CN107053179A - A kind of mechanical arm Compliant Force Control method based on Fuzzy Reinforcement Learning - Google Patents
A kind of mechanical arm Compliant Force Control method based on Fuzzy Reinforcement Learning Download PDFInfo
- Publication number
- CN107053179A CN107053179A CN201710263232.XA CN201710263232A CN107053179A CN 107053179 A CN107053179 A CN 107053179A CN 201710263232 A CN201710263232 A CN 201710263232A CN 107053179 A CN107053179 A CN 107053179A
- Authority
- CN
- China
- Prior art keywords
- mechanical arm
- fuzzy
- admittance
- reinforcement learning
- operator
- 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
- B25J9/1628—Programme controls characterised by the control loop
- B25J9/163—Programme controls characterised by the control loop learning, adaptive, model based, rule based expert control
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J13/00—Controls for manipulators
- B25J13/08—Controls for manipulators by means of sensing devices, e.g. viewing or touching devices
- B25J13/085—Force or torque sensors
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Human Computer Interaction (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a kind of mechanical arm Compliant Force Control method based on Fuzzy Reinforcement Learning, using fuzzy reinforcement algorithm, the real-time adjustable strategies of admittance parameter are trained by way of on-line study, the control that moment of face, current joint speed and the Acceleration Control motor that change admittance control strategy after convergence is applied according to operator actively comply with operator is intended to, task is actively followed with complete mechanical arm, corresponding task and environmental model need not be set up, with faster convergence rate and stable actual effect.This method can significantly reduce the working strength of operator, improve positioning precision, help to reduce mechanical arm physical dimension and deadweight, the control that people's machine power interaction models can respond operator well is intended to, with good adaptive ability, machine power interactive experience more remarkable fluency can be made one, power interaction impression when being operated closer in daily life to actual object.
Description
Technical field:
The invention belongs to interaction control technology field, a kind of mechanical arm based on Fuzzy Reinforcement Learning is specifically related to
Compliant Force Control method.
Background technology:
Before robot assisted Minimally Invasive Surgery is carried out, medical personnel need to formulate corresponding according to the personal feature of patient
Operation plan, is selected the incision site of Minimally Invasive Surgery and the initial attitude of each mechanical arm is set with this.In the process of implementation, it is necessary to
Each mechanical arm is drawn to mini-incision position and manually adjusts the joint angles of surgical arm, i.e. operator directly mechanical arm is applied
Plus external force, it is intended to adjust accordingly each connecting rod pose of mechanical arm according to operation.Generally, mechanical arm is used as machinery using decelerator
The transmission link of shoulder joint power, when gearing friction can make to lead diarthrodial pose and adjust to become difficult big retarding.
Solution common at present mainly has two kinds:One kind is that electromagnetic brake is installed after decelerator, passes through control
The disengaging and adhesive of decelerator and rear end power delivery section are realized in the action of electromagnetic brake, i.e., so-called passively to comply with control
Mode.If mechanical arm is pulled in this way, the gravity of mechanical arm itself is all undertaken by operator, and working strength increases
It is big and be difficult to control to performance accuracy.Further, since each joint departs from its motor during pulling, in order to obtain machine
Articulation angle after the adjustment of tool arm, in addition it is also necessary to which extra increase encoder records the change in location in joint.Electromagnetic brake and
The introducing of auxiliary coder can increase the physical dimension and own wt of mechanical arm, and the frequent adhesive of electromagnetic brake can also influence
The absolute positional accuracy of robot.It is on the other side another be achieved in that joint motor be in slave mode under, root
The control for estimating operator according to the stressing conditions of mechanical arm is intended to, and drives mechanical arm auxiliary operator to complete by joint motor
Expected pose adjustment, i.e., it is so-called actively to comply with control mode.Current control method of actively complying with is used at mechanical arm end
Position movement of the installing force approach sensor control machinery arm end effector in cartesian space is held, end is often concerned with
The location track rather than pose adjustment of ending tool, the power interaction position fixed in addition are also not easy to robot linkage posture
Independently adjust, therefore be not particularly suited for the position of actively putting of minimally invasive surgical operation robot and require.In addition, there is also one for such mode
The problem of determining, is then difficult to take into account control accuracy and operating experience, according to variable control according to fixed control parameter model
Parameter model processed is difficult to the submissive and smoothness for ensureing man-machine interactive operation again.
The content of the invention:
To solve the above problems, the present invention proposes a kind of mechanical arm Compliant Force Control side based on Fuzzy Reinforcement Learning
Method.
To reach above-mentioned purpose, technical scheme is as follows:
A kind of mechanical arm Compliant Force Control method based on Fuzzy Reinforcement Learning, comprises the following steps:
S1:Set up admittance Controlling model.
S2:Obtain the motion state of mechanical arm, the moment of face that operator applies and environment return value.
S3:In order to obtain the admittance model parameter strategy being adapted with current environment, according to what is obtained in step S2
Relevant information, the on-line training of admittance model parameter strategy is carried out by Fuzzy Reinforcement Learning, until algorithmic statement, to
Hope the change admittance Controlling model for obtaining and being adapted with current environment.
S4:Admittance parameter adjustable strategies after trained convergence in step S3 are applied to become among admittance Controlling model,
The moment of face and the feedback speed of joint of mechanical arm that admittance Controlling model after change parameter applies according to operator calculate joint
Current speed value is simultaneously sent to joint drive motor.
As the preferred of above-mentioned technical proposal, the Fuzzy Reinforcement Learning in the step S3 specifically includes following steps:
S31:The moment of face that the motion state of mechanical arm and operator are applied is as state variable, in each state variable
Domain in the range of divide multiple fuzzy sets, set up corresponding fuzzy rule and provide discrete movement set.
S32:The degree of membership of each state variable is calculated according to current state input, fuzzy division is carried out to state space,
Calculating has activated the weights corresponding to fuzzy rule.
S33:Discrete movement value is selected according to current admittance model parameter strategy.
S34:Admittance parameter adjustable strategies after trained convergence in step S3 are applied to become among admittance Controlling model,
The moment of face and the feedback speed of joint of mechanical arm applied according to operator calculates joint current speed value and sent to joint and drive
Dynamic motor, so that realizes micro-wound surgical operation mechanical arm actively puts bit function.
S35:Current admittance model parameter strategy is updated according to the environment return value currently obtained.
S36:Above-mentioned S32-S35 steps are repeated, until algorithmic statement.
As the preferred of above-mentioned technical proposal, also comprise the following steps:
S0:The integrated torque sensor in each joint of mechanical arm, the torque sensor be used for detect it is man-machine between connect
Touch square.
As the preferred of above-mentioned technical proposal, in the step S2:
The gravity compensation model of identified off-line mechanical arm by the way of linear regression, so as to obtain the outer of operator's application
Torque.
As the preferred of above-mentioned technical proposal, in the step S2:
The motion state of the mechanical arm includes the speed and acceleration of each joint of mechanical arm.
The beneficial effects of the present invention are:
Relative to passively mode is complied with, the working strength of operator can be significantly reduced, improves positioning precision, helps to subtract
Small physical dimension and deadweight.
Mode is actively complied with relative to preset parameter model, with good adaptive ability, when contact torque increase
When, power interactive controlling model can actively reduce the automatic virtual blocks parameter of environment, the movement velocity of mechanical arm is changed faster, energy
The movement tendency of human arm is enough quickly followed, the operating experience given people can be more laborsaving;Conversely, when contact force (amplitude) is gradually reduced
When, power interactive controlling model can correspondingly increase automatic virtual blocks parameter value to improve the control accuracy of man-machine interaction, auxiliary operation
Person positions, and reduces overshoot.
Mode is actively complied with relative to time-varying parameter model, people's machine power interaction models can respond operator's well
Control is intended to, and makes one machine power interactive experience more remarkable fluency, when being operated closer in daily life to actual object
Power interaction impression.
Brief description of the drawings:
The following drawings is only intended to, in doing schematic illustration and explanation to the present invention, not delimit the scope of the invention.Wherein:
Fig. 1 is a kind of master of mechanical arm Compliant Force Control method based on Fuzzy Reinforcement Learning of one embodiment of the invention
Dynamic Shared control flow chart;
Fig. 2 is the Fuzzy Reinforcement Learning flow chart of one embodiment of the invention.
Embodiment:
As shown in figure 1, a kind of mechanical arm Compliant Force Control method based on Fuzzy Reinforcement Learning of the present invention, including it is as follows
Step:
S1:Set up admittance Controlling model.
S2:Obtain the motion state of mechanical arm, the moment of face that operator applies and environment return value.The mechanical arm
Motion state includes the speed and acceleration in each mechanical arm active rotation joint.
S3:In order to obtain the admittance model parameter strategy being adapted with current environment, according to what is obtained in step S2
Relevant information, the on-line training of admittance model parameter strategy is carried out by Fuzzy Reinforcement Learning, until algorithmic statement, to
Hope the change admittance Controlling model for obtaining and being adapted with current environment.Fuzzy Reinforcement Learning in the step S3 specifically includes following step
Suddenly:
S31:The moment of face that the motion state of mechanical arm and operator are applied is as state variable (I), in each state
The domain X of variableiIn the range of divide multiple fuzzy sets, set up corresponding fuzzy rule and provide discrete movement set A={ u1,
u2,…,un, wherein, μiDiscrete transfer corresponding to the fuzzy rule (being determined as current fuzzy division) that has currently activated
Make.
S32:I is inputted according to current stateiCalculate the degree of membership μ of each state variablei(Ii), mould is carried out to state space
Paste is divided, and calculating has activated fuzzy rule fiCorresponding weight wi, wherein, fiRepresent i-th of fuzzy rule, wiTo be corresponding
Fuzzy rule activity, i.e., the weights of the corresponding discrete movement of current each fringe component.
S33:According to current admittance model parameter strategy selection discrete movement value ui。
S34:Discrete movement value in step S33 is integrated into final action output valve U and the value is used for admittance control
Simulation.
S35:Current admittance model parameter strategy is updated according to the environment return value currently obtained.
S36:Above-mentioned S32-S35 steps are repeated, until algorithmic statement.
S4:Admittance parameter adjustable strategies after trained convergence in step S3 are applied to become among admittance Controlling model,
The moment of face and the feedback speed of joint of mechanical arm applied according to operator calculates joint current speed value and sent to joint and drive
Dynamic motor, so that realizes micro-wound surgical operation mechanical arm actively puts bit function.
The control method of the present embodiment needs the integrated torque sensor in each joint of mechanical arm, and the torque sensor is used
Contact torque between detection is man-machine.In the present embodiment by the way of linear regression identified off-line mechanical arm gravity compensation
Model, so as to obtain the moment of face of operator's application.Different from the teaching mode of conventional industrial robot, micro-wound surgical operation machine
What the preoperative pendulum position process of device people needed regulation is the spatial attitude rather than end effector of each connecting rod of mechanical arm in world coordinates
It is the locus in (cartesian coordinate system), and generally such power interaction implementation is used and installed in mechanical arm execution end
The mode of six-dimension force sensor realizes reciprocal force information gathering, but so does the effect that people interacts with robot progress power that can limit
Position, is unfavorable for the independent adjustment of each connecting rod pose of surgery mechanical arm.Torque sensor is integrated into respectively to solve the above problems
Among mechanical arm active rotation joint, the power interaction locations that can make mechanical arm and external environment by such a mode extend to whole piece
Mechanical arm, moment inspecting and power interactive controlling are also more directly reliable.
In joint space, propose and a kind of be combined with nitrification enhancement based on fuzzy theory with reference to practical application
Become admittance Controlling model framework.In interactive process, make because people plays guiding among whole power interactive controlling loop
With, therefore the operating characteristic of people can have considerable influence to power interaction effect.In addition, Manipulator Dynamics characteristic can be with control mould
The change of shape parameter and change, also can on man-machine interaction produce influence.In order to by the human factor in interaction and dynamic
Mechanical change is considered actively to comply among Controlling model, is passed through using the intensified learning method based on multistep time difference online
The mode of study handles the problem of above-mentioned factor is brought.Meanwhile, the introducing of fuzzy theory helps to solve intensified learning state sky
Between general problem, Compliant Force Control algorithm is received continuous state and input and produce the output of continuous control parameter.This
The outer moment of face in order to extract operator's application, the gravity compensation model of identified off-line mechanical arm by the way of linear regression.
People's machine power interaction control method of proposition need not set up corresponding task and environmental model, with faster convergence rate and stably
Actual effect.
Active Compliance Control process is as shown in figure 1, Fuzzy Reinforcement Learning combines trained receive according to current motion state
Discrete movement selection strategy after holding back obtains current admittance Controlling model parameter, and admittance Controlling model is applied according to operator
The control that moment of face and current joint speed control motor actively comply with operator is intended to, to complete power interactive task.It is fuzzy strong
The single step training process that chemistry is practised calculates the degree of membership of each state variable simultaneously according to current state input first as shown in Fig. 2
Fuzzy division is carried out to state space, discrete movement weights and exploration strategy choosing according to corresponding to the fuzzy rule currently triggered
Select discrete movement value and integrate and export final admittance model parameter value.The new admittance model for changing parameter is used for current people
Machine interaction makes to expect to obtain in interactive process to obtain current environmental feedback and according to value of feedback corrective action weights
The performance indications obtained tend to be maximum, and continuous iteration said process is until algorithmic statement.
The present embodiment is described further by taking simple joint power interactive controlling as an example to the above method, as shown in figure 1,:
The speed currently measured with mechanical arm rotary jointAnd accelerationAnd suffered moment of face τhIt is defeated as state
Enter variable, equidistantly divide 7 fuzzy sets, corresponding number of fuzzy rules in the range of the domain of each state variable respectively
For 343 (7 × 7 × 7).Exported using the automatic virtual blocks parameter in admittance Controlling model as the action of intensified learning, if discrete dynamic
It is 3 to make collection element number, then corresponding fuzzy rule weights number is 1029 (7 × 7 × 7 × 3).Man-machine Compliant Force interactive controlling
Realize and include Strategies Training and interactive application two parts.During Strategies Training, using the minimum acceleration model of people as
Optimality criterion, the man-machine interaction task of execution needed for constantly repeating, nitrification enhancement can be according to interacting with operator
The experience of generation persistently changes the decision strategy of intelligent body until convergence.In man-machine interaction application process, based on intensified learning
Power interactive controlling algorithm fuzzy division is carried out to trigger corresponding fuzzy rule according to current state input, after convergence
Change admittance policy selection each activated the optimal working value component of fuzzy rule, finally by the corresponding activation of fuzzy rule
Spend (being represented by the T norms of corresponding fuzzy set degree of membership) and integrate each working value component, ultimately generate current time admittance control
The parameter value c that simulation is used.Change the moment of face τ that the admittance Controlling model after parameter applies according to operatorhAnd machinery
The feedback speed of shoulder jointCalculate joint current speed valueAnd send to joint drive motor.
A kind of mechanical arm Compliant Force Control method based on Fuzzy Reinforcement Learning described in the present embodiment, using fuzzy reinforcing
Learning algorithm, trains the real-time adjustable strategies of admittance parameter by way of on-line study, the change admittance control strategy after convergence
Moment of face, current joint speed and the Acceleration Control motor applied according to operator actively complies with the control meaning of operator
Figure, actively follows task, without setting up corresponding task and environmental model, with faster convergence rate with complete mechanical arm
With stable actual effect.This method can significantly reduce the working strength of operator, improve positioning precision, contribute to reduction machine
Tool arm configuration size and deadweight, the control that people's machine power interaction models can respond operator well are intended to, with it is good from
Adaptability, can make one machine power interactive experience more remarkable fluency, when being operated closer in daily life to actual object
Power interaction impression.
Obviously, above-described embodiment is only intended to clearly illustrate example, and the not restriction to embodiment.It is right
For those of ordinary skill in the art, can also make on the basis of the above description it is other it is various forms of change or
Change.There is no necessity and possibility to exhaust all the enbodiments.And the obvious change thus extended out or
Among changing still in the protection domain of the invention.
Claims (5)
1. a kind of mechanical arm Compliant Force Control method based on Fuzzy Reinforcement Learning, it is characterised in that comprise the following steps:
S1:Set up admittance Controlling model;
S2:Obtain the motion state of mechanical arm, the moment of face that operator applies and environment return value;
S3:In order to obtain the admittance model parameter strategy being adapted with current environment, according to the correlation obtained in step S2
Information, carries out the on-line training of admittance model parameter strategy until algorithmic statement, is obtained with expectation by Fuzzy Reinforcement Learning
The change admittance Controlling model being adapted with current environment;
S4:Admittance parameter adjustable strategies after trained convergence in step S3 are applied to become among admittance Controlling model, according to
The moment of face and the feedback speed of joint of mechanical arm that operator applies calculate joint current speed value and sent to joint drive electricity
Machine, so that realizes micro-wound surgical operation mechanical arm actively puts bit function.
2. the mechanical arm Compliant Force Control method according to claim 1 based on Fuzzy Reinforcement Learning, it is characterised in that institute
The Fuzzy Reinforcement Learning stated in step S3 specifically includes following steps:
S31:The moment of face that the motion state of mechanical arm and operator are applied is as state variable, in the opinion of each state variable
Multiple fuzzy sets are divided in the range of domain, corresponding fuzzy rule is set up and provides discrete movement set;
S32:The degree of membership of each state variable is calculated according to current state input, fuzzy division is carried out to state space, calculated
The weights corresponding to fuzzy rule are activated;
S33:Discrete movement value is selected according to current admittance model parameter strategy;
S34:Discrete movement value in step S33 is integrated into final action output valve U and the value is used for into admittance and controls mould
Type.
S35:Current admittance model parameter strategy is updated according to the environment return value currently obtained.
S36:Above-mentioned S32-S35 steps are repeated, until algorithmic statement.
3. the mechanical arm Compliant Force Control method according to claim 1 based on Fuzzy Reinforcement Learning, it is characterised in that also
Comprise the following steps:
S0:The integrated torque sensor in each joint of mechanical arm, the torque sensor be used for detect it is man-machine between contact force
Square.
4. the mechanical arm Compliant Force Control method according to claim 3 based on Fuzzy Reinforcement Learning, it is characterised in that institute
State in step S2:
The gravity compensation model of identified off-line mechanical arm by the way of linear regression, so as to obtain the external force of operator's application
Square.
5. the mechanical arm Compliant Force Control method according to claim 1 based on Fuzzy Reinforcement Learning, it is characterised in that institute
State in step S2:
The motion state of the mechanical arm includes the speed and acceleration of each joint of mechanical arm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710263232.XA CN107053179B (en) | 2017-04-21 | 2017-04-21 | A kind of mechanical arm Compliant Force Control method based on Fuzzy Reinforcement Learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710263232.XA CN107053179B (en) | 2017-04-21 | 2017-04-21 | A kind of mechanical arm Compliant Force Control method based on Fuzzy Reinforcement Learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107053179A true CN107053179A (en) | 2017-08-18 |
CN107053179B CN107053179B (en) | 2019-07-23 |
Family
ID=59600893
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710263232.XA Active CN107053179B (en) | 2017-04-21 | 2017-04-21 | A kind of mechanical arm Compliant Force Control method based on Fuzzy Reinforcement Learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107053179B (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109352648A (en) * | 2018-10-12 | 2019-02-19 | 北京地平线机器人技术研发有限公司 | Control method, device and the electronic equipment of mechanical mechanism |
CN109512518A (en) * | 2018-12-19 | 2019-03-26 | 苏州大学 | A kind of operating robot man-machine coordination motion blur model reference learning control method |
CN109605377A (en) * | 2019-01-21 | 2019-04-12 | 厦门大学 | A kind of joint of robot motion control method and system based on intensified learning |
CN110298854A (en) * | 2019-05-17 | 2019-10-01 | 同济大学 | The snakelike arm co-located method of flight based on online adaptive and monocular vision |
CN111390877A (en) * | 2020-04-23 | 2020-07-10 | 电子科技大学 | Exoskeleton device and exoskeleton admittance variation control method |
CN111881772A (en) * | 2020-07-06 | 2020-11-03 | 上海交通大学 | Multi-mechanical arm cooperative assembly method and system based on deep reinforcement learning |
CN112440276A (en) * | 2019-09-03 | 2021-03-05 | 发那科株式会社 | Control device for robot with arm |
CN112936278A (en) * | 2021-02-07 | 2021-06-11 | 深圳市优必选科技股份有限公司 | Man-machine cooperation control method and device for robot and robot |
CN113568313A (en) * | 2021-09-24 | 2021-10-29 | 南京航空航天大学 | Variable admittance auxiliary large component assembly method and system based on operation intention identification |
CN114344077A (en) * | 2021-12-07 | 2022-04-15 | 华南理工大学 | Flexible upper limb rehabilitation robot system based on SEMG movement intention recognition |
CN114569410A (en) * | 2022-05-06 | 2022-06-03 | 卓道医疗科技(浙江)有限公司 | Control method and device for rehabilitation robot training mode and storage medium |
CN114800513A (en) * | 2022-05-10 | 2022-07-29 | 上海交通大学 | System and method for automatically generating robot shaft hole assembly program based on single-time dragging teaching |
CN114932557A (en) * | 2022-06-24 | 2022-08-23 | 合肥工业大学 | Adaptive admittance control method based on energy consumption under kinematic constraint |
US11449016B2 (en) | 2017-12-22 | 2022-09-20 | Huawei Technologies Co., Ltd. | Action control method and apparatus |
CN116619393A (en) * | 2023-07-24 | 2023-08-22 | 杭州键嘉医疗科技股份有限公司 | Mechanical arm admittance variation control method, device and equipment based on SVM |
CN117084794A (en) * | 2023-10-20 | 2023-11-21 | 北京航空航天大学 | Respiration follow-up control method, device and controller |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101484086A (en) * | 2006-05-19 | 2009-07-15 | 马科外科公司 | Method and apparatus for controlling a haptic device |
JP2009208170A (en) * | 2008-02-29 | 2009-09-17 | Toyota Motor Corp | Power assisting device and its control method |
CN105242533A (en) * | 2015-09-01 | 2016-01-13 | 西北工业大学 | Variable-admittance teleoperation control method with fusion of multi-information |
CN105583824A (en) * | 2016-01-26 | 2016-05-18 | 清华大学 | Force control traction and swinging multi-degree-of-freedom mechanical arm control device and method |
-
2017
- 2017-04-21 CN CN201710263232.XA patent/CN107053179B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101484086A (en) * | 2006-05-19 | 2009-07-15 | 马科外科公司 | Method and apparatus for controlling a haptic device |
JP2009208170A (en) * | 2008-02-29 | 2009-09-17 | Toyota Motor Corp | Power assisting device and its control method |
CN105242533A (en) * | 2015-09-01 | 2016-01-13 | 西北工业大学 | Variable-admittance teleoperation control method with fusion of multi-information |
CN105583824A (en) * | 2016-01-26 | 2016-05-18 | 清华大学 | Force control traction and swinging multi-degree-of-freedom mechanical arm control device and method |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11449016B2 (en) | 2017-12-22 | 2022-09-20 | Huawei Technologies Co., Ltd. | Action control method and apparatus |
CN109352648B (en) * | 2018-10-12 | 2021-03-02 | 北京地平线机器人技术研发有限公司 | Mechanical mechanism control method and device and electronic equipment |
CN109352648A (en) * | 2018-10-12 | 2019-02-19 | 北京地平线机器人技术研发有限公司 | Control method, device and the electronic equipment of mechanical mechanism |
CN109512518A (en) * | 2018-12-19 | 2019-03-26 | 苏州大学 | A kind of operating robot man-machine coordination motion blur model reference learning control method |
CN109605377A (en) * | 2019-01-21 | 2019-04-12 | 厦门大学 | A kind of joint of robot motion control method and system based on intensified learning |
CN110298854A (en) * | 2019-05-17 | 2019-10-01 | 同济大学 | The snakelike arm co-located method of flight based on online adaptive and monocular vision |
CN110298854B (en) * | 2019-05-17 | 2021-05-11 | 同济大学 | Flight snake-shaped arm cooperative positioning method based on online self-adaption and monocular vision |
CN112440276A (en) * | 2019-09-03 | 2021-03-05 | 发那科株式会社 | Control device for robot with arm |
CN111390877B (en) * | 2020-04-23 | 2021-07-13 | 电子科技大学 | Exoskeleton device and exoskeleton admittance variation control method |
CN111390877A (en) * | 2020-04-23 | 2020-07-10 | 电子科技大学 | Exoskeleton device and exoskeleton admittance variation control method |
CN111881772A (en) * | 2020-07-06 | 2020-11-03 | 上海交通大学 | Multi-mechanical arm cooperative assembly method and system based on deep reinforcement learning |
CN111881772B (en) * | 2020-07-06 | 2023-11-07 | 上海交通大学 | Multi-mechanical arm cooperative assembly method and system based on deep reinforcement learning |
CN112936278A (en) * | 2021-02-07 | 2021-06-11 | 深圳市优必选科技股份有限公司 | Man-machine cooperation control method and device for robot and robot |
CN112936278B (en) * | 2021-02-07 | 2022-07-29 | 深圳市优必选科技股份有限公司 | Man-machine cooperation control method and device for robot and robot |
CN113568313A (en) * | 2021-09-24 | 2021-10-29 | 南京航空航天大学 | Variable admittance auxiliary large component assembly method and system based on operation intention identification |
CN114344077A (en) * | 2021-12-07 | 2022-04-15 | 华南理工大学 | Flexible upper limb rehabilitation robot system based on SEMG movement intention recognition |
CN114569410A (en) * | 2022-05-06 | 2022-06-03 | 卓道医疗科技(浙江)有限公司 | Control method and device for rehabilitation robot training mode and storage medium |
CN114800513A (en) * | 2022-05-10 | 2022-07-29 | 上海交通大学 | System and method for automatically generating robot shaft hole assembly program based on single-time dragging teaching |
CN114800513B (en) * | 2022-05-10 | 2024-03-29 | 上海交通大学 | System and method for automatically generating robot shaft hole assembly program based on single dragging teaching |
CN114932557A (en) * | 2022-06-24 | 2022-08-23 | 合肥工业大学 | Adaptive admittance control method based on energy consumption under kinematic constraint |
CN116619393A (en) * | 2023-07-24 | 2023-08-22 | 杭州键嘉医疗科技股份有限公司 | Mechanical arm admittance variation control method, device and equipment based on SVM |
CN116619393B (en) * | 2023-07-24 | 2023-11-14 | 杭州键嘉医疗科技股份有限公司 | Mechanical arm admittance variation control method, device and equipment based on SVM |
CN117084794A (en) * | 2023-10-20 | 2023-11-21 | 北京航空航天大学 | Respiration follow-up control method, device and controller |
CN117084794B (en) * | 2023-10-20 | 2024-02-06 | 北京航空航天大学 | Respiration follow-up control method, device and controller |
Also Published As
Publication number | Publication date |
---|---|
CN107053179B (en) | 2019-07-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107053179B (en) | A kind of mechanical arm Compliant Force Control method based on Fuzzy Reinforcement Learning | |
Li et al. | Asymmetric bimanual control of dual-arm exoskeletons for human-cooperative manipulations | |
US20200282558A1 (en) | System and method for controlling a robot with torque-controllable actuators | |
CN108883533B (en) | Robot control | |
CN108430375B (en) | Robot system and method for back driving the same | |
Dimeas et al. | Fuzzy learning variable admittance control for human-robot cooperation | |
CN107050763B (en) | Novel ankle joint rehabilitation robot and control method thereof | |
Chen et al. | Neural learning enhanced variable admittance control for human–robot collaboration | |
CN108897220B (en) | Self-adaptive stable balance control method and system and biped humanoid robot | |
Garrido et al. | Modular design and control of an upper limb exoskeleton | |
CN113001540A (en) | Man-machine interaction intelligent control method of load-mobile exoskeleton and exoskeleton system | |
CN108422421A (en) | The muscle of skeletal muscle formula robot controls and assembly method | |
CN111390877A (en) | Exoskeleton device and exoskeleton admittance variation control method | |
Wang et al. | A PSO-optimized fuzzy reinforcement learning method for making the minimally invasive surgical arm cleverer | |
Saini et al. | Adaptive control of a master-slave based robotic surgical system with haptic feedback | |
CN110181517B (en) | Double teleoperation training method based on virtual clamp | |
Ai et al. | Master-slave control technology of isomeric surgical robot for minimally invasive surgery | |
CN113500597B (en) | Multi-end teleoperation sliding mode impedance control method based on force translation mechanism | |
Zidane et al. | Trajectory control of a laparoscopic 3-PUU parallel manipulator based on neural network in SIMSCAPE SIMULINK environment | |
CN115755592B (en) | Multi-mode control method for adjusting motion state of three-degree-of-freedom exoskeleton and exoskeleton | |
JP3105694B2 (en) | Manipulator control method | |
Cao et al. | Adaptive sliding mode impedance control in lower limbs rehabilitation robotic | |
Xiang et al. | Learning-based high-precision force estimation and compliant control for small-scale continuum robot | |
CN113547524B (en) | Man-machine interaction control method of upper limb exoskeleton robot | |
Duan et al. | Design of Robot Control System for Knee Joint Replacement Surgery Based on STM32 |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20180611 Address after: 215000 Room 302, 15 building, 8 Jinfeng Road, Huqiu District, Suzhou, Jiangsu. Applicant after: Suzhou health multirobot company limited Address before: 150090 08, room 15, No. 368 Changjiang Road, Nangang District, Harbin, Heilongjiang. Applicant before: HARBIN SIZHERUI INTELLIGENT MEDICAL EQUIPMENT CO., LTD. |
|
GR01 | Patent grant | ||
GR01 | Patent grant |