CN108818530A - Stacking piston motion planing method at random is grabbed based on the mechanical arm for improving RRT algorithm - Google Patents
Stacking piston motion planing method at random is grabbed based on the mechanical arm for improving RRT algorithm Download PDFInfo
- Publication number
- CN108818530A CN108818530A CN201810602059.6A CN201810602059A CN108818530A CN 108818530 A CN108818530 A CN 108818530A CN 201810602059 A CN201810602059 A CN 201810602059A CN 108818530 A CN108818530 A CN 108818530A
- Authority
- CN
- China
- Prior art keywords
- mechanical arm
- piston
- point
- planning
- algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
- B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
Abstract
It is a kind of that stacking piston motion planing method at random is grabbed based on the mechanical arm for improving RRT algorithm.The pose that six shaft mechanical arms are first described using D-H method, is then calculated the D-H Mo Xing of six shaft mechanical arms；Resettle the ROS analogue system of piston blank material loading platform：The motion planning of mechanical arm is finally mainly completed using improved RRT algorithm：Planning process is divided into two stages by improved RRT algorithm, is solved the problems, such as avoidance caused by obstacle piston in the first stage and is improved the planning speed of algorithm, and three steps are divided into, and first step carries out 3D to commanded piston peripheral region；Second stage is to introduce transient test and refinement control function, and construct the most short principle in path, to improve the path quality of improved RRT algorithmic rule；The present invention is suitable for mechanical arm and grabs stacking piston at random, saves the time of planning.
Description
Technical field
The present invention relates to industrial robot motion planning technical fields, and in particular to a kind of based on the machine for improving RRT algorithm
Tool arm grabs stacking piston motion planing method at random.
Background technique
China is the first automobile production big country, and " heart " of the piston as automobile engine, is the most key zero
One of part, therefore the processing efficiency of piston is vital to the production of automobile.On piston production line, upper material process is relied on
Artificial crawl and placement, large labor intensity greatly reduce the degree of automation of the entire production line, so being badly in need of a kind of instead of people
The mode of work automatic charging.To solve this problem, a set of blank piston feeding system based on machine vision is established, it is main
It is divided into identification and crawl two parts, is identified by image procossing and identifies target and its posture to be captured；Crawl is to obtain
After taking target and its posture, is grabbed by mechanical arm and pose as required places piston.The motion planning of mechanical arm is to grab
The key technology taken.
In terms of the motion planning of mechanical arm, traditional Grid Method, such as A*, D* and Phi* etc. are multiple for higher-dimension
Miscellaneous problem, environmental information need very big amount of storage, and computational efficiency is very low；Particle group optimizing, ant group optimization and genetic algorithm etc.
Evolution algorithm is suitable for multi-objective problem；Other evolution algorithms, such as artificial bee colony, bacterium look for food optimizing and biology inspire nerve net
Network algorithm etc. is easily trapped into local minimum, and they are very sensitive to the variation of search space size and data；And based on
The motion planning of machine sampling is since its calculating is at low cost, so being more suitable for higher dimensional space problem.It is most popular to adopt at random
Sample algorithm is Quick Extended random tree (RRT) and its various innovatory algorithms, and RRT algorithm solves motion planning speed mistake
Slow problem, but obtained path quality and bad, and the motion planning at random for stacking piston is grabbed for mechanical arm and is asked
Topic, RRT algorithm not can solve the restricted problem of the introducing of the obstacle piston around commanded piston, and calculate when close to commanded piston
The convergence rate of method is also to be improved.
Summary of the invention
In order to overcome the disadvantages of the above prior art, the purpose of the present invention is to provide a kind of based on improvement RRT algorithm
Mechanical arm grabs stacking piston motion planing method at random, is suitable for mechanical arm and grabs stacking piston at random, saves planning
Time.
To achieve the goals above, the technical solution adopted by the present invention is that：
It is a kind of that stacking piston motion planing method at random is grabbed based on the mechanical arm for improving RRT algorithm, include the following steps：
Step 1, the pose that six shaft mechanical arms are described using D-H method, are established the link rod coordinate system of six shaft mechanical arms, determine six
Then the D-H Mo Xing of six shaft mechanical arms is calculated in the D-H parameter of shaft mechanical arm；
Step 2, the ROS analogue system for establishing piston blank material loading platform：
2.1) structure chart of entire ROS analogue system is designed；
2.2) by writing URDF file, each joint of the mechanical arm in analogue system and the position of each platform are described
Set and each part between connection state, redesign the node diagram of analogue system, control each node in analogue system
Between communication；
Step 3, the motion planning that mechanical arm is completed using improved RRT algorithm：Improved RRT algorithm is by planning process
It is divided into two stages, solve the problems, such as avoidance caused by obstacle piston in the first stage and improves the planning speed of algorithm；Second-order
Section is to introduce transient test and refinement control function, and construct the most short principle in path, to improve improved RRT algorithmic rule
Path quality, improved RRT algorithm include the following steps：
3.1) the point cloud Cloud of commanded piston peripheral region is obtainedgoalWith the initial attitude q of mechanical arminitAnd image
The pose p of the commanded piston identifiedgoal, figure is built by the point cloud progress 3D to commanded piston peripheral region and obtains Octrees
Model；
3.2) according to the pose p of current goal workpiecegoalWith the knot of mechanical arm gripper and the detection of Octrees model collision
Fruit determines posture q of the mechanical arm when grabbing target workpiecegoalWith crawl direction dirgoal；
3.3) the pre- posture q of a mechanical arm is introduced in trajectory planningpr, and remember that the location point is pre- posture point Pr, if
The distance between Pr point and target point are l, and the quantity of path point is Num, pass through calculating from Pr for space coordinate conversion formula
Point is to the path point between target point, to complete the planning of first stage；
3.4) second stage is to introduce transient test and refinement control function, and structure on the basis of original RRT algorithm
The most short principle in path is made, is cooked up from Pr point to the path operation origin.
Beneficial effects of the present invention are：
1. building figure by the 3D to commanded piston peripheral region, not needing the pose for identifying obstacle piston and its simplifying mould
Information is occupied in the space that type can obtain commanded piston peripheral obstacle.
2. the distance of the motion planning in the first stage of improved RRT algorithm is shorter, using this characteristic, by calculating
Gripper crawl direction, and then calculate posture of the mechanical arm at Pr point and target point, algorithm is by the time collection of collision detection
In at Pr point and target point, save the time of planning.
3. by space coordinate conversion formula calculate the path point between from Pr point to target point efficiency ratio RRT and its
Innovatory algorithm is higher.
4. improved RRT algorithm can be realized with Successful utilization on piston blank material loading platform and complete to live using mechanical arm
Fill in the feeding of blank.
Detailed description of the invention
Fig. 1 is the ROS system construction drawing of embodiment piston blank material loading platform.
Fig. 2 is the node diagram of embodiment ROS system.
Fig. 3 is embodiment algorithm flow chart.
Fig. 4 is that embodiment grabs direction calculating flow chart.
Fig. 5 is embodiment coordinate system.
Fig. 6 is embodiment success rate with planning time variation line chart.
Specific embodiment
The present invention is described in detail with reference to the accompanying drawings and examples.
It is a kind of that stacking piston motion planing method at random is grabbed based on the mechanical arm for improving RRT algorithm, include the following steps：
Step 1, the pose that six shaft mechanical arms are described using D-H method, are established the link rod coordinate system of six shaft mechanical arms, determine six
Then the D-H Mo Xing of six shaft mechanical arms is calculated in the D-H parameter of shaft mechanical arm；
Piston blank feeding experiment porch based on machine vision by model RS20N six shaft mechanical arm of Kawasaki, blank
The composition such as workbench, intermediate station, transmission belt, three PC machine, kinect camera, industrial camera；
On piston feeding experiment porch, first identifies the posture of a piston in blank workbench, then pass through
Improved RRT algorithm realizes the motion planning of mechanical arm, and piston is grabbed from blank workbench to intermediate station, and on intermediate station
After adjusting posture, finally piston crawl is placed into transmission belt；Choose the key dimension of piston blank experimental subjects：It is maximum
Diameter 120mm, height 125mm；
For six shaft mechanical arm of Kawasaki RS20N used in piston feeding system, each link joint of mechanical arm is initially set up
Coordinate system, the D-H Mo Xing of Kawasaki mechanical arm is then calculated, the results are shown in Table 1,
1 Kawasaki RS20N robot D-H parameter of table
In table 1, aiIndicate the length of i-th of connecting rod；αiThe corner of i-th of connecting rod；diThe offset distance of i-th of connecting rod；θiI-th
The corner in a joint；After the D-H parameter for obtaining mechanical arm, so that it may carry out motion planning to mechanical arm crawl piston；
Step 2, the ROS analogue system for establishing piston blank material loading platform：
2.1) structure chart of entire ROS analogue system is designed；
The ROS system for the piston material loading platform built under Ubuntu 16.04LTS, ROS kinetic platform, mainly
With Rviz as 3D visualization tool, and the motor function packet of mechanical arm is built using MoveIt, it can complete mechanical arm
Positive inverse kinematics, trajectory path planning, collision detection etc., MoveIt system structure of the piston feeding system in ROS
As shown in Figure 1, in user interface, the action control of mechanical arm is realized by move_group_interface, it be responsible for and
Move_group core carries out information interchange, movement, the opening and closing of robot arm end effector, target completed including needs
Piston information etc.；Connection relationship in ROS parameter module, between the URDF document definition physical parameter and component of all parts
Deng defining the motion planning group of mechanical arm in SRDF file, Config file stores the configuration such as path planning, PID control
Parameter；
2.2) by writing URDF file, each joint, large arm, forearm and the gripper of the mechanical arm in analogue system are described
And its connection state in environment between the simplified model of objective table and each part, the node diagram of analogue system is redesigned,
Control the communication between each node in analogue system；
Visual sensor difference sending point cloud and picture on blank platform and intermediate station give move_group core, move_
The mechanical arm state being calculated is sent to Kawasaki mechanical arm by group core, and last mechanical arm completes crawl movement, piston hair
Base material loading platform ROS system operation when node diagram as shown in Fig. 2, representative system oval in Fig. 2 node, every
Connecting line represents the topic of inter-node communication, and/BinPicking is that user writes the section instructed for issuing control mechanical arm
Point ,/move_group are the calculating core of the ROS system of entire piston blank material loading platform, and/joint_state is responsible for publication
The state in each joint of mechanical arm, including position, velocity and acceleration ,/robot_state are responsible for issuing the pose of mechanical arm；
Step 3, the motion planning that mechanical arm is completed using improved RRT algorithm：Improved RRT algorithm is by planning process
It is divided into two stages, solve the problems, such as avoidance caused by obstacle piston in the first stage and improves the planning speed of algorithm；Second-order
Section is to introduce transient test and refinement control function, and construct the most short principle in path, to improve improved RRT algorithmic rule
Path quality.In improved RRT algorithm, step 3.1) to the first stage building tree T 3.3) completed1, step 3.4) completes the
Two-stage building tree T2, finally merge tree T1With tree T2The path finally planned, flow chart such as Fig. 3 of improved RRT algorithm
It is shown, include the following steps：
3.1) the point cloud Cloud of commanded piston peripheral region is obtainedgoalWith the initial attitude q of mechanical arminitAnd image
The pose p of the commanded piston identifiedgoal, figure is built by the point cloud progress 3D to commanded piston peripheral region and obtains Octrees
Model；
Construct initial attitude qinitWith pre- posture qprBetween path tree T1When, Pr point is in the crawl direction of mechanical arm gripper
Upper selection, the length of distance objective point are l；The thinking that conventional motion planning algorithm solves this section of path is usually：Pass through
Image recognition goes out the pose of object, then calculates its simplified model, increase in the collision detection of motion planning the simplified model with
The collision pair of mechanical arm；But in the problem, the obstacle piston around commanded piston is numerous, and is difficult to ensure all correct
Identify posture；When mechanical arm is close to target point, due to carrying out a large amount of collision detection, algorithmic statement can be made slow；
The information of commanded piston posture and 3D point cloud is introduced directly into the first stage of improved RRT motion planning
In, to solve the above problems, it is advantageous that：1) figure is built by the 3D to commanded piston peripheral region, does not need to identify barrier
Hinder the pose of piston and its information is occupied in space that simplified model can obtain commanded piston peripheral obstacle；2) in the first stage
The distance of motion planning is shorter, using this characteristic, by the crawl direction of calculated gripper, and then calculates mechanical arm in Pr
Posture at point and target point, algorithm concentrate on the time of collision detection at Pr point and target point, save the time of planning；
3) higher by the efficiency ratio RRT algorithm for calculating the path point between from Pr point to target point of space coordinate conversion formula；
Because mechanical arm, when moving on this section of path, the direction of gripper remains unchanged, so calculating between from Pr point to target point
Path point process, it is roughly the same with the mode of Pr point is sought, it is only necessary to successively to change the distance value of relative end, so that it may calculate
T2Other path points in tree；
3.2) according to the pose p of current goal workpiecegoalWith the knot of mechanical arm gripper and the detection of Octrees model collision
Fruit determines posture q of the mechanical arm when grabbing target workpiecegoalWith crawl direction dirgoal；The calculating in mechanical arm crawl direction
Flow chart is as shown in Figure 4：
The process of calculating is first to establish the offline crawl direction collection (N of piston1,N2,…,Ns), preservation one is indicated with quaternary number
It is a in offline library；
A piston blank is obtained by the point cloud segmentation algorithm cut-point cloud based on minimum cut, then by Model Matching
Pose, the piston blank are the commanded piston in the motion planning of mechanical arm；It, will be offline according to the posture of commanded piston at this time
Grab direction collection (N1,N2,…,Ns) do a rotation transformation and obtain that commanded piston is corresponding under the pose to grab direction collection
(N1,N2,…,Ns)；Remember enFor the normal vector on ground；Postrotational direction collection is traversed, calculates N by formula (1)iWith enAngle
θi, such as θiThen directly give up the direction not less than 45 °；In formula (1), riAnd piTo grab direction NiThe R's and P of corresponding Eulerian angles
Component；
θi=arccos (cos (ri)cos(pi)) (1)
According to the position of commanded piston, the point cloud of fixed size is intercepted, 3D is then carried out to it and builds figure, obtains Octrees
Model；It carrying out building figure using the library OctoMap, the drawing method of building of the frame is based on Octree (octrees) data structure, and
Occupy situation using the space that probability occupies algorithm for estimating analysis site cloud；
Then the bounding box for acquiring mechanical arm gripper in advance is placed in the direction N for meeting angle requirementiOn, then with
Octrees model carries out collision detection, and this side up, whether movement can collide verifying gripper；If will not collide,
The safety coefficient for then calculating the direction chooses the maximum direction N of safety coefficient after traversal entire offline crawl direction collectioni
For the crawl direction of mechanical arm gripper, then appearance of the mechanical arm when grabbing commanded piston gone out by the computation of inverse- kinematics of mechanical arm
State qgoal；
3.3) the pre- posture q of a mechanical arm is introduced in trajectory planningpr, and remember that the location point is pre- posture point Pr, if
The distance between Pr point and target point are l, and the quantity of path point is Num.The calculating of mechanical arm pose at Pr point：
As shown in figure 5, note world is world coordinate system, target is commanded piston coordinate system, and Pre-tar is at Pr point
Piston coordinate system, Link6 are the coordinate system in the joint 6 of mechanical arm at Pr point,
Pose of the commanded piston under world coordinate system is Ttar：
[xtar ytar ztar] it is coordinate of the commanded piston under world coordinate system, RtarIt is commanded piston with respect to world coordinates
The spin matrix of system, from ΔtarTo ΔPre_tarCoordinate be transformed to Ttar-pre：
Remember that the coordinate from crawl point to joint 6 is transformed to Tholder：
Then mechanical arm is in Pr point, from ΔtarCoordinate system transformation to joint 6 is T12, shown in solution formula such as formula (2)；
T12=Ttar-preTholder (2)
It acquires shown in pose of the joint 6 in world coordinate system such as formula (3) at this time；
TF=TtarT12 (3)
The value in each joint when showing that mechanical arm is at Pr point finally by the inverse kinematics operation of mechanical arm；
3.4) in second stage, initial attitude q is constructedinitWith pre- posture qprBetween path tree T2When, wherein introducing
Refinement control (refinementControl) function and transient test (transitionTest) function, so as to improve road
Diameter, cost function used in transient test function are path machinery work function, definition such as (4) formula：
(1) in formula,It is the difference of two adjacent positive values, ε is a very small amount, and l indicates initial attitude qinitIt arrives
Pre- posture qprDistance；Finally using shortest path as criterion, iteration 20 times, path selection is shortest for tree T2；
In order to verify improved RRT motion planning, it is living from operation origin position to crawl target to test mechanical arm
The motion planning for filling in this path plans that speed is not a problem in the second stage of improved RRT motion planning, so
Mainly test its performance on the path quality of planning；And in the first stage, it mainly observes improved RRT algorithm and is planning
Superiority in speed.
From operation origin position is moved to Pr this stage of point for mechanical arm, in the ROS system of piston blank material loading platform
The effect of RRT-connect algorithm Yu improved RRT motion planning is tested in system.Estimate since algorithm is all based on probability
Meter, so the result planned every time is all not quite alike, chooses two kinds of algorithms and obtain typical rail in 100 planning experiments
Mark, the line smoothing of the geometric locus ratio RRT-connect of improved RRT algorithmic rule, and length of curve is short, shows improved
The quality in RRT algorithmic rule path is integrally better than RRT-connect algorithm.
For moving to this section of short path of target point from Pr point, existing RRT algorithm need to meet direction constraint condition from
And it avoids colliding with non-targeted piston, therefore computational efficiency is not high.Fig. 6 illustrates the planning of RRT-connect algorithm
Time and the line chart for being formulated for power relation.Defining planning time is in order to which successfully whether planning outbound path is (no for distinguished number
Then when planning failure, program can be run down always).From fig. 6, it can be seen that it is more than 6 seconds that RRT algorithm, which only works as planning time,
When, the success rate of planning can be only achieved 100%, and computational efficiency is very poor.Improved RRT algorithm because calculate crawl direction it
Afterwards, so that it may which guarantee, which can centainly be found out from Pr point, moves to this section of short path of target point, so not needing planning time
Whether distinguished number can plan outbound path.Table 2 compared improved RRT algorithm and RRT-connect algorithm and move from Pr point
Calculating effect on to this section of path of target point in 100 emulation experiments.
2 Riming time of algorithm of table compares
Note：1) improved RRT algorithm calculates 9 path points here
Improved RRT algorithm is in time than directly using fast 2 numbers of RRT-connect algorithm as can be seen from Table 2
Magnitude.Although improved RRT algorithm moves on this section of path of target point from Pr point, calculate the time can be with choosing on the path
The quantity of path point increases and increases, but since this path itself is shorter, does not need too many path point, so the time is not yet
It will increase too many.
Claims (1)
1. a kind of grab stacking piston motion planing method at random based on the mechanical arm for improving RRT algorithm, which is characterized in that including
Following steps：
Step 1, the pose that six shaft mechanical arms are described using D-H method, are established the link rod coordinate system of six shaft mechanical arms, determine six axis machines
Then the D-H Mo Xing of six shaft mechanical arms is calculated in the D-H parameter of tool arm；
Step 2, the ROS analogue system for establishing piston blank material loading platform：
2.1) structure chart of entire ROS analogue system is designed；
2.2) by writing URDF file, each joint of the mechanical arm in description analogue system and the position of each platform, with
And the connection state between each part, the node diagram of analogue system is redesigned, is controlled between each node in analogue system
Communication；
Step 3, the motion planning that mechanical arm is completed using improved RRT algorithm：Planning process is divided into two by improved RRT algorithm
A stage solves the problems, such as avoidance caused by obstacle piston in the first stage and improves the planning speed of algorithm；Second stage is to draw
Enter transient test and refinement control function, and construct the most short principle in path, to improve the path matter of improved RRT algorithmic rule
Amount, improved RRT algorithm include the following steps：
3.1) the point cloud Cloud of commanded piston peripheral region is obtainedgoalWith the initial attitude q of mechanical arminitAnd image recognition
The pose p of commanded piston outgoal, figure is built by the point cloud progress 3D to commanded piston peripheral region and obtains Octrees model；
3.2) according to the pose p of current goal workpiecegoalWith mechanical arm gripper with the detection of Octrees model collision as a result, determining
Posture q of the mechanical arm when grabbing target workpiece outgoalWith crawl direction dirgoal；
3.3) the pre- posture q of a mechanical arm is introduced in trajectory planningpr, and remember that the location point is pre- posture point Pr, if Pr point
The distance between target point is l, and the quantity of path point is Num, by space coordinate conversion formula calculate from Pr point to
Path point between target point, to complete the planning of first stage；
3.4) second stage is to introduce transient test and refinement control function, and construct road on the basis of original RRT algorithm
The most short principle of diameter, is cooked up from Pr point to the path operation origin.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810602059.6A CN108818530B (en) | 2018-06-12 | 2018-06-12 | Mechanical arm grabbing scattered stacking piston motion planning method based on improved RRT algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810602059.6A CN108818530B (en) | 2018-06-12 | 2018-06-12 | Mechanical arm grabbing scattered stacking piston motion planning method based on improved RRT algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108818530A true CN108818530A (en) | 2018-11-16 |
| CN108818530B CN108818530B (en) | 2020-05-15 |
Family
ID=64143698
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810602059.6A Active CN108818530B (en) | 2018-06-12 | 2018-06-12 | Mechanical arm grabbing scattered stacking piston motion planning method based on improved RRT algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108818530B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109940614A (en) * | 2019-03-11 | 2019-06-28 | 东北大学 | A kind of quick motion planning method of the more scenes of mechanical arm merging memory mechanism |
| CN109940616A (en) * | 2019-03-21 | 2019-06-28 | 佛山智能装备技术研究院 | One kind being based on brain-cerebella model intelligent grabbing system and method |
| CN110625612A (en) * | 2019-08-30 | 2019-12-31 | 南京理工大学 | Visual live working robot collision detection method based on ROS |
| CN112809671A (en) * | 2020-12-30 | 2021-05-18 | 华南理工大学 | Interactive kinematics modeling method of universal serial robot |
| CN113172631A (en) * | 2021-05-11 | 2021-07-27 | 西南交通大学 | Mechanical arm autonomous obstacle avoidance method based on improved RRT algorithm |
| CN113211447A (en) * | 2021-05-27 | 2021-08-06 | 山东大学 | Mechanical arm real-time perception planning method and system based on bidirectional RRT algorithm |
| CN113296498A (en) * | 2021-04-12 | 2021-08-24 | 山东科技大学 | Improved RRT path planning method based on adaptive resolution octree map |
| CN113319013A (en) * | 2021-07-08 | 2021-08-31 | 陕西科技大学 | Apple intelligent sorting method based on machine vision |
| CN113341978A (en) * | 2021-06-10 | 2021-09-03 | 西北工业大学 | Intelligent trolley path planning method based on ladder-shaped barrier |
| WO2022156749A1 (en) * | 2021-01-22 | 2022-07-28 | 熵智科技(深圳)有限公司 | Workpiece grabbing method and apparatus, and computer device and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100174435A1 (en) * | 2009-01-07 | 2010-07-08 | Samsung Electronics Co., Ltd. | Path planning apparatus of robot and method thereof |
| US20110035087A1 (en) * | 2009-08-10 | 2011-02-10 | Samsung Electronics Co., Ltd. | Method and apparatus to plan motion path of robot |
| JP2014073550A (en) * | 2012-10-04 | 2014-04-24 | Seiko Epson Corp | Path searching method, path searching device, robot control device, robot, and program |
| CN104155974A (en) * | 2013-07-29 | 2014-11-19 | 深圳信息职业技术学院 | Path planning method and apparatus for robot fast collision avoidance |
| CN106166750A (en) * | 2016-09-27 | 2016-11-30 | 北京邮电大学 | A kind of modified model D* mechanical arm dynamic obstacle avoidance paths planning method |
| CN106695802A (en) * | 2017-03-19 | 2017-05-24 | 北京工业大学 | Improved RRT<*> obstacle avoidance motion planning method based on multi-degree-of-freedom mechanical arm |
| CN106990777A (en) * | 2017-03-10 | 2017-07-28 | 江苏物联网研究发展中心 | Robot local paths planning method |
-
2018
- 2018-06-12 CN CN201810602059.6A patent/CN108818530B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100174435A1 (en) * | 2009-01-07 | 2010-07-08 | Samsung Electronics Co., Ltd. | Path planning apparatus of robot and method thereof |
| US20110035087A1 (en) * | 2009-08-10 | 2011-02-10 | Samsung Electronics Co., Ltd. | Method and apparatus to plan motion path of robot |
| JP2014073550A (en) * | 2012-10-04 | 2014-04-24 | Seiko Epson Corp | Path searching method, path searching device, robot control device, robot, and program |
| CN104155974A (en) * | 2013-07-29 | 2014-11-19 | 深圳信息职业技术学院 | Path planning method and apparatus for robot fast collision avoidance |
| CN106166750A (en) * | 2016-09-27 | 2016-11-30 | 北京邮电大学 | A kind of modified model D* mechanical arm dynamic obstacle avoidance paths planning method |
| CN106990777A (en) * | 2017-03-10 | 2017-07-28 | 江苏物联网研究发展中心 | Robot local paths planning method |
| CN106695802A (en) * | 2017-03-19 | 2017-05-24 | 北京工业大学 | Improved RRT<*> obstacle avoidance motion planning method based on multi-degree-of-freedom mechanical arm |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109940614B (en) * | 2019-03-11 | 2021-01-22 | 东北大学 | Mechanical arm multi-scene rapid motion planning method integrating memory mechanism |
| CN109940614A (en) * | 2019-03-11 | 2019-06-28 | 东北大学 | A kind of quick motion planning method of the more scenes of mechanical arm merging memory mechanism |
| CN109940616B (en) * | 2019-03-21 | 2022-06-03 | 佛山智能装备技术研究院 | Intelligent grabbing system and method based on brain-cerebellum mode |
| CN109940616A (en) * | 2019-03-21 | 2019-06-28 | 佛山智能装备技术研究院 | One kind being based on brain-cerebella model intelligent grabbing system and method |
| CN110625612A (en) * | 2019-08-30 | 2019-12-31 | 南京理工大学 | Visual live working robot collision detection method based on ROS |
| CN112809671A (en) * | 2020-12-30 | 2021-05-18 | 华南理工大学 | Interactive kinematics modeling method of universal serial robot |
| WO2022156749A1 (en) * | 2021-01-22 | 2022-07-28 | 熵智科技(深圳)有限公司 | Workpiece grabbing method and apparatus, and computer device and storage medium |
| CN113296498A (en) * | 2021-04-12 | 2021-08-24 | 山东科技大学 | Improved RRT path planning method based on adaptive resolution octree map |
| CN113296498B (en) * | 2021-04-12 | 2022-09-27 | 山东科技大学 | Improved RRT path planning method based on adaptive resolution octree map |
| CN113172631A (en) * | 2021-05-11 | 2021-07-27 | 西南交通大学 | Mechanical arm autonomous obstacle avoidance method based on improved RRT algorithm |
| CN113211447A (en) * | 2021-05-27 | 2021-08-06 | 山东大学 | Mechanical arm real-time perception planning method and system based on bidirectional RRT algorithm |
| CN113211447B (en) * | 2021-05-27 | 2023-10-27 | 山东大学 | Mechanical arm real-time perception planning method and system based on bidirectional RRT algorithm |
| CN113341978A (en) * | 2021-06-10 | 2021-09-03 | 西北工业大学 | Intelligent trolley path planning method based on ladder-shaped barrier |
| CN113341978B (en) * | 2021-06-10 | 2023-03-07 | 西北工业大学 | Intelligent trolley path planning method based on ladder-shaped barrier |
| CN113319013A (en) * | 2021-07-08 | 2021-08-31 | 陕西科技大学 | Apple intelligent sorting method based on machine vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108818530B (en) | 2020-05-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108818530A (en) | Stacking piston motion planing method at random is grabbed based on the mechanical arm for improving RRT algorithm | |
| CN112297013B (en) | Robot intelligent grabbing method based on digital twin and deep neural network | |
| CN108972549B (en) | Industrial mechanical arm real-time obstacle avoidance planning and grabbing system based on Kinect depth camera | |
| CN111251295B (en) | Visual mechanical arm grabbing method and device applied to parameterized parts | |
| CN110378325B (en) | Target pose identification method in robot grabbing process | |
| CN111085997A (en) | Capturing training method and system based on point cloud acquisition and processing | |
| Luo et al. | Collision-free path-planning for six-DOF serial harvesting robot based on energy optimal and artificial potential field | |
| CN108196453A (en) | A kind of manipulator motion planning Swarm Intelligent Computation method | |
| CN111462154A (en) | Target positioning method and device based on depth vision sensor and automatic grabbing robot | |
| CN114147708B (en) | Mechanical arm obstacle avoidance path planning method based on improved longhorn beetle whisker search algorithm | |
| Aleotti et al. | Perception and grasping of object parts from active robot exploration | |
| CN109108978A (en) | Three-dimensional space manipulator motion planning method based on study Generalization Mechanism | |
| Pavlichenko et al. | Autonomous dual-arm manipulation of familiar objects | |
| CN112207839A (en) | Mobile household service robot and method | |
| CN114299039B (en) | Robot and collision detection device and method thereof | |
| JP2022187983A (en) | Network modularization to learn high dimensional robot tasks | |
| CN115861780B (en) | Robot arm detection grabbing method based on YOLO-GGCNN | |
| CN116673963A (en) | Double mechanical arm cooperation flexible assembly system and method for unordered breaker parts | |
| Gonçalves et al. | Grasp planning with incomplete knowledge about the object to be grasped | |
| CN113822933B (en) | ResNeXt-based intelligent robot grabbing method | |
| JP2022187984A (en) | Grasping device using modularized neural network | |
| Xu et al. | Vision-Based Intelligent Perceiving and Planning System of a 7-DoF Collaborative Robot | |
| Tao et al. | An improved RRT algorithm for the motion planning of robot manipulator picking up scattered piston | |
| Kawagoshi et al. | Visual servoing using virtual space for both learning and task execution | |
| Li et al. | Path planning of a robotic manipulator for pruning apple trees based on RRT-connect algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |