CN110497405A - For controling the force feedback man-machine collaboration anticollision detection method and module of integral control system - Google Patents
For controling the force feedback man-machine collaboration anticollision detection method and module of integral control system Download PDFInfo
- Publication number
- CN110497405A CN110497405A CN201910746724.3A CN201910746724A CN110497405A CN 110497405 A CN110497405 A CN 110497405A CN 201910746724 A CN201910746724 A CN 201910746724A CN 110497405 A CN110497405 A CN 110497405A
- Authority
- CN
- China
- Prior art keywords
- robot
- collision
- control system
- module
- detection
- 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
- B25J19/00—Accessories fitted to manipulators, e.g. for monitoring, for viewing; Safety devices combined with or specially adapted for use in connection with manipulators
- B25J19/0095—Means or methods for testing manipulators
-
- 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
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
- B25J9/1607—Calculation of inertia, jacobian matrixes and inverses
Abstract
For controling the force feedback man-machine collaboration anticollision detection method and module of integral control system, wherein method includes S1, establishes robot dynamics' equation on predetermined robot platform；The disturbance observer of S2, construction based on the constant collision detection operator of robot energy and based on generalized momentum variable quantity；S3, it is based on robot system electric current Real-time Feedback, determines the relationship between each joint torque and impact force；S4, the testing result based on collision detection model work out different Safeguard tactics for different crash scenarios；S5, simulating, verifying and optimization are carried out to the validity of robot collision detection operator and the reasonability of Safeguard tactics based on ADAMS-Simulink union simulation platform；S6, avoidance protection safety strategy actual effect of the verifying assessment based on force feedback.The present invention has the advantages that there is collision detection capabilities under the premise of not increasing system complexity and overall cost.
Description
Technical field
The invention belongs to the robot field that cooperates, especially a kind of force feedback for controling integral control system is man-machine
The anticollision detection method that cooperates and module.
Background technique
With the progress for controling integrated technique, industrial robot performance is obviously improved, and function is more and more abundant, ring
Border adaptability is more and more stronger, and operating efficiency is higher and higher, has liberated labour significantly.However, industrial robot is more suitable at present
Do some labor-intensive repeated works, it is difficult to be competent at some work for needing priori knowledge and experience accumulation, and use people
Machine cooperative work mode can provide a kind of ideal solution.Then cooperation machine comes into being, under man-machine collaboration complex environment
Quickly detection and security protection high-speed decision are most important to support personnel's safety and robot security for robot collision.Due to association
Make that the usual inertia of robot is larger, and environment sensing ability is limited, manually makees movable randomness and uncertainty in addition, because of this person
There is interference even collision safety hidden danger in machine cooperating process, to integral control system is controled, more stringent requirements are proposed for this,
It is required that it can have certain collision detection capabilities under the premise of not increasing system complexity and overall cost, and can be according to reality
Border operating condition takes optimal Safeguard tactics.
Summary of the invention
To solve the above-mentioned problems, the present invention provides a kind of before not increasing system complexity and overall cost to society
Put with collision detection capabilities for control integral control system force feedback man-machine collaboration anticollision detection method and
Module.
The technical scheme is that providing a kind of for controling the force feedback man-machine collaboration anti-collision of integral control system
Detection method is hit, is included the following steps:
S1, on predetermined robot platform, link rod coordinate system is established using D-H parametric method, and according to lagrangian dynamics public affairs
Formula establishes robot dynamics' equation；
S2, according to robot dynamics' equation and the equation of momentum, construction based on the constant collision detection operator of robot energy and
Disturbance observer based on generalized momentum variable quantity；
S3, it is based on robot system electric current Real-time Feedback, determines the relationship between each joint torque and impact force, and provide machine
People's Jacobian matrix method for solving, and analyze the validity of its detection collision；
S4, the testing result based on collision detection model work out different peaces in conjunction with actual condition for different crash scenarios
Full protection strategy；
S5, based on ADAMS-Simulink union simulation platform to the validity and security protection plan of robot collision detection operator
Reasonability slightly carries out simulating, verifying and optimization；
S6, predetermined robot platform, avoidance protection safety strategy actual effect of the verifying assessment based on force feedback are based on.
As improvement of the present invention, the Safeguard tactics include: (1), collision after stop, i.e., robot control system
Control system allows servo-driver disconnection enabled at once after system detects collision alarm；Alternatively, robot control after (2), collision
System switching control mode processed, is converted to torque mode for mode position；Alternatively, robot changes original fortune after (3), collision
Dynamic rail mark, leaves collision area.
Further include following steps before S1 step as improvement of the present invention:
S11, monocular dual-view stereoscopic Matching Model of the building based on SVS, optimize geometry constraint conditions on loss function, pass through
Left and right View synthesis process and dual-view stereoscopic matching, realize the accurate estimation that target depth is detected in monocular image；
S12, the RGB image based on monocular cam acquisition carry out the feature extraction of depth convolution using ResNet model；
S13, according to skeleton joint geometry priori knowledge and interarticular correlativity, optimize double branch depth convolutional Neurals
Network structure design, realizes the synchronization process of artis and its joint incidence relation, wherein a branch is by probability thermal map and partially
The mode that shifting amount combines carries out skeleton key point recurrence, the joint related information of more people in a branch detection image, and leads to
It crosses bipartite graph matching and forms skeleton sequence data；
S14, based on Microsoft's COCO data set, in conjunction with industrial human-computer cooperation scene feature reconstruct human skeleton image data
Collection carries out joint point data mark using Shanghai Communications University open source alphapose, assists in conjunction with manually adjusting to obtain towards industry
Make the attitude data collection of scene.
The present invention also provides a kind of for controling the force feedback man-machine collaboration anticollision detection module of integral control system,
Include:
Kinetics equation establishes module, for establishing link rod coordinate system using D-H parametric method in predetermined robot platform, and
Robot dynamics' equation is established according to lagrangian dynamics formula；
Collision detection operator and disturbance observer establish module, and according to robot dynamics' equation and the equation of momentum, construction is based on
The constant collision detection operator of robot energy and the disturbance observer based on generalized momentum variable quantity；
Data analysis module is based on robot system electric current Real-time Feedback, determines the relationship between each joint torque and impact force,
And robot Jacobian matrix method for solving is provided, and analyze the validity of its detection collision；
Safeguard tactics work out module, based on the testing result of collision detection model, for different crash scenarios, in conjunction with reality
Operating condition works out different Safeguard tactics；
Simulating, verifying and optimization module have robot collision detection operator based on ADAMS-Simulink union simulation platform
Effect property and the reasonability of Safeguard tactics carry out simulating, verifying and optimization；
Actual effect authentication module is based on predetermined robot platform, avoidance protection safety strategy of the verifying assessment based on force feedback
Actual effect.
As improvement of the present invention, the Safeguard tactics include: (1), collision after stop, i.e., robot control system
Control system allows servo-driver disconnection enabled at once after system detects collision alarm；Alternatively, robot control after (2), collision
System switching control mode processed, is converted to torque mode for mode position；Alternatively, robot changes original fortune after (3), collision
Dynamic rail mark, leaves collision area.
As improvement of the present invention, the invention also includes:
Monocular dual-view stereoscopic matching module, for constructing the monocular dual-view stereoscopic Matching Model based on SVS, in loss function
Upper optimization geometry constraint conditions are matched by left and right View synthesis process and dual-view stereoscopic, are realized in monocular image and are detected mesh
Mark the accurate estimation of depth；
It is special to carry out depth convolution using ResNet model based on the RGB image of monocular cam acquisition for convolution characteristic extracting module
Sign is extracted；
Skeleton key point processing module, it is excellent according to skeleton joint geometry priori knowledge and interarticular correlativity
Change double branch depth convolutional neural networks structure designs, the synchronization process of artis and its joint incidence relation is realized, wherein one
Branch carries out skeleton key point recurrence, more people in a branch detection image in such a way that probability thermal map and offset combine
Joint related information, and pass through bipartite graph matching formed skeleton sequence data；
Human skeleton image data processing module, based on Microsoft's COCO data set, in conjunction with industrial human-computer cooperation scene feature
Human skeleton image data set is reconstructed, point data mark in joint is carried out using Shanghai Communications University open source alphapose, in conjunction with people
Work adjustment obtains the attitude data collection towards industrial collaboration scene.
The present invention has the advantages that there is collision detection capabilities under the premise of not increasing system complexity and overall cost.
Detailed description of the invention
Fig. 1 is a kind of process blocks schematic diagram of embodiment of the method for the present invention.
Fig. 2 is the frame structure schematic diagram of the embodiment of the method for the present invention.
Fig. 3 is a kind of structural schematic diagram of embodiment of module of the present invention.
Fig. 4 is the refinement structural schematic diagram of Fig. 3.
Specific embodiment
Referring to Figure 1, what Fig. 1 was disclosed is a kind of for controling the force feedback man-machine collaboration anticollision of integral control system
Detection method,
S1 establishes link rod coordinate system in predetermined robot platform, using improved D-H parametric method, and according to Lagrangian power
It learns formula and establishes robot dynamics' equation；
S2, according to robot dynamics' equation and the equation of momentum, construction based on the constant collision detection operator of robot energy and
Disturbance observer based on generalized momentum variable quantity；
S3, it is based on robot system electric current Real-time Feedback, determines the relationship between each joint torque and impact force, and provide machine
People's Jacobian matrix method for solving, and analyze the validity of its detection collision；
S4, collision detection model inspection is based on as a result, being directed to different crash scenarios, in conjunction with actual condition, it is anti-to work out different safety
Shield strategy is minimized and is adversely affected caused by robot collision；
S5, based on ADAMS-Simulink union simulation platform to the validity and security protection plan of robot collision detection operator
Reasonability slightly carries out simulating, verifying and optimization；
S6, avoidance protection safety strategy actual effect of the assessment based on force feedback is verified based on predetermined robot platform.
In the present invention, the response policy after the machine person to person that cooperates is collided in cooperating process is divided into following three
Kind:
(a) stop after colliding, i.e., robot control system detects control system after collision alarm and allows servo-driver at once
It disconnects and enabling, actual effect is as pressed scram button.It is in response to that speed is fast, and real-time is good the advantages of this mode；Its
The disadvantage is that impact and pressure that collision generates can not unload, there are certain security risks.
(b) robot control system switching control mode after colliding, is converted to torque mode for mode position, hence into
Zero-g mode, the motor work in each joint of robot at this time is in torque mode, and the size of torque is for overcoming robot certainly
The gravitational moment and joint-friction torque of body, so that robot is unlikely to fall down hair in the case where brake is not worked
Raw dangerous, actual effect is similar to robot and is in dragging teaching process.Its advantage is that robot has certain flexibility at this time,
Impact force is unloaded.
(c) robot changes original motion profile after colliding, and leaves collision area.Its advantage is that realizing that non-shutdown is kept away
Barrier more intelligently is conducive to guarantee working efficiency；The disadvantage is that, the mistake in motion switch path more demanding to Motion trajectory
There are still risk of collision in journey.
Man-machine coordination, which controls integrated controller, should be able to be applicable in a variety of robot manipulating task environment, and be directed to above-mentioned collision protection
Corresponding switching interface is arranged in security strategy, according to operating environment spatial variations and danger classes, recommends prevention policies for user.
Refer to Fig. 2, for further promoted man-machine collaboration security performance, the present invention can also using it is contactless, non-by
The machine vision method of control is carried out 3D human body operation behaviour attitude detection and man-machine collision under industrial scene from accident source and is advised
Technical research is kept away, man-machine collision accident is avoided to occur in advance.
Have many advantages, such as to facilitate deployment, at low cost according to monocular cam, thought estimated based on binocular depth, is realized:
S11, monocular dual-view stereoscopic Matching Model of the building based on SVS, optimize geometry constraint conditions on loss function, pass through left and right
View synthesis process and dual-view stereoscopic matching, realize the accurate estimation that target depth is detected in monocular image；
S12, the RGB image based on monocular cam acquisition carry out the feature extraction of depth convolution using ResNet model, according to people
Body skeletal joint geometry priori knowledge and interarticular correlativity optimize double branch depth convolutional neural networks structure designs；
S13, the synchronization process for realizing artis and its joint incidence relation, wherein a branch passes through probability thermal map and offset knot
The mode of conjunction carries out skeleton key point recurrence, the joint related information of more people in a branch detection image, and passes through two points
Figure matching forms skeleton sequence data；
It S14, is to guarantee model applicability and reliability, based on Microsoft's COCO data set, and for industrial human-computer cooperation field
Scape feature reconstructs human skeleton image data set, while making workload to reduce data set, is increased income using Shanghai Communications University
Alphapose carries out joint point data mark, obtains the attitude data collection towards industrial collaboration scene in conjunction with manually adjusting.
Fig. 3 and Fig. 4 are referred to, the present invention also provides a kind of for controling the force feedback man-machine collaboration of integral control system
Anticollision detection module, comprising:
Kinetics equation establishes module 1, for establishing link rod coordinate system using D-H parametric method in predetermined robot platform, and
Robot dynamics' equation is established according to lagrangian dynamics formula；
Collision detection operator and disturbance observer establish module 2, and according to robot dynamics' equation and the equation of momentum, construction is based on
The constant collision detection operator of robot energy and the disturbance observer based on generalized momentum variable quantity；
Data analysis module 3 is based on robot system electric current Real-time Feedback, determines the pass between each joint torque and impact force
System, and robot Jacobian matrix method for solving is provided, and analyze the validity of its detection collision；
Safeguard tactics work out module 4, based on the testing result of collision detection model, for different crash scenarios, in conjunction with reality
Border operating condition works out different Safeguard tactics；
Simulating, verifying and optimization module 5, based on ADAMS-Simulink union simulation platform to robot collision detection operator
Validity and the reasonability of Safeguard tactics carry out simulating, verifying and optimization；
Actual effect authentication module 6 is based on predetermined robot platform, avoidance protection safety strategy of the verifying assessment based on force feedback
Actual effect.
As improvement of the present invention, the invention also includes:
Monocular dual-view stereoscopic matching module 7, for constructing the monocular dual-view stereoscopic Matching Model based on SVS, in loss letter
Optimize geometry constraint conditions on number, matched by left and right View synthesis process and dual-view stereoscopic, realizes and detected in monocular image
The accurate estimation of target depth；
Convolution characteristic extracting module 8 carries out depth convolution using ResNet model based on the RGB image of monocular cam acquisition
Feature extraction；
Skeleton key point processing module 9, it is excellent according to skeleton joint geometry priori knowledge and interarticular correlativity
Change double branch depth convolutional neural networks structure designs, the synchronization process of artis and its joint incidence relation is realized, wherein one
Branch carries out skeleton key point recurrence, more people in a branch detection image in such a way that probability thermal map and offset combine
Joint related information, and pass through bipartite graph matching formed skeleton sequence data；
Human skeleton image data processing module 10, it is special in conjunction with industrial human-computer cooperation scene based on Microsoft's COCO data set
Point reconstruct human skeleton image data set carries out joint point data mark using Shanghai Communications University open source alphapose, in conjunction with
It manually adjusts and obtains the attitude data collection towards industrial collaboration scene.
Preferably, in the present embodiment, the Safeguard tactics include: 1, stop after collision, i.e. robot control system
Control system allows servo-driver disconnection enabled at once after detecting collision alarm；Alternatively, robot control system after 2, collision
System switching control mode, is converted to torque mode for mode position；Alternatively, robot changes original movement rail after 3, collision
Mark leaves collision area.
Claims (6)
1. a kind of for controling the force feedback man-machine collaboration anticollision detection method of integral control system, which is characterized in that packet
Include following steps:
S1, on predetermined robot platform, link rod coordinate system is established using D-H parametric method, and according to lagrangian dynamics public affairs
Formula establishes robot dynamics' equation；
S2, according to robot dynamics' equation and the equation of momentum, construction based on the constant collision detection operator of robot energy and
Disturbance observer based on generalized momentum variable quantity；
S3, it is based on robot system electric current Real-time Feedback, determines the relationship between each joint torque and impact force, and provide machine
People's Jacobian matrix method for solving, and analyze the validity of its detection collision；
S4, the testing result based on collision detection model work out different peaces in conjunction with actual condition for different crash scenarios
Full protection strategy；
S5, based on ADAMS-Simulink union simulation platform to the validity and security protection plan of robot collision detection operator
Reasonability slightly carries out simulating, verifying and optimization；
S6, predetermined robot platform, avoidance protection safety strategy actual effect of the verifying assessment based on force feedback are based on.
2. according to claim 1 for controling the force feedback man-machine collaboration anticollision detection side of integral control system
Method, it is characterised in that: the Safeguard tactics include: (1), collision after stop, i.e., robot control system, which detects, touches
Control system allows servo-driver disconnection enabled at once after hitting signal；Alternatively, robot control system switching control after (2), collision
Mode position is converted to torque mode by molding formula；Alternatively, robot changes original motion profile after (3), collision, leave
Collision area.
3. the force feedback man-machine collaboration anticollision detection according to claim 1 or 2 for controling integral control system
Method, it is characterised in that: further include following steps before S1 step:
S11, monocular dual-view stereoscopic Matching Model of the building based on SVS, optimize geometry constraint conditions on loss function, pass through
Left and right View synthesis process and dual-view stereoscopic matching, realize the accurate estimation that target depth is detected in monocular image；
S12, the RGB image based on monocular cam acquisition carry out the feature extraction of depth convolution using ResNet model；
S13, according to skeleton joint geometry priori knowledge and interarticular correlativity, optimize double branch depth convolutional Neurals
Network structure design, realizes the synchronization process of artis and its joint incidence relation, wherein a branch is by probability thermal map and partially
The mode that shifting amount combines carries out skeleton key point recurrence, the joint related information of more people in a branch detection image, and leads to
It crosses bipartite graph matching and forms skeleton sequence data；
S14, based on Microsoft's COCO data set, in conjunction with industrial human-computer cooperation scene feature reconstruct human skeleton image data
Collection carries out joint point data mark using Shanghai Communications University open source alphapose, assists in conjunction with manually adjusting to obtain towards industry
Make the attitude data collection of scene.
4. a kind of for controling the force feedback man-machine collaboration anticollision detection module of integral control system, which is characterized in that packet
It includes:
Kinetics equation establishes module (1), for establishing link rod coordinate system using D-H parametric method in predetermined robot platform,
And robot dynamics' equation is established according to lagrangian dynamics formula；
Collision detection operator and disturbance observer establish module (2), according to robot dynamics' equation and the equation of momentum, construct base
In the constant collision detection operator of robot energy and disturbance observer based on generalized momentum variable quantity；
Data analysis module (3) is based on robot system electric current Real-time Feedback, determines the pass between each joint torque and impact force
System, and robot Jacobian matrix method for solving is provided, and analyze the validity of its detection collision；
Safeguard tactics work out module (4), based on the testing result of collision detection model, for different crash scenarios, in conjunction with
Actual condition works out different Safeguard tactics；
Simulating, verifying and optimization module (5), based on ADAMS-Simulink union simulation platform to robot collision detection operator
Validity and the reasonability of Safeguard tactics carry out simulating, verifying and optimization；
Actual effect authentication module (6) is based on predetermined robot platform, avoidance protection safety plan of the verifying assessment based on force feedback
Slightly actual effect.
5. the force feedback man-machine collaboration anticollision detection mould according to claim 4 for controling integral control system
Block, which is characterized in that the Safeguard tactics include: (1), collision after stop, i.e., robot control system, which detects, touches
Control system allows servo-driver disconnection enabled at once after hitting signal；Alternatively, robot control system switching control after (2), collision
Mode position is converted to torque mode by molding formula；Alternatively, robot changes original motion profile after (3), collision, leave
Collision area.
6. the force feedback man-machine collaboration anticollision detection according to claim 4 or 5 for controling integral control system
Module, which is characterized in that further include:
Monocular dual-view stereoscopic matching module (7) is losing for constructing the monocular dual-view stereoscopic Matching Model based on SVS
Optimize geometry constraint conditions on function, matched by left and right View synthesis process and dual-view stereoscopic, realizes and examined in monocular image
Survey the accurate estimation of target depth；
Convolution characteristic extracting module (8) carries out depth volume using ResNet model based on the RGB image of monocular cam acquisition
Product feature extraction；
Skeleton key point processing module (9), according to skeleton joint geometry priori knowledge and interarticular correlativity,
Optimize double branch depth convolutional neural networks structure designs, realizes the synchronization process of artis and its joint incidence relation, wherein
One branch carries out skeleton key point recurrence in such a way that probability thermal map and offset combine, more in a branch detection image
The joint related information of people, and skeleton sequence data is formed by bipartite graph matching；
Human skeleton image data processing module (10), based on Microsoft's COCO data set, in conjunction with industrial human-computer cooperation scene
Feature reconstructs human skeleton image data set, carries out joint point data mark, knot using Shanghai Communications University open source alphapose
Conjunction, which manually adjusts, obtains the attitude data collection towards industrial collaboration scene.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910746724.3A CN110497405B (en) | 2019-08-14 | 2019-08-14 | Force feedback man-machine cooperation anti-collision detection method and module for driving and controlling integrated control system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910746724.3A CN110497405B (en) | 2019-08-14 | 2019-08-14 | Force feedback man-machine cooperation anti-collision detection method and module for driving and controlling integrated control system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110497405A true CN110497405A (en) | 2019-11-26 |
| CN110497405B CN110497405B (en) | 2022-05-24 |
Family
ID=68587352
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910746724.3A Active CN110497405B (en) | 2019-08-14 | 2019-08-14 | Force feedback man-machine cooperation anti-collision detection method and module for driving and controlling integrated control system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110497405B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110919633A (en) * | 2019-12-06 | 2020-03-27 | 泉州市微柏工业机器人研究院有限公司 | Robot position deviation method based on torque control |
| CN111152264A (en) * | 2020-01-03 | 2020-05-15 | 北京理工大学 | Precision measurement module for detecting collision force and power of cooperative robot |
| CN113021340A (en) * | 2021-03-17 | 2021-06-25 | 华中科技大学鄂州工业技术研究院 | Robot control method, device, equipment and computer readable storage medium |
| CN114750168A (en) * | 2022-06-14 | 2022-07-15 | 苏州上舜精密工业科技有限公司 | Mechanical arm control method and system based on machine vision |
| CN117409517A (en) * | 2023-10-19 | 2024-01-16 | 光谷技术有限公司 | Voice alarm system and method based on video AI behavior analysis |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160001446A1 (en) * | 2013-02-14 | 2016-01-07 | Commissariat A L'energie Atomique Et Aux Energies Alternatives | Method for the improved detection of the collision of a robot with its environment, system and computer program product implementing said method |
| CN108582070A (en) * | 2018-04-17 | 2018-09-28 | 上海达野智能科技有限公司 | robot collision detecting system and method, storage medium, operating system |
| CN108772838A (en) * | 2018-06-19 | 2018-11-09 | 河北工业大学 | A kind of mechanical arm safety collision strategy based on outer force observer |
| US20190193269A1 (en) * | 2017-12-27 | 2019-06-27 | Hanwha Precision Machinery Co., Ltd. | Robot control system and method of controlling a robot |
| CN110103222A (en) * | 2019-05-24 | 2019-08-09 | 北京无线电测量研究所 | A kind of industrial robot collision checking method |
-
2019
- 2019-08-14 CN CN201910746724.3A patent/CN110497405B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160001446A1 (en) * | 2013-02-14 | 2016-01-07 | Commissariat A L'energie Atomique Et Aux Energies Alternatives | Method for the improved detection of the collision of a robot with its environment, system and computer program product implementing said method |
| US20190193269A1 (en) * | 2017-12-27 | 2019-06-27 | Hanwha Precision Machinery Co., Ltd. | Robot control system and method of controlling a robot |
| CN108582070A (en) * | 2018-04-17 | 2018-09-28 | 上海达野智能科技有限公司 | robot collision detecting system and method, storage medium, operating system |
| CN108772838A (en) * | 2018-06-19 | 2018-11-09 | 河北工业大学 | A kind of mechanical arm safety collision strategy based on outer force observer |
| CN110103222A (en) * | 2019-05-24 | 2019-08-09 | 北京无线电测量研究所 | A kind of industrial robot collision checking method |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110919633A (en) * | 2019-12-06 | 2020-03-27 | 泉州市微柏工业机器人研究院有限公司 | Robot position deviation method based on torque control |
| CN111152264A (en) * | 2020-01-03 | 2020-05-15 | 北京理工大学 | Precision measurement module for detecting collision force and power of cooperative robot |
| CN113021340A (en) * | 2021-03-17 | 2021-06-25 | 华中科技大学鄂州工业技术研究院 | Robot control method, device, equipment and computer readable storage medium |
| CN114750168A (en) * | 2022-06-14 | 2022-07-15 | 苏州上舜精密工业科技有限公司 | Mechanical arm control method and system based on machine vision |
| CN114750168B (en) * | 2022-06-14 | 2022-09-20 | 苏州上舜精密工业科技有限公司 | Mechanical arm control method and system based on machine vision |
| CN117409517A (en) * | 2023-10-19 | 2024-01-16 | 光谷技术有限公司 | Voice alarm system and method based on video AI behavior analysis |
| CN117409517B (en) * | 2023-10-19 | 2024-05-07 | 光谷技术有限公司 | Voice alarm system and method based on video AI behavior analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110497405B (en) | 2022-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110497405A (en) | For controling the force feedback man-machine collaboration anticollision detection method and module of integral control system | |
| CN106593534B (en) | A kind of intelligent tunnel construction safety monitoring system | |
| CN106250867B (en) | A kind of implementation method of the skeleton tracking system based on depth data | |
| KR101355974B1 (en) | Method and devices for tracking multiple object | |
| CN109822579A (en) | Cooperation robot security's control method of view-based access control model | |
| CN103995747B (en) | Distributed pedestrian detecting system and method based on mobile robot platform | |
| CN108245385B (en) | A kind of device helping visually impaired people's trip | |
| CN107092252A (en) | A kind of robot automatic obstacle avoidance method and its device based on machine vision | |
| CN106426186A (en) | Electrified operation robot autonomous operation method based on multi-sensor information fusion | |
| CN107742097B (en) | Human behavior recognition method based on depth camera | |
| CN103716324B (en) | Virtual mine risk-taking behavior implementation system and method based on multiple agents | |
| CN107229922A (en) | A kind of fatigue driving monitoring method and device | |
| CN107038422A (en) | The fatigue state recognition method of deep learning is constrained based on space geometry | |
| CN101511550A (en) | Method for observation of a person in an industrial environment | |
| CN109875568A (en) | A kind of head pose detection method for fatigue driving detection | |
| CN107932560A (en) | A kind of man-machine safety guard system and means of defence | |
| CN104570731A (en) | Uncalibrated human-computer interaction control system and method based on Kinect | |
| CN104933734A (en) | Multi-Kinect-based human body gesture data fusion method | |
| CN104931057B (en) | A kind of any position localization method, the apparatus and system of robot | |
| CN110689725A (en) | Security robot system for crossing comprehensive inspection and processing method thereof | |
| CN114885133B (en) | Depth image-based equipment safety real-time monitoring method and system and related equipment | |
| EP2476999B1 (en) | Method for measuring displacement, device for measuring displacement, and program for measuring displacement | |
| CN107067018A (en) | A kind of hot line robot bolt recognition methods based on random Hough transformation and SVM | |
| JP2011186576A (en) | Operation recognition device | |
| CN106127723A (en) | Peripheral space characteristics information extraction method |
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 |