CN105467926B - A kind of motion control method and device, artificial intelligence equipment - Google Patents
A kind of motion control method and device, artificial intelligence equipment Download PDFInfo
- Publication number
- CN105467926B CN105467926B CN201410459517.7A CN201410459517A CN105467926B CN 105467926 B CN105467926 B CN 105467926B CN 201410459517 A CN201410459517 A CN 201410459517A CN 105467926 B CN105467926 B CN 105467926B
- Authority
- CN
- China
- Prior art keywords
- information
- movable information
- target
- movement
- movable
- 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
Landscapes
- Manipulator (AREA)
- Toys (AREA)
Abstract
A kind of motion control method of present invention offer and device, artificial intelligence equipment, wherein method include：Multiple movable informations of sensor acquisition are obtained, multiple movable information corresponds to multiple individual of sample respectively；According to the multiple movable information, the target movable information for optimizing corresponding movement is determined；The target movable information is sent to artificial intelligence equipment, so that the artificial intelligence equipment executes corresponding movement according to the target movable information.The present invention improves the optimization efficiency of artificial intelligence equipment.
Description
Technical field
The present invention relates to smart machine technology, more particularly to a kind of motion control method and device, artificial intelligence equipment.
Background technology
With the continuous development of science and technology, currently the mankind can be replaced to hold by artificial intelligence equipment such as robot
Row some tasks, such as robot can cook either robot can also do washing clothes or robot can also be in the factory
It does some and has dangerous action for the mankind, to avoid the danger of the mankind.Therefore, robot is to human lives or work
Tool is very helpful.
But in current technology, the movement control mode of robot is relatively easy, such as the embedded fixation in robot
Execution program, the controller of robot interior after the exogenous data for receiving sensor acquisition, by the gathered data with hold
Correlated condition in line program compares, and the action of corresponding execution is obtained by simple logic judgment, and the action
Execution is the action parameter that foundation is stored in advance in robot, such as the walking stride value of foundation storage to walk.It can
It is that robot differs according to the action that the parameter executes and surely obtains preferable movement effects, such as since walking stride value does not conform to
Suitable so that robot is carrying out of the task can not be completed well.At this time can by change robot internal processes and
Corresponding action parameter optimizes its action so that and the movement of robot itself can improve and improve, it is obvious that
, this mode needs to carry out modification of program and debugging, it is possible to which multiple modification and debugging can be only achieved a preferable fortune
Dynamic effect, very complicated so that the action optimization efficiency of robot is very low, influences business execution.
Invention content
In view of this, a kind of motion control method of present invention offer and device, artificial intelligence equipment, to improve artificial intelligence
The optimization efficiency of equipment.
Specifically, the present invention is achieved through the following technical solutions：
In a first aspect, a kind of motion control method is provided, including：
Multiple movable informations of sensor acquisition are obtained, multiple movable information corresponds to multiple individual of sample respectively；
According to the multiple movable information, the target movable information for optimizing corresponding movement is determined；
The target movable information is sent to artificial intelligence equipment, so that the artificial intelligence equipment is according to the mesh
It marks movable information and executes corresponding movement.
Second aspect provides a kind of motion control method, including：
Target movable information is obtained, the target movable information is multiple fortune that motion control device is acquired according to sensor
Dynamic information obtains, and the multiple movable information corresponds to multiple individual of sample respectively；
Corresponding movement is executed according to the target movable information, it is the multiple to be optimized by the target movable information
The corresponding movement of movable information.
The third aspect provides a kind of motion control device, including：
Information receiving module, for multiple movable informations of receiving sensor acquisition, the multiple movable information is right respectively
Answer multiple individual of sample；
Message processing module, for according to the multiple movable information, determining the target movement for optimizing corresponding movement
Information；
Information sending module, for the target movable information to be sent to artificial intelligence equipment, so that described artificial
Smart machine executes corresponding movement according to the target movable information.
Fourth aspect provides a kind of artificial intelligence equipment, including：
Data obtaining module, for obtaining target movable information, the target movable information be motion control device according to
Multiple movable informations of sensor acquisition obtain, and the multiple movable information corresponds to multiple individual of sample respectively；
Motor execution module, for executing corresponding movement according to the target movable information, to be transported by the target
The corresponding movement of the multiple movable information of dynamic Advance data quality.
Motion control method and device provided in an embodiment of the present invention, artificial intelligence equipment, are acquired according to sensor
The movable information of multiple individual of sample, to obtain the target movable information of optimization movement, because of the foundation of optimization so that
The target movable information can obtain faster, and also be more in line with the movement of individual of sample, compared with the existing technology in
The continuous modification of individual equipment is debugged, and the optimization efficiency of artificial intelligence equipment is improved.
Description of the drawings
Fig. 1 is the application scenarios one of motion control method provided in an embodiment of the present invention；
Fig. 2 is a kind of flow chart of motion control method provided in an embodiment of the present invention；
Fig. 3 is the flow chart of another motion control method provided in an embodiment of the present invention；
Fig. 4 is the application scenarios two of motion control method provided in an embodiment of the present invention；
Fig. 5 is the data coupling principle figure in motion control method provided in an embodiment of the present invention；
Fig. 6 is the structural schematic diagram of motion control device provided in an embodiment of the present invention；
Fig. 7 is the structural schematic diagram of artificial intelligence equipment provided in an embodiment of the present invention；
Fig. 8 is the entity structure schematic diagram of motion control device provided in an embodiment of the present invention；
Fig. 9 is the entity structure schematic diagram of artificial intelligence equipment provided in an embodiment of the present invention.
Specific implementation mode
Artificial intelligence equipment such as robot is needed when executing some action according to some action datas, example
Such as, it is assumed that robot lifts its arm, needs the parameters such as the angle that the height lifted according to arm, arm lift；Assuming that
Robot will walk, and need according to parameters such as step pitch, the speed of walking walked.The parameters such as above-mentioned height, speed are
It is stored in robot interior, such as the memory module being stored in the controller for controlling robot motion.
But in existing routine techniques, the above-mentioned parameter for controlling robot motion is relatively-stationary, once it will
In robot, robot just needs to move according to the parameter parameter setting, and it is also comparatively laborious to change parameter.This Shen
Please embodiment movement control mode, seek to obtain the above-mentioned action parameter for optimizing robot motion as soon as possible so that
Robot is quickly improved and perfect.It should be noted that in each embodiment of follow-up the application, artificial intelligence is set
It is standby to be illustrated by taking robot as an example, it is not limited to this in actual implementation, can also be the similar artificial intelligence of other control principles
It can equipment.
The motion control method of the embodiment of the present application, by acquiring the movable information of a large amount of individual of sample, and comprehensive root
Quickly the control parameter of robot motion is obtained according to these information.Specifically, referring to Fig. 1, the movement of the present embodiment is shown
A kind of optional application scenarios of control method briefly describe the lower scene first.
The individual of sample for acquiring movable information is chosen, which is, for example, human body, such as three shown in Fig. 1
Individual 11,12 and 13.In actual implementation, whom selects as individual of sample, it can be according to the movement of robot to be optimized
For foundation.For example, it is assumed that be cooked using robot, that can select some cooks as individual of sample, alternatively, also may be used
To be that ordinary people is not necessarily cook, but movement of the ordinary people when cooking to be selected to be acquired and (subsequently will be described)；Again
For example, it is assumed that use robot to carry out the workpiece handling of factory, some porters can be selected as individual of sample.
It is merely exemplary in Fig. 1 to show three in addition, the quantity of selected individual of sample can also flexibly be set
People, the quantity for being practically used for being elected to be the people of individual of sample can be very much, such as 50 people, 100 people.Under normal conditions, sample
This quantity obtained target datas more meet the actual motion conditions of human body more, for example, choosing 10 porter's tests
Obtained carrying action parameter will be easier accurately to summarize compared with choosing 2 and carrying the carrying action parameter that work obtains
Go out the preferable action parameter of carrying action.
As shown in Figure 1, being also provided with server 14, the data of above-mentioned individual of sample acquisition need to be transmitted to server 14,
Data are handled by server 14 to obtain target data.For example, in above-mentioned example, many porters' of acquisition removes
Fortune action data is transmitted to server 14, and server 14 carries out analyzing processing according to these data, it is general to obtain these workers
Move action parameter, for example everybody is essentially all arm to raise up a angles, and lift b height when carrying, in some instances it may even be possible to
It is c newton also to collect the arm strength usually applied when carrying, is equivalent to some for summing up most people movement
General character, as obtained target data.The target data is transmitted to robot 15 by server 14, and robot 15 is according to the mesh
When marking data motion, so that it may which, to obtain preferable movement effects, robot is quickly improved.
By foregoing description, as shown in Fig. 2, server 14 can execute following flow：
201, the multiple movable informations for corresponding to multiple individual of sample respectively of receiving sensor acquisition；
202, according to multiple movable informations, the target movable information that movement is corresponded to for optimizing movable information is obtained；
203, the target movable information is sent to artificial intelligence equipment, so that the artificial intelligence equipment is according to institute
It states target movable information and executes corresponding movement.
As shown in figure 3, robot 15 can execute following flow：
301, target movable information is obtained, the target movable information is point that motion control device is acquired according to sensor
The multiple movable informations for not corresponding to multiple individual of sample obtain；
302, corresponding movement is executed according to the target movable information, with by described in target movable information optimization
The corresponding movement of multiple movable informations.
Specifically, as follows will be in conjunction with some actual application scenarios or application mode, to above-mentioned server and robot
Processing be described in detail：
Using one：Assuming that robot is to be exclusively used in laundry clothes (perhaps the example is not very appropriate, but is merely to illustrate
The principle of the motion control method of the application), it is possible to choose usually the more female individual of laundry work is undertaken in family
As individual of sample, and only data of the acquisition individual when laundry takes.
Specifically, in conjunction with Fig. 1, sensor is set on female individual, sensor 16 is installed on individual 11, is pacified on individual 12
Sensor 17 is filled, sensor 18 is installed on individual 13.Since laundry clothes are mainly arm force, it is possible to by these sensors
It is arranged on arm.Laundry action parameter may include：Pendulum of the dynamics, arm that arm applies in laundry when giving the clothes a scrubbing
The corresponding various sensors for acquiring these parameters can be arranged in dynamic amplitude etc..In addition, in order not to be caused to the life of tester
These sensors only can be just installed and activated in inconvenience when laundry takes.
In the present embodiment, the data of sensor acquisition are properly termed as movable information, i.e., it is relevant that this movement is taken with laundry
Some information；Server 14 can be transferred to after the motion information acquisition, transfer mode is not limited, and can pass through network transmission
Information.In 14 side of server, many movable informations being received, these information correspond to different individual of sample respectively, such as
Three people shown in Fig. 1, everyone can transmit the movable information for the laundry clothes that the correspondence tester is collected.It is as follows
Table 1 illustrates optional information recording method formula：
The movable information of 1 sensor of table acquisition
| Sample identification | Acquisition time | Dynamics information | Velocity information | Amplitude information | Environmental information |
| Y1 | 8:29 | N1 | V1 | B1 | S1 |
| Y2 | 16:40 | N2 | V2 | B2 | S2 |
| Y3 | 20:12 | N3 | V3 | B3 | S3 |
It should be noted that be example some optional information in table 1, in actual implementation can according to actual conditions into
Row change.Ginseng is shown in Table 1, and Y1, Y2 and Y3 are, for example, to correspond respectively to three test human bodies 11,12 and 13 shown in Fig. 1.
Acquisition time can be that tester starts sensor, the startup time of sensor when laundry takes, and expression starts to wash in the time
Clothes；Optionally, which can also be a period, such as Y1 corresponding 8:29~9:01, therein 9:01 is
The time that laundry clothes terminate.Velocity information therein indicates the swing speed of tester's arm when laundry takes, and environmental information can
To be the humidity for the surrounding air that sensor acquires, it is generally the case that the environmental air humidity for clothes of doing washing can be relatively high.This
In embodiment, for example, Y1 is properly termed as an individual of sample, then corresponding to (Y1,8 of the Y1 in above-mentioned table 1:29、N1、V1、B1
And S1) parameter combination be properly termed as a movable information of corresponding Y1.
Server 14 will obtain after the above-mentioned multiple movable informations for getting sensor acquisition according to these movable informations
The target movable information that movement is corresponded to for optimizing movable information acquires many test samples for example, in above-mentioned example
It does washing the kinematic parameters of clothes, then clothes of what in the end doing washing could wash clothes cleaner, for example arm needs the width swung
Spend much, speed that arm is swung is how many etc., and it is just more suitable that the Data Summary by studying acquisition goes out how to move.Specific place
In reason, server can (since movable information is the human body of acquisition, the present embodiment claims according to multiple body motion informations
For body motion information) common feature, obtain the target body movable information of corresponding common feature, target body movement letter
Breath is exactly the clothes action more preferably parameter of doing washing that can allow to be found.
Still illustrate by taking table 1 as an example：Assuming that individual of sample has selected 10 people, 10 parts of fortune have actually been got in table 1
Dynamic information, wherein everyone has oneself in the kinematic parameter (such as speed, amplitude) when taking of doing washing, but server passes through
It is found after statistics, arm amplitude of fluctuation when having 7 people's laundries to take in this 10 people is in BmLeft and right, swing speed is in vmLeft and right, power
Degree is in Nm, then these parameters (Bm, vm, Nm) it can be known as common feature, that is, most people is all done so,
It should (Bm, vm, Nm) it can serve as target movable information.
As described above, a kind of selection mode that 7 personal data are selected from the sample data of 10 people is described, it is this
The mode of common feature is found it is also understood that according to preset data screening algorithm, is removed from movable information insincere
Data；According to the trust data that can not be except letter data in movable information, analysis obtains the target movable information.For example,
In this corresponding data of 10 people, the swing speed for clothes of doing washing is v1, v2, v3, v4 ... ... v10 respectively, wherein v2 to v8
Speed values (assuming that being identical speed unit) assume all 15 or so, have plenty of 14, have plenty of 16, have plenty of 15, and
It is the numerical value of 25, v10 is 10 that the numerical value of v1, which is the numerical value of 6, v9, then by statistics, it is therefore an objective to which searching meets most people
The characteristics of motion, then the numerical value of v1, v9 and v10 can be removed, it is believed that these be can not letter data, be equivalent to abnormal data,
Only retain the speed values of v2 to v8, it is believed that be trust data.And obtain target velocity number according to the speed values of v2 to v8
Value, for example 15 are can be set as, or it is set as the range section of 14-16.And which kind of preset data screening algorithm specifically uses
Algorithm the present embodiment is not limited, as long as the data screening function of above-mentioned principle can be realized.
Optionally, in actual test, everyone data may be not quite similar, and can not find duplicate multiple numbers
According to server can be according to certain algorithm at this time, and it is which type of range and then to derive one in find most of data
Common denominator data range or some point value within the scope of this can also.The target movable information to be obtained in a word is most of
Test sample used in parameter, in addition by above-mentioned it can be seen that, although by (Y1,8 in description above-mentioned:29、N1、V1、
B1 and S1) parameter combination as a movable information, but obtained target movable information can be part therein ginseng
Number (Bm, vm, Nm), guidance machine people movement can be used for.
Server 14 sends this information to artificial intelligence equipment such as robot, machine after obtaining target movable information
Device people can execute the action of laundry clothes according to the information, such as according to (Bm, vm, Nm) carry out robot arm swing and
Force.Since server passes to the target movable information of robot, obtained according to the actual motion Information Statistics of test human body
It arrives, is the corresponding parameter of action that most people is carried out, therefore the action that robot is executed according to the parameter can obtain
Relatively good effect；And the data of the not individual individual of sample of movable information of the present embodiment acquisition, many samples
The big data of the corresponding movable information of individual, can count to obtain preferable kinematic parameter.
In addition, the robot motion optimization of this mode, due to being the common denominator data for directly acquiring most samples, can make
The speed for obtaining robot motion's optimization improves, compared to traditional isolated debugging so that the acquisition efficiency of more excellent data is than very fast；
Also, it is sample data to be obtained by server automatic collection, and be transmitted to after statistics obtains target movable information in which
Robot, compared to traditional continuous modification and debugging to robot program so that the mode of robot optimization is more convenient
With it is quick, such as from the perspective of the user of robot, no longer need oneself to debug and modification robot parameter, but
Target movable information is obtained from server by robot oneself, which is that server is responsible for acquisition and processing obtains, and is not needed
User participates in, in short, the motion control method of the present embodiment not only increases the optimal speed of robot, and due to comparing
Meet the actual motion situation of individual of sample, effect of optimization will be more preferable.
From the perspective of robot 15, the target movable information that server process obtains can be stored in robot 15
Memory module in；When temporarily not passing through network connection between robot 15 and server 14, robot can will store mould
The foundation that the newest target movable information that the last time stored in block obtains is executed as action, alternatively, robot can also be with clothes
Device be engaged in by network connection, corresponding action is executed again after obtaining newest target movable information from server.
It is exemplified below：User may not want that the robot of oneself all the time by network attached server, then using
Person can be No. 1 morning 8:00, start the data acquisition functions of robot so that robot networking is downloaded newest from server
Target movable information, according to the information execute laundry clothes action.If user feels the pager according to this download
The effect of device people movement is relatively good, can use the information always；Alternatively, when user feels that robot motion is bad, clothes
It does not wash clean, it is desirable to when further increasing its movement effects, the data acquisition functions of robot can be again started up so that machine
People's networking downloads newer target movable information from server；Or user can also control robot periodically from service
Device obtains information.Certainly optional, server regularly can also push its newer target according to preset time to robot
Movable information, for example every other day pushed.
As described above, the data acquisition and processing (DAP) of server side is not only to carry out once, and the reception of data can be
It carries out always, or periodically carry out, so that target movable information is constantly updated.Specifically, for example, server can be set
A time interval is set, the calculating of performance objective movable information when preset time interval reaches, for example, it can be set to every
It 22：00, according to the movable information of the individual of sample of acquisition, calculates target movable information.For another example, server can also be
The calculating of performance objective movable information when multiple movable information updates are completed, such as the individual of sample of setting is 10, then this
The acquisition movable information of 10 individual of sample is to constantly update, and can may all have update, server receiving a number daily
According to rear, above-mentioned processing can be executed until the data of 10 all individual of sample all complete again by update.It can certainly
It is other setting means, no longer illustrates.
Using two：It is such as the mesh by taking some specific intended application scene as an example in the example of above-mentioned application one
Mark application scenarios are laundry clothes；It obtains taking the corresponding target movable information of scene with laundry, the laundry for optimizing robot takes
Movement.In the application two of the present embodiment, scene can not also be distinguished, such as the moment using someone as individual of sample, it can be with
Acquire various movements of this people in one day, such as laundry take, cooks, runs, dances etc., acquires moving for the tester always
Make；It is corresponding, robot receive target movable information also include many scenes under information, such as the data for clothes of doing washing,
The data etc. cooked, then the robot can be the robot for being able to carry out many motor tasks.
It is described in detail below：It can be arranged by taking the tester 11 in Fig. 1 as an example, on the tester for acquiring its movement letter
The various sensors of breath, the set-up mode of sensor can flexibly set, as possible so that tester carry it is more convenient and not
It can bring inconvenience to the action of tester.Tester 11 in Fig. 1 only illustrates one of setting sensor with it
16, for acquiring its arm movement, can also the biography for acquiring corresponding parameter be set in other body parts of the tester
Sensor.Sensor can be constantly in working condition, such as can get up to come into play to evening from tester on daytime and sleep
Between period, sensor can continual carry out motion pick.The movable information of acquisition is similar to shown in table 1, such as
Can also include walking step pitch, the speed of travel of tester, environmental information can also include the air themperature etc. of surrounding, specifically
It is arranged which kind of sensor acquires which kind of kinematic parameter, can be determined according to robot movement to be optimized.
In the present embodiment, server can first press these data after the above-mentioned data for receiving sensor acquisition
It is distinguished according to application scenarios, and corresponding target movable information is analyzed to each application scenarios respectively.It is exemplified below：With table
2 be to exemplify a part of acquisition parameter therein.
The movable information of 2 sensor of table acquisition
As shown in table 2, it is assumed that have the movable information of 6 people, including multi-motion parameter；May be in these movable informations
Corresponding different moving scene, include that the tester washes for example, when one day movable information of sensor collecting test people
Parameter when clothes, some actions when further including tester culinary art, further include action of the tester when arranging housework
Etc..Different moving scenes, tester would generally take different motion modes, such as arm when taking of doing washing is that have amplitude of fluctuation
Degree and swing speed, when hanging out, arm needs are raised, and when in culinary art, the movement of arm may be frying pan cooking
It moves up and down；Server can sort out these parameters according to moving scene, and the corresponding data of same class scene are just easier to find
Its common feature, the data of inhomogeneity scene are typically to be not easily found common feature.
By taking table 2 as an example, two kinds of moving scenes are illustrated, laundry clothes and culinary art, it is assumed that the movable information of 6 above-mentioned people is
The information of both movements is corresponded to respectively, and server can be distinguished according to environmental parameter.For example, air humidity meeting when laundry takes
It is relatively somewhat higher, and ambient air temperature when cooking can be relatively high, is based on this, server finds the movement of Y1, Y2 and Y5
Humidity parameter (S1, S2 and S5) in information takes all in laundry in corresponding humidity range, all relatively high, accordingly by Y1,
The movable information of Y2 and Y5 is classified as " laundry clothes move corresponding information ".In another example the air themperature in kitchen can phase when culinary art
To higher, based on this, server finds the temperature parameter (T3, T4 and T6) in the movable information of Y3, Y4 and Y6 all in air
It is all relatively high within the scope of temperature corresponding to temperature, the movable information of Y3, Y4 and Y6 are classified as " culinary art movement correspondence accordingly
Information ".
As described above, humidity parameter (S1, S2 and S5), temperature parameter (T3, T4 and T6) all can serve as to distinguish movement
This kind of parameter can be known as type of sports parameter, be determined for determining type of sports, such as according to humidity by the foundation of scene
Type of sports is laundry clothes, determines that type of sports is culinary art according to temperature.It, can be from collecting according to these type of sports parameters
Multiple movable informations in, obtain and belong to the movable information of same type of sports as reference movement information, for example, Y1, Y2 and
Y5 belongs to laundry clothes movement, and the corresponding movable information of these samples (such as dynamics, speed and amplitude etc.) is known as with reference to fortune
Dynamic information determines the target movable information for optimizing laundry clothes movement according to these reference movement information.It is same determining
After the reference movement information of type of sports, the mode that can be crossed as described above, according to preset data screening algorithm, from ginseng
It can not letter data according to removing in movable information；According to the trust data that can not be except letter data in reference movement information, analysis
Obtain the target movable information.
Certainly above-mentioned only to illustrate a kind of possible scene differentiating method, in specific implementation, server can also use
Other methods carry out scene differentiation, for example, the equipment for shooting tester's sport video can also be arranged, which will
The sport video of shooting is sent to server, and server can distinguish scene in conjunction with video analysis, for example server passes through picture
Face identifies that time when obtaining tester and cooking, then obtaining the culinary art is 7:07, tester upload is then searched again
The time point corresponding kinematic parameter in movable information, and it is culinary art to record the corresponding scene of these parameters.
In another example can also be to be controlled by tester, tester's do-it-yourself some moving scene (such as laundry clothes) it
Before, select corresponding scene (multiple scene options can be pre-set for surveying from the sensory-control system that it is carried first
Examination person selects) so that when sensor gathered data reports server, which is also transmitted to server together.
Other scenes are distinguished mode and are no longer itemized.
Server to the movable information of each scene, will find its common feature correspondence respectively after distinguishing moving scene
Target movable information, method detailed may refer to the embodiment using one, is no longer described in detail.Server will correspond to each field respectively
The target movable information of scape is sent to robot, and robot carries out pair according to these information when executing different scene motions
It should be executed according to parameter.
For example, robot interior stores the movement of laundry clothes and culinary art according to parameter, when robot executes laundry clothes fortune
When dynamic, then therefrom the parameter of acquisition laundry clothes corresponds to and executes action, when robot, which executes culinary art, to be moved, then therefrom obtains culinary art
Parameter correspond to and execute action.Robot how to distinguish oneself on earth execute which kind of movement, equally can there are many scene know
Other mode, can be artificial by the controller when needing robot to do which kind of movement for example, robot can have controller
Selection setting is carried out, then robot recalls corresponding movable information from memory module and executes accordingly；In another example robot also may be used
To perceive ambient enviroment by sensor, such as when humidity is higher, it is believed that be laundry clothes, then the movable information for selecting laundry to take,
It is considered to cook when temperature is higher, then selects the movable information of culinary art.Equally, can also judge in conjunction with video.
Optionally, if server does not differentiate between scene, other data analysing methods can also be taken, find target movement
Information, so that the movement of robot has better effect.
Using three：All it is the sample of selection by taking the movable information for obtaining human body as an example in above-mentioned application one and application two
This individual is people；In the application three of the present embodiment, the movable information for the individual of sample that server obtains is the movement of robot
Information, that is to say, that individual of sample further includes robot in addition to including tester, also, the movement acquired from robot counts greatly
Robot is re-entered according to the optimization that can also carry out common feature, and after optimizing and carrys out guidance machine people movement, forms robot fortune
The closed loop feedback of dynamic optimization, in this way but also the optimal speed to robot motion is further promoted, effect of optimization further carries
It is high.
Specifically, referring to Fig. 4, show that the principle of the method for the present embodiment, server 14 are obtained according to the parameter of acquisition
To after target body movable information, which can be sent to multiple robots, such as including robot 15, robot
It is only to illustrate three robots in 19 and robot 20, certain Fig. 4, can is many robot in actual implementation, to obtain
Obtain the sample of big data.
That is, in the present embodiment, the individual of sample for test includes not only people, further includes artificial intelligence equipment
Such as robot.It is also setting sensor in robot, acquires the movable information of the robot, for example, by taking laundry takes as an example, clothes
The relevant parameter of laundry clothes is transmitted to robot 15 by business device 14, and robot 15 executes the action of laundry clothes, acting accordingly
Cheng Zhong, the sensor being arranged in robot start to acquire corresponding parameter, obtain some movements of robot when laundry takes
Information, such as the swing speed of the mechanical arm of robot, amplitude of fluctuation etc., the acquisition of the movable information of robot and the information of people
The method of acquisition is essentially identical, is no longer described in detail.If being known as body motion information using people as the parameter acquired when individual of sample,
So it can be known as equipment moving information using robot as the parameter acquired when individual of sample.
It should be noted that the time that different robots obtains target movable information from server may not be to synchronize
, for example, robot 15 is set in the daily time 8:00 obtains newer target movable information from server 14, and robot
19 are set in the daily time 19：00 obtains newer target movable information from server 14, and robot 20 does not have the set time,
May be to control robot from server 14 when it is desirable that obtaining fresh information by the owner (or referred to as user) of robot
Obtain newer target movable information.So this mode is since the information of server side is to constantly update, in difference
Time obtain information each robot action based on parameter may be different, by sensor can acquire according to
According to the corresponding movable information of movement that different information executes, server is fed back to.
Server 14 can also be taken and believe with human motion after the equipment moving information for receiving many robot feedbacks
Identical method is ceased, the common feature of these equipment moving information is analyzed, and obtains corresponding target device movable information.And
And target body movable information and target device movable information can be carried out data coupling by server 14, and it is comprehensive to obtain a two
Close obtained newer target movable information.
Referring to Fig. 5, shows that server carries out the principle of data coupling, specifically transported according to body motion information and equipment
Dynamic information respectively in movable information each account for comparing and target body movable information and target device movable information are into line number
According to coupling, target movable information is obtained, referring to following formula：
Target movable information=target body movable information * body motion informations accounting+target device movable information * equipment
Movable information accounting；
It is exemplified below：For arm swing speed when laundry clothes movement, it is assumed that there are 20 personal accomplishment individual of sample, point
Not Ju You corresponding swing speed, and server analysis goes out corresponding target body movable information (the i.e. mesh of these speed
Mark swing speed) it is v1；Moreover, it is assumed that there are 10 robots as individual of sample, it is respectively provided with corresponding swing speed,
And it is v2 that server analysis, which goes out the corresponding target device movable information of these speed (i.e. target swing speed), then,
Body motion information accounting=20/ (20+10)=2/3；
Equipment moving information accounting=10/ (20+10)=1/3；
Target swing speed=v1* (2/3)+v2* (1/3)
Parameter in other movable informations can carry out the calculating of data coupling according to the method described above.In the present embodiment
In movement control mode, the big data acquired from a fairly large number of individual of sample that server obtains both had included being acquired from human body
Data, also include the data acquired from robot, and coupled to obtain new target according to this two parts data and moved
Information, server can be by the new target motion information transmissions to robot, to advanced optimize the action of robot.
Using four：In above-mentioned application three, server combines body motion information and equipment moving information, comprehensive
To the target movable information for guidance machine human action；In the application four of the present embodiment, it is further improved on this basis,
On the basis of server is optimized according to above two information, the proportion shared by body motion information is gradually reduced, and final
Form self-teaching and the closed loop training of robot itself.
Optionally, server can be adjusted body motion information according to preset information regularization condition, so that
Accounting of the body motion information in total movable information reduces, that is, gradually reduces the proportion shared by body motion information.It is above-mentioned
Preset information regularization condition, than if so, server just reduces every a week proportion of body motion information, such as
One week, proportion of the body motion information in total information was 70%, and the second week adjustment body motion information is moved always
Proportion in information is 50%, etc.；Alternatively, the proportion for how reducing body motion information can also be arranged in the condition, such as often
10% is reduced every a period of time, or reduces by 5% etc., various ways can flexibly be set.
Wherein, the accounting of body motion information reduces, and following mode may be used：It can gradually reduce as sample
The quantity of the test human body of body, for example, initially having selected 20 personal accomplishment individual of sample, after a period of time, eliminate 5
People is changed to the individual of sample of 15 people, using be reduced to for a period of time include 10 people individual of sample, gradually reduce until
The final individual of sample for cancelling human body.Or can also be the individual of sample of acquisition quantity it is constant, only server is being analyzed
When handling data, the data of the individual of sample used gradually reduce.When body motion information is reduced to zero, what server received
The movable information of individual of sample acquisition, the equipment moving information only acquired from robot by sensor, server is according to this
A little equipment moving information searching common features, obtain target device movable information, and feed back to robot, robot is according to the mesh
Marking device movable information executes corresponding movement again, feeds back to server again after collecting movable information, forms robot
The closed loop self-teaching of data.
In addition, during the training of self closed loop of robot, if the user of robot feels the dynamic of robot
Make the requirement that cannot meet user, user can also slightly modify to the kinematic parameter of robot, and correspondence is held in this way
The movable information of capable action can also change, and after being transmitted to server, be found altogether according to these movable informations by server
Property feature, the action of further guidance machine people.
It is not the modification tune only for single robot motion's parameter in the motion control method of the embodiment of the present application
Examination, but the movable information sample of the big data by the automatic collecting sensor acquisition of server, and according to these big data information
Sample finds common feature and obtains preferably kinematic parameter, and the action of guidance machine people is carried out with this so that the optimization of robot is imitated
Rate is improved, and effect of optimization is also preferable.
Furthermore, it is necessary to illustrate, the movement performed by robot, it can be not only laundry clothes, cook or execute certain
Kind factory task dispatching can help the mankind to handle the action of some task, can also be the robot for amusement, for example dance
Robot, or imitate the robot etc. run of the mankind, its fortune done also be intended to for these robots mankind
Kinetic energy is enough closer to the mankind because in this way the mankind can using these robots as the partner of the daily life of the mankind,
It accompanies the mankind to entertain or move together etc., the life of the mankind is also helpful in fact.In this case, so that it may with using above-mentioned
Method described in embodiment selects human body as individual of sample, acquires the movable information that many human bodies are danced, server evidence
This counts used action parameter when most of human bodies are danced, guidance machine people movement so that robot dances more like people
Class.Also, in the motion control method of the embodiment of the present application, the big data of collection of server is that continuous acquisition is constantly updated,
In this way so that for instructing the parameter of movement also constantly to be optimized.
On the basis of above description motion control method, the embodiment of the present application also provides a kind of motion control device,
The device can be arranged on the server 14, so that server executes corresponding motion control side in above method embodiment
Method.Referring to Fig. 6, which may include：Information receiving module 61, message processing module 62 and information sending module
62；Wherein,
Information receiving module 61, for multiple movable informations of receiving sensor acquisition, the multiple movable information difference
Corresponding multiple individual of sample；
Message processing module 62, for according to the multiple movable information, determining the target fortune for optimizing corresponding movement
Dynamic information；
Information sending module 63, for the target movable information to be sent to artificial intelligence equipment, so that the people
Work smart machine executes corresponding movement according to the target movable information.
Further, message processing module 62 are specifically used for, according to the type of sports parameter in the movable information, determining
The corresponding type of sports of the movable information；From the multiple movable information, the movement letter for belonging to same type of sports is obtained
Breath is used as reference movement information, and determines the target movable information for optimizing corresponding movement according to the reference movement information.
Further, the individual of sample includes：Human body；The movable information includes：The human body of sensor acquisition
Corresponding body motion information.The individual of sample further includes：Smart machine；The movable information further includes：Sensor acquires
The corresponding equipment moving information of smart machine.Message processing module 62, specifically for determining target according to body motion information
Body motion information；Target device movable information is determined according to equipment moving information；And according to the target body movable information,
Target device movable information and the body motion information and equipment moving information are respective in the movable information respectively
Accounting carries out data coupling, obtains the target movable information.
The embodiment of the present application also provides a kind of artificial intelligence equipment, which is, for example, robot；Referring to Fig. 7, the people
Work smart machine may include：Data obtaining module 71 and Motor execution module 72；Wherein,
Data obtaining module 71, for obtaining target movable information, the target movable information is motion control device root
The multiple movable informations acquired according to sensor obtain, and the multiple movable information corresponds to multiple individual of sample respectively；
Motor execution module 72, for executing corresponding movement according to the target movable information, to pass through the target
Movable information optimizes the corresponding movement of the multiple movable information.
Optionally, data obtaining module 71 are specifically used for：Acquisition is stored in advance in the local target movable information；
Alternatively, obtaining the target movable information from the motion control device.
Referring to FIG. 8, the embodiment of the present application also provides a kind of entity structure schematic diagrames of motion control device.The movement
Control device may be the host server personal computer comprising computing capability or portable portable
Computer or terminal etc., the present embodiment do not limit the specific implementation of the device.As shown in figure 8, the motion control fills
It sets and may include：Processor (processor) 810, communication interface (Communications Interface) 820, memory
(memory) 830 and bus 840.
Wherein, processor 810, communication interface 820, memory 830 complete mutual communication by bus 840.Communication
Interface 820 communicates the transmission for carrying out movable information for being carried out with the communication interface in sensor or robot.
Processor 810 a, it may be possible to central processor CPU or specific integrated circuit ASIC (Application
Specific Integrated Circuit), or be arranged to implement one or more integrated circuits of the present embodiment.
Memory 830 may include high-speed RAM memory, it is also possible to further include nonvolatile memory (non-volatile
Memory), a for example, at least magnetic disk storage, the memory 830 is for storing program instruction.Processor 810 can execute
Program instruction in memory 830, for executing following work：Receiving sensor acquisition corresponds to multiple individual of sample respectively
Multiple movable informations；According to the multiple movable information, the target fortune that movement is corresponded to for optimizing the movable information is obtained
Dynamic information；The target movable information is sent to artificial intelligence equipment, so that the artificial intelligence equipment is according to the mesh
It marks movable information and executes corresponding movement.
Referring to FIG. 9, the embodiment of the present application also provides a kind of entity structure schematic diagrames of artificial intelligence equipment.This is artificial
Artificial intelligence equipment is, for example, robot.As shown in figure 9, the artificial smart machine may include：Processor (processor)
910, communication interface (Communications Interface) 920, memory (memory) 930 and bus 940.
Wherein, processor 910, communication interface 920, memory 930 complete mutual communication by bus 940.Communication
Interface 920 communicates the transmission for carrying out movable information for being carried out with server.
Processor 910 a, it may be possible to central processor CPU or specific integrated circuit ASIC (Application
Specific Integrated Circuit), or be arranged to implement one or more integrated circuits of the present embodiment.
Memory 930 may include high-speed RAM memory, it is also possible to further include nonvolatile memory (non-volatile
Memory), a for example, at least magnetic disk storage, the memory 930 is for storing program instruction.Processor 910 can execute
Program instruction in memory 930, for executing following work：Target movable information is obtained, the target movable information is
Motion control device is obtained according to the multiple movable informations for corresponding to multiple individual of sample respectively that sensor acquires；According to the mesh
It marks movable information and executes corresponding movement, to optimize the corresponding fortune of the multiple movable information by the target movable information
It is dynamic.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, it can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of division of logic function, formula that in actual implementation, there may be another division manner, such as multiple units or component
It can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown or
The mutual coupling, direct-coupling or communication connection discussed can be by some communication interfaces, between device or unit
Coupling or communication connection are connect, can be electrical, machinery or other forms.
The unit illustrated as separating component may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, you can be located at a place, or may be distributed over multiple
In network element.Some or all of unit therein can be selected according to the actual needs to realize the mesh of this embodiment scheme
's.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, can also be
Each unit physically exists alone, can also be during two or more units are integrated in one unit.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer read/write memory medium.Based on this understanding, technical scheme of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be expressed in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be
People's computer, server or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes：USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic disc or CD.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
With within principle, any modification, equivalent substitution, improvement and etc. done should be included within the scope of protection of the invention god.
Claims (16)
1. a kind of motion control method, which is characterized in that including：
Multiple movable informations of sensor acquisition are obtained, the multiple movable information corresponds to multiple individual of sample respectively；
According to the multiple movable information, the target movable information for optimizing corresponding movement, the target movable information are determined
Common feature in corresponding the multiple movable information；
The target movable information is sent to artificial intelligence equipment, so that the artificial intelligence equipment is transported according to the target
Dynamic information executes corresponding movement.
2. according to the method described in claim 1, it is characterized in that, described according to the multiple movable information, determination is for excellent
Change the target movable information of corresponding movement, specially：
When preset time interval reaches, or when the multiple movable information updates completion, according to the multiple movement
Information calculates the target movable information for optimizing corresponding movement.
3. according to the method described in claim 1, it is characterized in that, described according to the multiple movable information, determination is for excellent
Change the target movable information of corresponding movement, including：
According to the type of sports parameter in the movable information, the corresponding type of sports of the movable information is determined；
From the multiple movable information, obtains and belong to the movable information of same type of sports as reference movement information, and root
The target movable information for optimizing corresponding movement is determined according to the reference movement information.
4. according to the method described in claim 3, it is characterized in that, described determine according to the reference movement information for optimizing
The target movable information of corresponding movement, including：
According to preset data screening algorithm, removing from the reference movement information can not letter data；
According to the trust data that can not be except letter data in the reference movement information, analysis obtains the target movement letter
Breath.
5. according to the method described in claim 1, it is characterized in that, the individual of sample includes：Human body；The movable information packet
It includes：The corresponding body motion information of the human body of sensor acquisition.
6. according to the method described in claim 5, it is characterized in that, the individual of sample further includes：Smart machine；The movement
Information further includes：The corresponding equipment moving information of the smart machine of sensor acquisition；It is described to be believed according to the multiple movement
Breath obtains the target movable information for optimizing corresponding movement, including：
Target body movable information is determined according to body motion information；
Target device movable information is determined according to equipment moving information；
According to the target body movable information, target device movable information and the body motion information and equipment moving
Information each accounts for comparing in the movable information respectively, carries out data coupling, obtains the target movable information.
7. according to the method described in claim 6, it is characterized in that, determining that target body is transported according to body motion information described
Before dynamic information, further include：
According to preset information regularization condition, the body motion information is adjusted, so that the body motion information
Accounting in the movable information reduces.
8. a kind of motion control method, which is characterized in that including：
Target movable information is obtained, the target movable information is that motion control device is believed according to multiple movements that sensor acquires
Breath obtains, and the multiple movable information corresponds to multiple individual of sample respectively；The target movable information corresponds to the multiple movement
Common feature in information；
Corresponding movement is executed according to the target movable information, to optimize the multiple movement by the target movable information
The corresponding movement of information.
9. according to the method described in claim 8, it is characterized in that, the acquisition target movable information, including：Acquisition is deposited in advance
It stores up in the local target movable information；Alternatively, obtaining the target movable information from the motion control device.
10. the method stated according to claim 9, which is characterized in that described to obtain the target fortune from the motion control device
Dynamic information, including：
According to preset time the newer target movable information is received from the motion control device.
11. a kind of motion control device, which is characterized in that including：
Information receiving module, for multiple movable informations of receiving sensor acquisition, correspondence is more respectively for the multiple movable information
A individual of sample；
Message processing module, for according to the multiple movable information, determining the target movable information for optimizing corresponding movement；
The target movable information corresponds to the common feature in the multiple movable information；
Information sending module, for the target movable information to be sent to artificial intelligence equipment, so that the artificial intelligence
Equipment executes corresponding movement according to the target movable information.
12. according to the devices described in claim 11, which is characterized in that
Described information processing module is specifically used for, according to the type of sports parameter in the movable information, determining the movement letter
Cease corresponding type of sports；From the multiple movable information, the movable information for belonging to same type of sports is obtained as reference
Movable information, and the target movable information for optimizing corresponding movement is determined according to the reference movement information.
13. according to the devices described in claim 11, which is characterized in that the individual of sample includes：Human body；The movable information
Including：The corresponding body motion information of the human body of sensor acquisition.
14. device according to claim 13, which is characterized in that the individual of sample further includes：Smart machine；The fortune
Dynamic information further includes：The corresponding equipment moving information of smart machine of sensor acquisition；
Described information processing module, specifically for determining target body movable information according to body motion information；It is transported according to equipment
Dynamic information determines target device movable information；And according to the target body movable information, target device movable information, Yi Jisuo
It states body motion information and equipment moving information to each account for comparing in the movable information respectively, carries out data coupling, obtain
The target movable information.
15. a kind of artificial intelligence equipment, which is characterized in that including：
Data obtaining module, for obtaining target movable information, the target movable information is motion control device according to sensing
Multiple movable informations of device acquisition obtain, and the multiple movable information corresponds to multiple individual of sample respectively；The target movement letter
Common feature in the corresponding the multiple movable information of breath；
Motor execution module is believed for executing corresponding movement according to the target movable information with being moved by the target
Breath optimizes the corresponding movement of the multiple movable information.
16. artificial intelligence equipment according to claim 15, which is characterized in that
Described information acquisition module, is specifically used for：Acquisition is stored in advance in the local target movable information；Alternatively, from institute
It states motion control device and obtains the target movable information.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410459517.7A CN105467926B (en) | 2014-09-10 | 2014-09-10 | A kind of motion control method and device, artificial intelligence equipment |
| CN201810672928.2A CN109048889B (en) | 2014-09-10 | 2014-09-10 | Method and device for obtaining target motion information of artificial intelligence equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410459517.7A CN105467926B (en) | 2014-09-10 | 2014-09-10 | A kind of motion control method and device, artificial intelligence equipment |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810672928.2A Division CN109048889B (en) | 2014-09-10 | 2014-09-10 | Method and device for obtaining target motion information of artificial intelligence equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105467926A CN105467926A (en) | 2016-04-06 |
| CN105467926B true CN105467926B (en) | 2018-08-07 |
Family
ID=55605737
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810672928.2A Active CN109048889B (en) | 2014-09-10 | 2014-09-10 | Method and device for obtaining target motion information of artificial intelligence equipment |
| CN201410459517.7A Active CN105467926B (en) | 2014-09-10 | 2014-09-10 | A kind of motion control method and device, artificial intelligence equipment |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810672928.2A Active CN109048889B (en) | 2014-09-10 | 2014-09-10 | Method and device for obtaining target motion information of artificial intelligence equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN109048889B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108242102B (en) * | 2017-12-28 | 2020-04-24 | 纳恩博（北京）科技有限公司 | Information processing method and device, electronic device and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1830635A (en) * | 2005-03-10 | 2006-09-13 | 北京理工大学 | Human imitation robot action similarity evaluation based on human body motion track |
| DE102006017945A1 (en) * | 2005-04-15 | 2006-12-14 | Fanuc Robotics America, Inc., Rochester Hills | Robot control system, has personal computer connected with robot control and robot power data monitoring unit so that computer responds to power data to use optimization routines and to change robot control system parameter in real time |
| JP2009502527A (en) * | 2005-08-01 | 2009-01-29 | トヨタ自動車株式会社 | Robot control system and robot control method |
| CN103605375A (en) * | 2013-10-24 | 2014-02-26 | 国家电网公司 | Bionic robot control method and bionic robot control device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101266619B (en) * | 2008-05-12 | 2010-07-21 | 腾讯科技（深圳）有限公司 | User information excavation method and system |
-
2014
- 2014-09-10 CN CN201810672928.2A patent/CN109048889B/en active Active
- 2014-09-10 CN CN201410459517.7A patent/CN105467926B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1830635A (en) * | 2005-03-10 | 2006-09-13 | 北京理工大学 | Human imitation robot action similarity evaluation based on human body motion track |
| DE102006017945A1 (en) * | 2005-04-15 | 2006-12-14 | Fanuc Robotics America, Inc., Rochester Hills | Robot control system, has personal computer connected with robot control and robot power data monitoring unit so that computer responds to power data to use optimization routines and to change robot control system parameter in real time |
| JP2009502527A (en) * | 2005-08-01 | 2009-01-29 | トヨタ自動車株式会社 | Robot control system and robot control method |
| CN103605375A (en) * | 2013-10-24 | 2014-02-26 | 国家电网公司 | Bionic robot control method and bionic robot control device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105467926A (en) | 2016-04-06 |
| CN109048889A (en) | 2018-12-21 |
| CN109048889B (en) | 2021-03-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ryu et al. | Automated action recognition using an accelerometer-embedded wristband-type activity tracker | |
| CN105726267B (en) | Auxiliary torque setting method and equipment | |
| KR101942617B1 (en) | Task scheduling method and system in cloud computing | |
| Uddin et al. | Human activity recognition from wearable sensors using extremely randomized trees | |
| CN104007822A (en) | Large database based motion recognition method and device | |
| CN105912142B (en) | A kind of note step and Activity recognition method based on acceleration sensor | |
| CN106796665A (en) | Method and system for generating instructions for a robotic system to perform a task | |
| CN105205436A (en) | Gesture identification system based on multiple forearm bioelectric sensors | |
| CN109889592A (en) | A kind of intelligence manufacture method and device based on edge calculations | |
| CN103616034A (en) | Network pedometer based on Bluetooth and step calculation method | |
| CN111288986B (en) | Motion recognition method and motion recognition device | |
| CN110936370A (en) | Cleaning robot control method and device | |
| CN103699795A (en) | Exercise behavior identification method and device and exercise intensity monitoring system | |
| CN105320653A (en) | Consuming behavior pattern collecting system and method | |
| CN106123911A (en) | A kind of based on acceleration sensor with the step recording method of angular-rate sensor | |
| CN107077651A (en) | Robot cooperation method, device, robot and computer program product | |
| CN201978444U (en) | Motion detecting and analyzing system | |
| CN105807903A (en) | Control method and device of intelligent equipment | |
| CN103517118B (en) | The action identification method of a kind of remote controller and system | |
| CN105138127A (en) | Left and right hand distinguishing method and device for wearable device and wearable device | |
| CN108008151A (en) | A kind of moving state identification method and system based on 3-axis acceleration sensor | |
| CN109460792A (en) | A kind of artificial intelligence model training method and device based on image recognition | |
| CN105467926B (en) | A kind of motion control method and device, artificial intelligence equipment | |
| CN105589977A (en) | Movement frequency monitoring method and device | |
| CN106994238A (en) | Data processing method and the electronic equipment for performing the data processing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200923 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Patentee after: Innovative advanced technology Co.,Ltd. Address before: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Patentee before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20200923 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Patentee after: Advanced innovation technology Co.,Ltd. Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands Patentee before: Alibaba Group Holding Ltd. |