exploring and constructing map by mobile device
CN112180910A (en) * 2019-06-18 2021-01-05 北京京东尚科信息技术有限公司 Obstacle sensing method and device for mobile robot
CN111813102B (en) * 2020-06-06 2023-11-21 浙江中力机械股份有限公司 Distributed autonomous robot environment map construction method
CN114148350A (en) * 2021-12-21 2022-03-08 北京三快在线科技有限公司 Control method and device for unmanned equipment

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101619985A (en) * 2009-08-06 2010-01-06 上海交通大学 Service robot autonomous navigation method based on deformable topological map
KR101207535B1 (en) * 2010-12-31 2012-12-03 한양대학교 산학협력단 Image-based simultaneous localization and mapping for moving robot
CN103278170A (en) * 2013-05-16 2013-09-04 东南大学 Mobile robot cascading map building method based on remarkable scenic spot detection
CN103512579A (en) * 2013-10-22 2014-01-15 武汉科技大学 Map building method based on thermal infrared camera and laser range finder
CN103809597A (en) * 2014-02-18 2014-05-21 清华大学 Flight path planning method for unmanned plane and unmanned plane
CN103914068A (en) * 2013-01-07 2014-07-09 中国人民解放军第二炮兵工程大学 Service robot autonomous navigation method based on raster maps
CN104199428A (en) * 2014-09-17 2014-12-10 上海畔慧信息技术有限公司 Swarm robot management server and method implemented by same

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101619985A (en) * 2009-08-06 2010-01-06 上海交通大学 Service robot autonomous navigation method based on deformable topological map
KR101207535B1 (en) * 2010-12-31 2012-12-03 한양대학교 산학협력단 Image-based simultaneous localization and mapping for moving robot
CN103914068A (en) * 2013-01-07 2014-07-09 中国人民解放军第二炮兵工程大学 Service robot autonomous navigation method based on raster maps
CN103278170A (en) * 2013-05-16 2013-09-04 东南大学 Mobile robot cascading map building method based on remarkable scenic spot detection
CN103512579A (en) * 2013-10-22 2014-01-15 武汉科技大学 Map building method based on thermal infrared camera and laser range finder
CN103809597A (en) * 2014-02-18 2014-05-21 清华大学 Flight path planning method for unmanned plane and unmanned plane
CN104199428A (en) * 2014-09-17 2014-12-10 上海畔慧信息技术有限公司 Swarm robot management server and method implemented by same

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
智能移动机器人路径规划方法研究;隋岩;《中国优秀硕士学位论文全文数据库信息科技辑》;20110515;论文第14-23页 *

Also Published As

Publication number Publication date
CN104932494A (en) 2015-09-23

Similar Documents

Publication Publication Date Title
CN104932494B (en) The build mechanism of distribution of obstacles figure in a kind of probabilistic type room
CN105511457B (en) Robot static path planning method
CN109144102A (en) A kind of Path Planning for UAV based on improvement bat algorithm
CN106228819B (en) A kind of traffic signal optimization control method and device of multi-intersection
CN110502032A (en) A kind of unmanned plane cluster formation flight method of Behavior-based control control
CN106598070B (en) More obstacles, barrier-avoiding method and unmanned plane under small obstacle during a kind of agricultural plant protection unmanned plane sprays
CN110991972B (en) Cargo transportation system based on multi-agent reinforcement learning
CN104407619B (en) Multiple no-manned plane under uncertain environment reaches multiple goal approachs simultaneously
CN109974737B (en) Route planning method and system based on combination of safety evacuation signs and reinforcement learning
CN108762264A (en) The dynamic obstacle avoidance method of robot based on Artificial Potential Field and rolling window
CN110332943A (en) A kind of robot complete coverage path planning method quickly traversed
CN107885209A (en) Obstacle avoidance method based on dynamic window and virtual target point
CN111290398B (en) Unmanned ship path planning method based on biological heuristic neural network and reinforcement learning
CN109521794A (en) A kind of multiple no-manned plane routeing and dynamic obstacle avoidance method
CN110471444A (en) UAV Intelligent barrier-avoiding method based on autonomous learning
CN107037809A (en) A kind of unmanned boat collision prevention method based on improvement ant group algorithm
CN109782807A (en) A kind of AUV barrier-avoiding method under back-shaped obstacle environment
CN109407705A (en) A kind of method, apparatus, equipment and the storage medium of unmanned plane avoiding barrier
CN104757911B (en) The cleaning method and Intelligent robot for sweeping floor of Intelligent robot for sweeping floor
Palacios-Gasós et al. Optimal path planning and coverage control for multi-robot persistent coverage in environments with obstacles
CN111562785A (en) Path planning method and system for collaborative coverage of cluster robots
CN113341984A (en) Robot path planning method and device based on improved RRT algorithm
CN110345948A (en) Dynamic obstacle avoidance method based on neural network in conjunction with Q learning algorithm
CN109300144A (en) A kind of pedestrian track prediction technique of mosaic society&#39;s power model and Kalman filtering
CN107992040A (en) The robot path planning method combined based on map grid with QPSO algorithms

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant