CN103815991B - Virtual training system and the method for doing evil through another person of dual pathways operation perception - Google Patents
Virtual training system and the method for doing evil through another person of dual pathways operation perception Download PDFInfo
- Publication number
- CN103815991B CN103815991B CN201410079225.0A CN201410079225A CN103815991B CN 103815991 B CN103815991 B CN 103815991B CN 201410079225 A CN201410079225 A CN 201410079225A CN 103815991 B CN103815991 B CN 103815991B
- Authority
- CN
- China
- Prior art keywords
- virtual
- another person
- signal
- doing evil
- control
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
Virtual training system and the method for doing evil through another person of dual pathways operation perception, relates to a kind of EMG-controlling prosthetic hand training field.It uses the EMG-controlling prosthetic hand initial stage to be difficult to maybe cannot complete the problem of crawl task to solve people with disability.It comprises electromyographic electrode, data collecting plate card, virtual reality module and electrostimulator.Electromyographic electrode is used for the filtering of human body electromyographic signal and amplification, data collecting plate card is used for the A/D conversion of electromyographic electrode output signal, virtual reality module includes virtual scene display, people's hand control joint rotation angle, joint rotation angle position measurement display, capture scene selects and instruction, electromyographic signal classification and decoding controls, joint motor moment size controls, virtually in crawl process do evil through another person and capture thing collision checking function, capture joint moment and measure display, and electrostimulator then carries out the output of corresponding electrical stimulation signal according to the virtual moment values of contact object of doing evil through another person.The present invention is applicable to operation training when people with disability has just brought into use EMG-controlling prosthetic hand.
Description
Technical field
The present invention relates to a kind of EMG-controlling prosthetic hand training field, especially a kind of EMG-controlling prosthetic hand training devices of combined with virtual reality and method.
Background technology
Virtual reality (Virtual Reality) is a kind of computer technology with the experiencing virtual world of can setting up, it by computer technology generate one true to nature, can based on the immersion interactive environment of computing information, thus be widely used in multiple fields such as machine-building, biologic medical, urban planning, military simulation and Aero-Space, and create huge economic benefit and social benefit.The current training system based on virtual reality concentrates on the recovery training of apoplexy, brain injury, parkinson and surgical site infections mostly.Francesca Cordella uses kinect to carry out the record of staff movement locus in crawl process as the motion capture device of staff, rehabilitation assessment is carried out in joint trajectories contrast patient and healthy staff being completed same action, and utilizes reality environment to carry out corresponding rehabilitation training.
EMG-controlling prosthetic hand is a kind of muscle signal of telecommunication (Electromyography utilizing human body forearm, EMG) as information source, by certain information decoding strategy, the control of people intention is converted into the rehabilitation equipment that motion control information completes multiple crawl task, the such as Sensorhandspeed of OttoBock company, Michelangelo Hand etc.
For forearm defect or wrist from disconnected people with disability, the symptom such as atrophy, necrosis is often there is in residual arm muscle after amputation, and along with user is for the raising gradually of performance requirement of doing evil through another person, the degree of freedom of doing evil through another person is also more, therefore adjoint control difficulty also can increase, comparatively difficult when these all cause people with disability just to bring into use EMG-controlling prosthetic hand, occur that operation is unskilled, or the situation of crawl task cannot be completed at all.Therefore training of doing evil through another person just seems particularly important, but has that complexity is high, maintenance cost is high, exemplary difference is demonstrated in action based on the training of truly doing evil through another person and the shortcomings such as statistics training result difficulty.In order to improve training efficiency, reduce costs, a kind of virtual reality system that patient can be helped to complete pattern drill seems very necessary.
Summary of the invention
The present invention uses the EMG-controlling prosthetic hand initial stage to be difficult to maybe cannot complete the problem of crawl task to solve people with disability, thus provides a kind of dual pathways to operate virtual training system and the method for doing evil through another person of perception.
The virtual training system of doing evil through another person of dual pathways operation perception, it comprises electromyographic electrode, data collecting plate card, PC and electrostimulator; PC inside embeds virtual reality module;
The electromyographic signal collected for gathering the electromyographic signal of human body, and is carried out filtering and amplification by electromyographic electrode, and filtering and the signal after amplifying are sent to data collecting plate card;
Data collecting plate card is used for carrying out analog digital conversion by through electromyographic electrode filtering and the electromyographic signal after amplifying, and signal is sent to the virtual reality module in PC;
The electricity irritation control instruction that electrostimulator sends for the virtual reality module received in PC, and produce electrical stimulation signal;
Virtual reality module comprises virtual scene display module, staff control module, virtual location sensor assembly, captures scene selection and indicating module, myoelectricity decoding and automatic control module, motor torque output control module, collision detection module and virtual moment sensor assembly;
Virtual scene display module captures calling and replacing of object for providing different, carries out virtual doing evil through another person and to simulate with the crawl of corresponding object;
Staff control module is used for providing virtual joint position manual control signal of doing evil through another person, crawl object replaces control signal, electromyographic signal collection trains enabling signal and staff myoelectricity control signal coding controls enabling signal automatically;
Virtual location sensor assembly is used for detecting virtual joint rotation angle of doing evil through another person;
Crawl scene is selected and indicating module is used for selecting virtual crawl scene of doing evil through another person and indicating;
Myoelectricity decoding and automatic control module are used for using grader myoelectricity control signal to be classified to the electromyographic signal that data collecting plate card sends, and decode for the coded signal produced, according to the joint moment of current manual input, finger kinetics equation joint rotation angle numerical solution array mode is adopted to control virtual joint action of doing evil through another person in real time;
Motor torque output control module is used for virtual control motor output torque control signal of doing evil through another person, and dynamic numerical solution of doing evil through another person solves assignment;
Collision detection module is used for carrying out the collision detection of different objects crawl for virtual doing evil through another person;
Virtual moment sensor assembly is used for collision rift being detected, captures the value produced after collision occurs obtain corresponding moment sensor values according to grasp force Epidemiological Analysis or reality of doing evil through another person.
The virtual training method of doing evil through another person of dual pathways operation perception, it comprises the following steps:
Adopt electromyographic electrode for gathering the electromyographic signal of human body, and the electromyographic signal collected is carried out filtering and amplification, and filtering and the signal after amplifying are sent to the step of data collecting plate card;
Adopt data collecting plate card to be used for carrying out analog digital conversion by through electromyographic electrode filtering and the electromyographic signal after amplifying, and signal is sent to the step of the virtual reality module in PC;
Adopt the electricity irritation control instruction that electrostimulator sends for the virtual reality module received in PC, and produce the step of electrical stimulation signal;
The signal processing method of virtual reality module comprises:
Capturing calling and replacing of object for providing different, carrying out the virtual virtual scene display step simulated with the crawl of corresponding object of doing evil through another person;
For providing virtual joint position manual control signal of doing evil through another person, capturing the staff rate-determining steps that object replacement control signal, electromyographic signal collection training enabling signal and staff myoelectricity control signal coding control enabling signal automatically;
For the step detected virtual joint rotation angle of doing evil through another person;
For carrying out the step of Action Selection and instruction to virtual crawl scene of doing evil through another person:
Electromyographic signal for sending data collecting plate card uses grader myoelectricity control signal to be classified, and decode for the coded signal produced, according to the joint moment of current manual input, finger kinetics equation joint rotation angle numerical solution array mode is adopted to control the myoelectricity decoding of virtual joint action of doing evil through another person and automatic rate-determining steps in real time;
Control for exporting to the motor torque of virtual control motor output torque control signal of doing evil through another person, and dynamic numerical solution of doing evil through another person solves assignment procedure;
For the collision detection step for the virtual collision detection of carrying out different objects crawl of doing evil through another person;
For collision rift being detected, capturing according to grasp force Epidemiological Analysis or reality of doing evil through another person the acquisition step that the value produced after collision occurs obtains corresponding moment sensor values.
Myoelectricity decoding and the signal processing of automatic control module are: myoelectricity is decoded and automatic control module is first curved by staff, loosen, the electromyographic signal of stretching is classified, the time continued according to electromyographic signal is again segmented electromyographic signal, according to the length of persistent period and the kind of muscle, electromyographic signal is divided into and shortly stretches signal, short exor signal, long stretch signal and long exor signal; Finally the coded signal of muscle movement sequence is decoded, use the corresponding 6 kinds of different grasping movement of 6 kinds of different muscle movement combined sequence.
Staff control module comprises the virtual action module of doing evil through another person of Non-follow control, virtual crawl scene switches selection module, collection is trained and automatically control beginning control knob;
The virtual action module of doing evil through another person of Non-follow control comprises the input control starting manual control button and corresponding joint rotation angle;
Virtual crawl scene switching is selected module to comprise different typical case and is captured corresponding object importing button; Collection training comprises gathering with automatic control beginning control knob to be trained and automatically controls start button;
Virtual motor torque adjustment module of doing evil through another person comprises virtual motor torque manual input control of doing evil through another person and solves assignment module with dynamic numerical solution of doing evil through another person;
Virtual motor torque input control of doing evil through another person comprises the input control of virtual joint motor moment of doing evil through another person;
Dynamic numerical solution of doing evil through another person solves assignment module by dynamic analysis of doing evil through another person, set up the relation of do evil through another person motor torque and joint rotation angle, angular velocity, angular acceleration, and try to achieve the numerical solution array of joint rotation angle, the assignment of joint rotation angle in capturing according to joint rotation angle numerical solution array, shows the dynamic crawl process of corresponding joint motor moment in virtual scene.
Collision detection module is used for being specially for virtual process of doing evil through another person the collision detection of carrying out different objects crawl:
Set up the step of DualSceneCollide class: for calling collision detection class and instantiation collision detection pair;
Calling the definition of SoPath class, to be grabbed object be static searching route, and then defining 11 dactylus is respectively 11 News Search paths, and sets up the right step of collision detection with static searching route respectively;
Call the collision detection that SoMaterial class in Open Inventor defines the different finger of the point of impingement labelling five of five kinds of different colours, the checkCollision function calling the SoDualSceneCollider apoplexy due to endogenous wind that Open Inventor carries again to carrying out collision detection, and returns the step of the coordinate figure of collision origination point to often pair of collision detection.
The moment values that stimulus frequency and the virtual moment sensor assembly of electrostimulator obtain is directly proportional.
Virtual scene display module is that immersive VR captures scene module, for the switching carrying out virtual scene of doing evil through another person, typicality captures object scene, and the setting of dummy object texture color and illumination.
The virtual training system of doing evil through another person of dual pathways operation perception of the present invention is by the collection to human body EMG control signal, classification, automatic identification human body EMG coding-control intention carries out the crawl of respective type object, and merged virtual joint position and torque sensor, joint driven torque control and collision detection can be carried out simultaneously, add the sense of reality of training system of doing evil through another person, enable trainer immerse wherein to carry out to do evil through another person the training of product, improve the efficiency of training of doing evil through another person, good help is had for the electromyographic signal and control proficiency level that improve the trainer that does evil through another person, the self-confidence of trainer can also be improved simultaneously.Successfully solving people with disability uses the EMG-controlling prosthetic hand initial stage to be difficult to maybe cannot complete the problem of crawl task.
Accompanying drawing explanation
Fig. 1 is structural principle block diagram of the present invention;
Fig. 2 is the two-way man-machine interaction schematic diagram of the present invention.
Detailed description of the invention
Detailed description of the invention one, composition graphs 1 illustrate this detailed description of the invention, the virtual training system of doing evil through another person of dual pathways operation perception, and it comprises electromyographic electrode, data collecting plate card, PC and electrostimulator; PC inside embeds virtual reality module;
The electromyographic signal collected for gathering the electromyographic signal of human body, and is carried out filtering and amplification by electromyographic electrode, and filtering and the signal after amplifying are sent to data collecting plate card;
Data collecting plate card is used for carrying out analog digital conversion by through electromyographic electrode filtering and the electromyographic signal after amplifying, and signal is sent to the virtual reality module in PC;
The electricity irritation control instruction that electrostimulator sends for the virtual reality module received in PC, and produce electrical stimulation signal;
Virtual reality module comprises virtual scene display module, staff control module, virtual location sensor assembly, captures scene selection and indicating module, myoelectricity decoding and automatic control module, motor torque output control module, collision detection module and virtual moment sensor assembly;
Virtual scene display module captures calling and replacing of object for providing different, carries out virtual doing evil through another person and to simulate with the crawl of corresponding object;
Staff control module is used for providing virtual joint position manual control signal of doing evil through another person, crawl object replaces control signal, electromyographic signal collection trains enabling signal and staff myoelectricity control signal coding controls enabling signal automatically;
Virtual location sensor assembly is used for detecting virtual joint rotation angle of doing evil through another person;
Crawl scene is selected and indicating module is used for selecting virtual crawl scene of doing evil through another person and indicating;
Myoelectricity decoding and automatic control module are used for using grader myoelectricity control signal to be classified to the electromyographic signal that data collecting plate card sends, and decode for the coded signal produced, according to the joint moment of current manual input, finger kinetics equation joint rotation angle numerical solution array mode is adopted to control virtual joint action of doing evil through another person in real time;
Motor torque output control module is used for virtual control motor output torque control signal of doing evil through another person, and dynamic numerical solution of doing evil through another person solves assignment;
Collision detection module is used for carrying out the collision detection of different objects crawl for virtual doing evil through another person;
Virtual moment sensor assembly is used for collision rift being detected, captures the value produced after collision occurs obtain corresponding moment sensor values according to grasp force Epidemiological Analysis or reality of doing evil through another person.
The virtual training method of doing evil through another person of detailed description of the invention two, dual pathways operation perception, it comprises the following steps:
Adopt electromyographic electrode for gathering the electromyographic signal of human body, and the electromyographic signal collected is carried out filtering and amplification, and filtering and the signal after amplifying are sent to the step of data collecting plate card;
Adopt data collecting plate card to be used for carrying out analog digital conversion by through electromyographic electrode filtering and the electromyographic signal after amplifying, and signal is sent to the step of the virtual reality module in PC;
Adopt the electricity irritation control instruction that electrostimulator sends for the virtual reality module received in PC, and produce the step of electrical stimulation signal;
The action executing method of virtual reality module comprises:
Capturing calling and replacing of object for providing different, carrying out the virtual virtual scene display step simulated with the crawl of corresponding object of doing evil through another person;
For providing virtual joint position manual control signal of doing evil through another person, capturing the staff rate-determining steps that object replacement control signal, electromyographic signal collection training enabling signal and staff myoelectricity control signal coding control enabling signal automatically;
For the step detected virtual joint rotation angle of doing evil through another person;
For to the virtual step of carrying out Action Selection and instruction of doing evil through another person:
Electromyographic signal for sending data collecting plate card uses grader myoelectricity control signal to be classified, and decode for the coded signal produced, according to the joint moment of current manual input, finger kinetics equation joint rotation angle numerical solution array mode is adopted to control the myoelectricity decoding of virtual joint action of doing evil through another person and automatic rate-determining steps in real time;
Control for exporting to the motor torque of virtual control motor output torque control signal of doing evil through another person, and dynamic numerical solution of doing evil through another person solves assignment procedure;
For the collision detection step for the virtual collision detection of carrying out different objects crawl of doing evil through another person;
For collision rift being detected, capturing according to grasp force Epidemiological Analysis or reality of doing evil through another person the acquisition step that the value produced after collision occurs obtains corresponding moment sensor values.
The signal processing of myoelectricity decoding and automatic control module is: the time that myoelectricity decoding and automatic control module continue according to electromyographic signal is segmented electromyographic signal, according to the length of persistent period and the kind of muscle, electromyographic signal is divided into and shortly stretches signal, short exor signal, long stretch signal and long exor signal; Finally muscle movement sequence is encoded, use the corresponding 6 kinds of different grasping movement of 6 kinds of different muscle movement combined sequence.
Staff control module comprises the virtual action module of doing evil through another person of Non-follow control, virtual crawl scene selects module, gather training and automatic control starts control knob;
The virtual action module of doing evil through another person of Non-follow control comprises the input control starting manual control button and corresponding joint rotation angle;
Virtual crawl scene switching is selected module to comprise different typical case and is captured corresponding object importing button; Collection training comprises gathering with automatic control beginning control knob to be trained and automatically controls start button;
Virtual motor torque adjustment module of doing evil through another person comprises virtual motor torque manual input control of doing evil through another person and solves assignment module with dynamic numerical solution of doing evil through another person;
Virtual motor torque input control of doing evil through another person comprises the input control of virtual joint motor moment of doing evil through another person;
Dynamic numerical solution of doing evil through another person solves assignment module by dynamic analysis of doing evil through another person, set up the relation of do evil through another person motor torque and joint rotation angle, angular velocity, angular acceleration, and try to achieve the numerical solution array of joint rotation angle, the assignment of joint rotation angle in capturing according to joint rotation angle numerical solution array, shows the dynamic crawl process of corresponding joint motor moment in virtual scene.
Collision detection module is used for being specially for virtual process of doing evil through another person the collision detection of carrying out different objects crawl:
Set up the step of DualSceneCollide class: for calling collision detection class and instantiation collision detection pair;
Calling the definition of SoPath class, to be grabbed object be static searching route, and then defining 11 dactylus is respectively 11 News Search paths, and sets up the right step of collision detection with static searching route respectively;
Call the collision detection that SoMaterial class in Open Inventor defines the different finger of the point of impingement labelling five of five kinds of different colours, the checkCollision function calling the SoDualSceneCollider apoplexy due to endogenous wind that Open Inventor carries again to carrying out collision detection, and returns the step of the coordinate figure of collision origination point to often pair of collision detection.
The moment values that stimulus frequency and the virtual moment sensor assembly of electrostimulator obtain is directly proportional.
Virtual scene display module is that immersive VR captures scene module, for the switching carrying out virtual scene of doing evil through another person, typicality captures object scene, and the setting of dummy object texture color and illumination.
Operation principle: electromyographic electrode is used for electromyographic signal filter and amplification, is sent to data collecting plate card;
The electromyographic signal received is carried out A/D conversion by data collecting plate card, and the electromyographic signal after conversion is passed to PC;
PC is under virtual reality operation panel clicks the prerequisite of sample train, accept the electromyographic signal that staff is curved, loosen, stretch, and classified by the electromyographic signal of grader by 3 kinds of patterns, then the time continued according to electromyographic signal is segmented electromyographic signal, according to length and the kind of persistent period, electromyographic signal is divided into shortly stretches signal, short exor signal, long stretches signal and long exor signal.Click the automatic control knob of grasp mode and the operation interface that will do, staff carries out the coding of bending, stretching action according to set in advance 6 kinds of different grasping movement, PC is according to the coded command of now staff action and the classification results gathering training early stage, carry out classification and the decoding of control instruction, the final crawl task realizing corresponding instruction automatically, in crawl process, joint rotation angle carries out real-time assignment, to produce dynamical motion effect true to nature according to virtual numerical solution array of doing evil through another person finger kinetics equation;
In course of action, carry out real time collision detection, the process of collision detection is: (1) sets up DualSceneCollide class, for calling collision detection class and instantiation collision detection pair simultaneously.
(2) calling the definition of SoPath class, to be grabbed object be static searching route, and then defining 11 dactylus is respectively 11 News Search paths, and sets up collision detection pair with static searching route respectively.
(3) because a new generation does evil through another person as the finger design that is coupled, therefore under usual state, the contact point of each finger and dummy object only has one, and therefore the collision centering of each finger only can have a collision to colliding simultaneously.Call the collision detection that SoMaterial class in OpenInventor defines the different finger of the point of impingement labelling five of five kinds of different colours, the checkCollision function calling the SoDualSceneCollider apoplexy due to endogenous wind that Open Inventor carries again to carrying out collision detection, and returns the coordinate figure of collision origination point to often pair of collision detection.In the process, by element value assignment corresponding for the corner numerical solution array of current dynamics calculation storage of doing evil through another person to virtual location sensor, and show in real time, after collision occurs, the angle value of the previous time point after record collision occurs, in addition, the electromyographic signal numerical value that corresponding electromyographic electrode gathers, adopt the staff encode control commands of coding-control, information of classifier also can show in real time, after collision occurs, each joint of finger stops action, now virtual torque sensor according to grasp force Epidemiological Analysis or do evil through another person reality capture collision occur after produce worth go out corresponding moment sensor values, and show in real time, as shown in Figure 2, electrostimulator carries out the adjustment of electrical stimulation signal according to the value of virtual torque sensor.
When selecting the manual control button of virtual reality interface, now manually given virtual corresponding joint rotation angle value of doing evil through another person carries out the control of current joint rotation angle of doing evil through another person, and the virtual joint rotation angle value assignment of doing evil through another person of manual setting to virtual location sensor, show in real time, 6 kinds of switching push buttons capturing scene are for providing different crawl simulated scenario conversions, the virtual LED lantern festival that simultaneously often kind of scene selectes rear correspondence is lit, and instruction captures scene handover success.Gather training and the unlatching of automatic control mode, all need control panel click corresponding release the button just can, the LED gathering training is used to indicate staff correspondence coded command sample collection and completes.The moment of motor can manually be arranged, and solves according to the kinetics equation that staff settings carry out corresponding joint moment value and calculate and control, and corresponding joint moment value also can show in real time.
Virtual reality scenario display module selected element light source, as the environment light source of virtual training system, can not produce interference to the gauge point of collision detection again like this while bringing display effect more true to nature.Use SoTexture2 node in Open Inventor to complete the importing of texture, enhance verity and the feeling of immersion of virtual reality system.
Claims (7)
1. the virtual training system of doing evil through another person of dual pathways operation perception, is characterized in that: it comprises electromyographic electrode, data collecting plate card, PC and electrostimulator; PC inside embeds virtual reality module;
The electromyographic signal collected for gathering the electromyographic signal of human body, and is carried out filtering and amplification by electromyographic electrode, and filtering and the signal after amplifying are sent to data collecting plate card;
Data collecting plate card is used for carrying out analog digital conversion by through electromyographic electrode filtering and the electromyographic signal after amplifying, and signal is sent to the virtual reality module in PC;
The electricity irritation control instruction that electrostimulator sends for the virtual reality module received in PC, and produce electrical stimulation signal;
Virtual reality module comprises virtual scene display module, staff control module, virtual location sensor assembly, captures scene selection and indicating module, myoelectricity decoding and automatic control module, motor torque output control module, collision detection module and virtual moment sensor assembly;
Virtual scene display module captures calling and replacing of object for providing different, carries out virtual doing evil through another person and to simulate with the crawl of corresponding object;
Staff control module is used for providing virtual joint position manual control signal of doing evil through another person, crawl object replaces control signal, electromyographic signal collection trains enabling signal and staff myoelectricity control signal coding controls enabling signal automatically;
Virtual location sensor assembly is used for detecting virtual joint rotation angle of doing evil through another person;
Crawl scene is selected and indicating module is used for selecting virtual crawl scene of doing evil through another person and indicating;
Myoelectricity decoding and automatic control module are used for using grader myoelectricity control signal to be classified to the electromyographic signal that data collecting plate card sends, and decode for the coded signal produced, according to the joint moment of current manual input, finger kinetics equation joint rotation angle numerical solution array mode is adopted to control virtual joint action of doing evil through another person in real time;
Motor torque output control module is used for virtual control motor output torque control signal of doing evil through another person, and dynamic numerical solution of doing evil through another person solves assignment;
Collision detection module is used for carrying out the collision detection of different objects crawl for virtual doing evil through another person;
Virtual moment sensor assembly is used for collision rift being detected, captures the value produced after collision occurs obtain corresponding moment sensor values according to grasp force Epidemiological Analysis or reality of doing evil through another person.
2., based on the training method of the virtual training system of doing evil through another person of dual pathways operation perception according to claim 1, it is characterized in that: it comprises the following steps:
Adopt electromyographic electrode for gathering the electromyographic signal of human body, and the electromyographic signal collected is carried out filtering and amplification, and filtering and the signal after amplifying are sent to the step of data collecting plate card;
Adopt data collecting plate card to be used for carrying out analog digital conversion by through electromyographic electrode filtering and the electromyographic signal after amplifying, and signal is sent to the step of the virtual reality module in PC;
Adopt the electricity irritation control instruction that electrostimulator sends for the virtual reality module received in PC, and produce the step of electrical stimulation signal;
The signal processing method of virtual reality module comprises:
Capturing calling and replacing of object for providing different, carrying out the virtual virtual scene display step simulated with the crawl of corresponding object of doing evil through another person;
For providing virtual joint position manual control signal of doing evil through another person, capturing the staff rate-determining steps that object replacement control signal, electromyographic signal collection training enabling signal and staff myoelectricity control signal coding control enabling signal automatically;
For the step detected virtual joint rotation angle of doing evil through another person;
For carrying out the step capturing scene selection and instruction to virtual crawl scene of doing evil through another person:
Electromyographic signal for sending data collecting plate card uses grader myoelectricity control signal to be classified, and decode for the coded signal produced, according to the joint moment of current manual input, finger kinetics equation joint rotation angle numerical solution array mode is adopted to control the myoelectricity decoding of virtual joint action of doing evil through another person and automatic rate-determining steps in real time;
Control for exporting to the motor torque of virtual control motor output torque control signal of doing evil through another person, and dynamic numerical solution of doing evil through another person solves assignment procedure;
For the collision detection step for the virtual collision detection of carrying out different objects crawl of doing evil through another person;
For collision rift being detected, capturing according to grasp force Epidemiological Analysis or reality of doing evil through another person the acquisition step that the value produced after collision occurs obtains corresponding moment sensor values.
3. the virtual training method of doing evil through another person of dual pathways operation perception according to claim 2, it is characterized in that myoelectricity decoding and the signal processing of automatic control module are: myoelectricity is decoded and automatic control module is first curved by staff, loosen, the electromyographic signal of stretching is classified, the time continued according to electromyographic signal is again segmented electromyographic signal, according to the length of persistent period and the kind of muscle, electromyographic signal is divided into and shortly stretches signal, short exor signal, long stretch signal and long exor signal; Finally the coded signal of muscle movement sequence is decoded, use the corresponding 6 kinds of different grasping movement of 6 kinds of different muscle movement combined sequence.
4. the virtual training method of doing evil through another person of dual pathways operation perception according to claim 2, is characterized in that staff control module comprises the virtual action module of doing evil through another person of Non-follow control, virtual crawl scene switches selection module, gathers training and automatically control beginning control knob;
The virtual action module of doing evil through another person of Non-follow control comprises the input control starting manual control button and corresponding joint rotation angle;
Virtual crawl scene switching is selected module to comprise different typical case and is captured corresponding object importing button; Collection training comprises gathering with automatic control beginning control knob to be trained and automatically controls start button.
5. the virtual training method of doing evil through another person of dual pathways operation perception according to claim 2, is characterized in that collision detection module is for being specially for virtual process of doing evil through another person the collision detection of carrying out different objects crawl:
Set up the step of DualSceneCollide class: for calling collision detection class and instantiation collision detection pair;
Calling the definition of SoPath class, to be grabbed object be static searching route, and then defining 11 dactylus is respectively 11 News Search paths, and sets up the right step of collision detection with static searching route respectively;
Call the collision detection that SoMaterial class in Open Inventor defines the different finger of the point of impingement labelling five of five kinds of different colours, the checkCollision function calling the SoDualSceneCollider apoplexy due to endogenous wind that Open Inventor carries again to carrying out collision detection, and returns the step of the coordinate figure of collision origination point to often pair of collision detection.
6. the virtual training method of doing evil through another person of dual pathways operation perception according to claim 2, is characterized in that the moment values that the stimulus frequency of electrostimulator and virtual moment sensor assembly obtain is directly proportional.
7. the virtual training method of doing evil through another person of dual pathways operation perception according to claim 2, it is characterized in that virtual scene display module is that immersive VR captures scene module, for the switching carrying out virtual scene of doing evil through another person, typicality captures object scene, and the setting of dummy object texture color and illumination.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410079225.0A CN103815991B (en) | 2014-03-06 | 2014-03-06 | Virtual training system and the method for doing evil through another person of dual pathways operation perception |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410079225.0A CN103815991B (en) | 2014-03-06 | 2014-03-06 | Virtual training system and the method for doing evil through another person of dual pathways operation perception |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103815991A CN103815991A (en) | 2014-05-28 |
| CN103815991B true CN103815991B (en) | 2015-10-28 |
Family
ID=50751486
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410079225.0A Active CN103815991B (en) | 2014-03-06 | 2014-03-06 | Virtual training system and the method for doing evil through another person of dual pathways operation perception |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103815991B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105943206A (en) * | 2016-06-01 | 2016-09-21 | 上海师范大学 | Prosthetic hand control method based on MYO armlet |
| CN106420124B (en) * | 2016-10-10 | 2018-08-24 | 上海电机学院 | A kind of myoelectricity control virtual robot is done evil through another person the method for analogue system |
| CN106530926B (en) * | 2016-11-29 | 2019-03-05 | 东南大学 | Virtual artificial hand training platform and its training method based on Myo armband and Eye-controlling focus |
| CN106974749A (en) * | 2017-04-25 | 2017-07-25 | 北京展翼计划科技发展有限公司 | Bionical prosthetic hand and device based on 3D printing |
| CN107411857B (en) * | 2017-09-21 | 2019-06-11 | 哈尔滨工业大学 | A kind of control system of distribution wrist hands prostheses system |
| CN113771081A (en) * | 2021-07-06 | 2021-12-10 | 清华大学 | Physical-based virtual human hand automatic grabbing method and device |
| CN115153983B (en) * | 2022-06-15 | 2024-04-12 | 哈尔滨工业大学 | Smart artificial hand control system, device, method and storage medium based on machine vision and eye tracking |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006086504A2 (en) * | 2005-02-09 | 2006-08-17 | Alfred E. Mann Institute For Biomedical Engineering At The University Of Southern California | Method and system for training adaptive control of limb movement |
| CN101183242A (en) * | 2007-11-16 | 2008-05-21 | 哈尔滨工业大学 | Embedded multi-freedom degree muscle-electric artificial hand control system based on double digital signal processor |
| CN101305939A (en) * | 2008-05-26 | 2008-11-19 | 哈尔滨工业大学 | Electrical stimulator for sensation feedback of apery myoelectricity artificial hand |
| CN101667346A (en) * | 2009-09-18 | 2010-03-10 | 国家康复辅具研究中心 | Rehabilitation training system of amputation upper limb based on virtual reality |
| CN202235784U (en) * | 2011-08-25 | 2012-05-30 | 东南大学 | Electromyography prosthetic hand power system with tactile feedback |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003041790A2 (en) * | 2001-11-10 | 2003-05-22 | Arizona Board Of Regents | Direct cortical control of 3d neuroprosthetic devices |
-
2014
- 2014-03-06 CN CN201410079225.0A patent/CN103815991B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006086504A2 (en) * | 2005-02-09 | 2006-08-17 | Alfred E. Mann Institute For Biomedical Engineering At The University Of Southern California | Method and system for training adaptive control of limb movement |
| CN101183242A (en) * | 2007-11-16 | 2008-05-21 | 哈尔滨工业大学 | Embedded multi-freedom degree muscle-electric artificial hand control system based on double digital signal processor |
| CN101305939A (en) * | 2008-05-26 | 2008-11-19 | 哈尔滨工业大学 | Electrical stimulator for sensation feedback of apery myoelectricity artificial hand |
| CN101667346A (en) * | 2009-09-18 | 2010-03-10 | 国家康复辅具研究中心 | Rehabilitation training system of amputation upper limb based on virtual reality |
| CN202235784U (en) * | 2011-08-25 | 2012-05-30 | 东南大学 | Electromyography prosthetic hand power system with tactile feedback |
Non-Patent Citations (1)
| Title |
|---|
| 杨大鹏等.三自由度仿人型假手及其肌电控制策略.《哈尔滨工程大学学报》.2009,第30卷(第7期), * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103815991A (en) | 2014-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103815991B (en) | Virtual training system and the method for doing evil through another person of dual pathways operation perception | |
| CN103892945B (en) | Myoelectric limb control system | |
| CN101667346B (en) | Rehabilitation training system of amputation upper limb based on virtual reality | |
| CN107397649A (en) | A kind of upper limbs exoskeleton rehabilitation robot control method based on radial base neural net | |
| EP3235540A1 (en) | Auxiliary device for training and auxiliary method for training | |
| CN106779045A (en) | Rehabilitation training robot system and its application method based on virtual scene interaction | |
| CN102499797B (en) | Artificial limb control method and system | |
| CN102831380A (en) | Body action identification method and system based on depth image induction | |
| CN104107134A (en) | Myoelectricity feedback based upper limb training method and system | |
| CN103358311B (en) | Robot control method and system based on bioelectricity signal decoding | |
| CN109215774B (en) | Real-time remote pulse-taking system and working method thereof | |
| Curcie et al. | Biomimetic finger control by filtering of distributed forelimb pressures | |
| CN105708587A (en) | Lower-limb exoskeleton training method and system triggered by brain-computer interface under motion imagination pattern | |
| CN107281710A (en) | A kind of method of remedial action error | |
| Song et al. | Activities of daily living-based rehabilitation system for arm and hand motor function retraining after stroke | |
| CN108022248A (en) | A kind of lower limb gait rehabilitation assessment system of view-based access control model collecting device | |
| CN106491251A (en) | One kind is based on non-intrusion type brain-computer interface robotic arm control system and its control method | |
| CN103631381A (en) | Man-machine interaction system and man-machine interaction method for electric transmission line patrolling simulation system | |
| Suhaimi et al. | Analysis of EMG-based muscles activity for stroke rehabilitation | |
| CN109223441A (en) | A kind of human upper limb rehabilitation training and movement auxiliary system based on Kinect sensor | |
| CN116386414A (en) | Digital mirror image-based ergonomic adjustment line training system and method | |
| CN109446957B (en) | EMG signal identification method | |
| Silva et al. | Teleoperation training environment for new users of electric powered wheelchairs based on multiple driving methods | |
| Kwon et al. | On muscle selection for EMG based decoding of dexterous, in-hand manipulation motions | |
| Geng et al. | Recognition of combined arm motions using support vector machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |