CN103499972A - Subsumption hierarchical structure-based universal robot multi-scene control method - Google Patents

Subsumption hierarchical structure-based universal robot multi-scene control method Download PDF

Info

Publication number
CN103499972A
CN103499972A CN201310461873.8A CN201310461873A CN103499972A CN 103499972 A CN103499972 A CN 103499972A CN 201310461873 A CN201310461873 A CN 201310461873A CN 103499972 A CN103499972 A CN 103499972A
Authority
CN
China
Prior art keywords
instruction
hierarchical structure
priority
socket
control method
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201310461873.8A
Other languages
Chinese (zh)
Other versions
CN103499972B (en
Inventor
申富饶
干强
赵金熙
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University
Original Assignee
Nanjing University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing University filed Critical Nanjing University
Priority to CN201310461873.8A priority Critical patent/CN103499972B/en
Publication of CN103499972A publication Critical patent/CN103499972A/en
Application granted granted Critical
Publication of CN103499972B publication Critical patent/CN103499972B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention discloses a subsumption hierarchical structure-based universal robot multi-scene control method, which is used for controlling a service robot by using a priority mechanism and a prohibition mechanism through a keyboard instruction, a voice instruction, a socket instruction and a mail instruction on a plurality of scenes, and has a simple obstacle avoidance walking function. The method has the obvious advantages that each control mode is executed in parallel in a hierarchical structure, the simple priority mechanism and the prohibition mechanism are adopted between hierarchies to coordinate so as to realize good real-time property; the control method is software design and is unrelated to a robot platform so as to realize good transplantability; no master control module is arranged, and a new function is added without rewriting the conventional module so as to realize good extensibility; a coordination mechanism is simple, and the requirements on a processor are low compared with a complex master control module, so that the hardware cost is reduced, and the generalization is easy.

Description

A kind of many scenes of all-purpose robot control method based on the containment type hierarchical structure
Technical field
The present invention relates to the robot control field, particularly a kind of many scenes of all-purpose robot control method based on the containment type hierarchical structure.
Background technology
Mention " intellect service robot is that the integrated intellectuality equipment of multiple hi-tech of essential service is provided for the mankind under non-structure environment " in China's " National Program for Medium-to Long-term Scientific and Technological Development (2006~the year two thousand twenty) ".
Along with the development of Robotics, robot control method also changes at development.Traditional control method adopts " perception → plan → action " (" Sense → Plan → Act ") framework, first by the sensor senses surrounding environment, receive dissimilar instruction, then information is delivered to main control module and carried out analysis-by-synthesis, finally show that the corresponding module execution is transferred in next step action.A kind of multi-mode (multimodal) robot control system proposed such as the people such as Felipe Trujillo-Romero of Technological University of the Mixteca, at first this system obtains image and voice messaging from camera and microphone, through sending into main control module after identification, then main control module is used Voting Algorithm to determine next step behavior, and last CD-ROM drive motor is carried out.The HERB that domestic and international famous research institution proposes in recent years, ARMAR-III, the robots such as PengPengII use this framework, and have obtained good intelligent effect.
" but perception → plan → action " (" Sense → Plan → Act ") framework not too is suitable as the control system of service robot.Mainly contain three reasons:
1,, when adding or delete sensor and auxiliary function module, need the analysis-by-synthesis algorithm in the redesign main control module.This makes the existing robot system of renewal become of a high price.
2, service robot is mainly used to receives user's and carries out, and without the need for very strong capacity of will, but requires good real-time.Yet along with the continuous increase of robot sensor and auxiliary function module, the analysis-by-synthesis algorithm of main control module can become very complicated, will affect the real-time of service robot.
3, complicated main control module can be very high to processor requirement, and the hardware cost that this will improve service robot, affect the popularization of service robot.
The Brooks of MIT once proposed a kind of containment type level framework (subsumption) and overcame " perception → plan → action "
The shortcoming of (" Sense → Plan → Act ") framework.In containment type level framework (subsumption), robot control system is divided level by the Task-decomposing parallel processing, between level, use the simple mechanism (suppression and inhibition) that suppresses to be coordinated, there is no main control module.Use the robot of containment type level framework that Myrmix etc. is arranged, have very high real-time performance.
But there are two major defects in containment type level framework (subsumption):
1,, when system level increases, be difficult to design inhibition mechanism and decompose well parallel by task.
2, from circuit aspect control, so robot control system does not possess portability.
In recent years, company of many robots is (as Japanese Vstone Co., Ltd., Beijing Universal Pioneering Technology Co., Ltd. etc.) by the abstract interface that becomes the C/C++ language of the control of hardware view, this separates with robot hardware's platform for robot control system, realizes that the portability of robot system provides technical support.
Summary of the invention
Goal of the invention: technical matters to be solved by this invention is for the deficiencies in the prior art, and a kind of many scenes of all-purpose robot control method based on the containment type hierarchical structure is provided.
In order to solve the problems of the technologies described above, the invention discloses a kind of many scenes of containment type hierarchical service robot control method, for under a plurality of scenes, utilize priority mechanism and forbid mechanism, control service robot by keyboard instruction, phonetic order, socket (Socket) instruction and E-mail instructions, and have and simply keep away the barrier walking function, comprise the following steps:
(1) the people's platform that starts the machine, start external or Embedded computing machine, and camera, microphone are connected to computing machine;
(2) by the computing machine access network, and set up socket as server end and client and be connected;
(3) create four instruction queues, for depositing four steering orders under scene;
(4) define the priority of four instruction queues;
(5) start four control modules for receiving instruction;
(6) control module is by the instruction the received corresponding instruction queue of packing into;
(7) take out the highest instruction of priority from four instruction queues;
(8) recognition instruction;
(9) carry out instruction, return to step (6).
Wherein step (8) is further comprising the steps:
(81) if the instruction None-identified provides " order of Unknown Command(None-identified) " and points out and abandon controlling interface;
(82) if instruction can be identified, enter step (9)
Wherein step (9) is further comprising the steps:
(91) if instruction can't be carried out, at the control interface, provide " barrier appears in Obstacle ahead(the place ahead) "
Point out and abandon;
(92) if instruction can be carried out, execution is returned to step (6) after finishing.
In said method, robot platform refers to provide all robots of C/C++ language interface of the control of hardware view.
In said method, external or Embedded computing machine refers to: windows operating system, camera and microphone driver, Microsoft's speech recognition software are installed, and are had the computing machine of network connecting function.
In said method, control scenes for four and refer to respectively: the external or embedded computer (being designated as scene-0) of direct control, with robot at same room with them (being designated as scene-1), with robot can surf the Net outside house, same building (being designated as scene-2), house, (be designated as scene-3) all over the world.
In said method, four kinds of control modes refer to respectively: keyboard instruction, phonetic order, socket (Socket) instruction, E-mail instructions.
In said method, to take out the highest instruction of priority and completed by priority mechanism, the effect of realization is: the priority between instruction queue is defined by the user, and it is high that the user is defined as high instruction queue priority; The priority of same instruction queue inside determines by time sequencing, and the instruction priority of first joining the team is high.
In said method, four control modules refer to respectively: keyboard receiver module, sound identification module, socket (Socket) communication module, mail reception module.
In said method, the Starting mode of control module is: with form the concurrent or executed in parallel of separate threads.
In said method, recognition instruction refers to: search the processing mode of present instruction in system, if there is no definition process mode, None-identified.
In said method, carry out instruction and comprise that calling service software carries out instruction and CD-ROM drive motor execution instruction.Service software comprises: weather forecast, mail send, visiting notice.CD-ROM drive motor comprises: advance, walk left, walk to the right, turn left, turn right, retreat.
In said method, keep away the barrier walking function by the obstacle detection module and forbid that mechanism completes, when obstacle detection module output obstacle information, can trigger and forbid mechanism, make motor to take action.
The present invention overcomes the shortcoming of tradition " perception → plan → action " (" Sense → Plan → Act ") framework in the service robot application, introduce containment type hierarchical structure (subsumption) technology, robot control system is carried out to Hierarchical Design, use the simple coordination system, cancel the setting of main control module, also overcome inhibition mechanism in containment type hierarchical structure (subsumption) and be difficult to design simultaneously, portable poor shortcoming, propose a kind ofly under several scenes, can control, use priority mechanism and forbid two kinds of simple coordination systems of mechanism, there is better real-time, the service robot control method of portability and extensibility.
Beneficial effect: remarkable advantage of the present invention be every kind of control mode with the hierarchical structure executed in parallel, uses simple priority mechanism between level and forbid that mechanism is coordinated, so real-time is good; This control method is Software for Design, irrelevant with robot platform, therefore portable good; Main control module is not set, and increasing new function does not need again to write existing module, so extensibility is good; The coordination system is simple, compares complicated main control module low to processor requirement, so the hardware cost reduction, is easy to popularize.
The accompanying drawing explanation
Below in conjunction with the drawings and specific embodiments, the present invention is done further and illustrates, above-mentioned and/or otherwise advantage of the present invention will become apparent.
Fig. 1 is Organization Chart of the present invention.
Fig. 2 is work operational flow diagram of the present invention.
Fig. 3 is the sound identification module schematic diagram.
Fig. 4 is socket communication module schematic diagram.
Fig. 5 is the mail reception module diagram.
Fig. 6 is the obstacle detection module diagram.
Fig. 7 is the priority mechanism shop drawing.
Fig. 8 forbids mechanism works figure.
Embodiment
Fig. 1 is Organization Chart of the present invention.(it is example higher than scene-2 higher than scene-3 higher than scene-1 that the priority of take is set to scene-0)
The present invention is designed to control mode four levels of corresponding scene:
(1) scene-0 of the corresponding direct control of Keyboard Control robot.Receive information by keyboard, by the queue-0 of keyboard instruction load; The implementation of keyboard receiver module is: use scanf () function to read user's keyboard input from control desk, be packaged into the instruction of text formatting, the push function of call instruction queue-0 is by the keyboard instruction queue of packing into.
(2) correspondence and the scene-1 of robot at same room with them controlled in voice.Receive information by microphone, at sound identification module by the queue-1 of phonetic order load; As shown in Figure 3, the implementation of sound identification module is: Microsoft Speech SDK5.1 and Chinese language bag are installed, the speech recognition function that uses Microsoft Speech SDK5.1 to carry is write the audio monitoring device, call CComPtr<ISpRecoResult > in GetText () function the voice messaging listened to is converted to Word message, Word message is extracted to key word, be packaged into the instruction of text formatting, the push function of call instruction queue-1 is by the phonetic order queue of packing into.
(3) socket (Socket) is controlled correspondence and the scene-2 of robot in house, same building.By socket (Socket) reception information, socket (Socket) communication module is by the queue-2 of socket (Socket) instruction load; As shown in Figure 4, the implementation of socket (Socket) communication module is: computing machine as the service end of socket (Socket) communication by server address, bound socket (bind function), listening port (listen function) and client connect (accept function) are set, then receive socket (Socket) information (recv function) from client, again to socket (Socket) information extraction key word, be packaged into the instruction of text formatting, the push function of call instruction queue-2 is by the Socket instruction queue of packing into.
(4) mail is controlled the corresponding scene all over the world-3 that can network.Receive information by mailing system, the mail control module is E-mail instructions, load queue-3; As shown in Figure 5, the implementation of mail control module is: install and register the Jmail4.4 assembly, Connect () function by the Jmail4.4 assembly connects mail server, from Messages member's mail tabulation, poor according to last number of mail and number of mail this time, obtain the unread mail of specifying mailbox, then Mail Contents extracted to key word, be packaged into the instruction of text formatting, load queue-3.
In the robot control system shown in Fig. 1, priority mechanism one co-exists in three places.The first place is the obstruction of instruction queue-0 for instruction queue-1, as long as there is instruction output instruction queue-0, instruction queue-1 just gets clogged, and identification is preferentially sent in the instruction of instruction queue-0.The second place is instruction queue-0 and 1 obstruction for instruction queue-2, as long as there is instruction output instruction queue-0 or 1, instruction queue-2 just get clogged, and identification is preferentially sent in instruction queue 0 or 1 instruction.The 3rd place is instruction queue-0,1 and 2 obstructions for instruction queue-3, as long as there is instruction output instruction queue-0,1 or 2, instruction queue 3 just gets clogged, and identification is preferentially sent in instruction queue 0,1 or 2 instruction.
Fig. 2 is work operational flow diagram of the present invention.
Step S01: the people's platform that starts the machine, start external or Embedded computing machine, camera, microphone are connected to computing machine.
In this step, the robot external world or embedded computer need to be installed the computing machine of windows operating system, camera and microphone driver, Microsoft's speech recognition software.
Step S02: by the computing machine access network, and set up socket as server end and client and be connected.
Computing machine is as server end in this step, and in house, other control ends are as client, and the socket that sets up one-to-many connects.
Step S03: create four instruction queues, corresponding four control modes of controlling under scene.
In this step, the establishment mode of instruction queue is: newly-built four C++ template class queue, the element of queue template is the instruction of text formatting.
Step S04: the priority that defines four instruction queues.
In this step, definite mode of priority is: the priority between instruction queue is defined by the user, and it is high that the user is defined as high instruction queue priority.
Step S05: start four control modules for receiving instruction.
In this step, the Starting mode of control module is: use four threads of pthread_create function creation, each control module is with form the concurrent or executed in parallel of separate threads.
Step S06: control module is by the instruction the received corresponding instruction queue of packing into.
In this step, the mode that instruction is joined the team is: the push function that control module is used the queue template is by the instruction corresponding instruction queue of packing into.
Step S07: take out the highest instruction of priority from four instruction queues.
In this step, as shown in Figure 7, four instruction queues are under the coordination of priority mechanism, and the instruction of limit priority is the instruction of first taking-up, use queue template pop function to take out instruction.
Step S08: recognition instruction.
In this step, search the processing mode of present instruction in system, if there is no the definition process mode, enter step S09; Otherwise enter step S10.
Step S09: be given in the control interface and provide the prompting of " Unknown Command ", and abandon, return to step S07.
Step S10: whether decision instruction can be carried out.
In this step, if present instruction is CD-ROM drive motor, and the information that can't carry out of obstacle detection module output, triggers and forbid that as shown in Figure 8 mechanism forbids CD-ROM drive motor, enter step S11; Otherwise enter step S12.
Step S11: provide " Obstacle ahead " prompting at the control interface, and abandon, return to step S07.
Step S12: carry out instruction, and return to step S07.
As shown in Figure 1, robot behavior can comprise the service software module, comprising: weather forecast, mail send, visiting notice.
The implementation of weather forecast is: use gethostbyname () function to obtain the address of the weather forecast website of www.webxml.com.cn by name, setting up Socket with this weather forecast website is connected, use send () function to send request message to this weather forecast website, import the city code into as parameter, calculate the side-play amount of required Weather information in returned packet (recv function), thereby obtain weather forecast, the Speak function re-used in Microsoft SDK5.1 carries out reading aloud of weather forecast.
The implementation that mail sends is: the Jmail4.4 assembly is installed, is used the function SendMail that sends out fast mail in Jmail4.4, sender's mailbox, addressee's mailbox, mail matter topics, text, smtp server are imported into parametric form.
The implementation of visiting notice is: Jmail4.4 assembly and OpenCV2.3.1 storehouse are installed, use the function cvCaptureFromCAM () that takes pictures in the OpenCV2.3.1 storehouse to take pictures to the visitor, use the function Send () of hair band annex mail in Jmail4.4, sender's mailbox, addressee's mailbox, mail matter topics, text, smtp server are imported into the form of member variable, visitor's photo of taking is imported into attachment version (AddAttachment function).
The effect of drive motor module is shielding robot platform details, realizes the portability of system.Concrete implementation is: the interface interchange sequence that each motor action is converted to concrete robot platform.The motor action of the robot system that we have realized comprises: advance, walk left, walk to the right, turn left, turn right, retreat.The implementation of each action is the motor port of calling the C/C++ language that robot platform provides.Artificially routine with the Robovie_X machine, the realization of each action is first uses VSRC003_LoadMotion () function to be written into motor port, re-uses VSRC003_PlayMotion () function and performs an action.The correspondence table of instruction-action-interface is:
Instruction Action Interface
forward Advance Go_Forward(advanced).txt
goleft Walk left Side?step?left(1step).txt
goright Walk to the right Side?step?right(1step).txt
turnleft Turn left Turn?in?place?left.txt
turnright Turn right Turn?in?place?right.txt
back Retreat Walk?backwards.txt
Transplanting mode of the present invention is as follows:
The transplanting condition provides the C/C++ language interface for robot platform.
When the present invention is transplanted to new engine people platform R, existing motor is moved: advance, walk left, walk to the right, turn left, turn right, retreat, be converted to respectively the motor port calling sequence of R robot.For example, when the R robot is anthropomorphic robot, the action of " advancing " is converted to that leg " is lifted " by the R robot and alternately the calling of " putting leg " interface; When the R robot is the spider-shaped robot, the action of " advancing " is converted to R robot repeatedly the calling of interface of " creeping ".
Through experimental verification, after replacing the calling interface in drive motor module, can realize the transplanting of system.As Fig. 9 is two kinds of robot platforms (left figure is Robovie-X, and right figure is VstoneTichno) of experiment use, following table is the hardware combinations of two groups of experiments.
The present invention has also designed and has kept away the barrier walking function, while avoiding the robot action and barrier collision.As shown in Figure 6, the implementation of obstacle detection module is: the OpenCV2.3.1 storehouse is installed, obtaining video by OpenCV2.3.1 built-in function cvCreateCameraCapture () re-uses cvQueryFrame () function and captures a two field picture, use cvCvtColor () function that the photo of shooting is changed into to gray level image, through level and smooth (cvSmooth function), after Extract contour (extractContourImg function), by greyscale image transitions, be bianry image (cvThreshold function) again, calculate the number percent that black pixel (barrier) accounts for the total pixel of picture, if result of calculation is greater than threshold value (determining after many experiments that 6% is comparatively suitable), there is barrier in judgement, the output obstacle information, mechanism is forbidden in activation, all instructions of sending to drive motor module all are dropped, forbid carrying out.
Fig. 8 forbids mechanism works figure.Forbid that machine-processed effect is that inhibit command is carried out.If high-rise module does not have output information, forbid machine-processed un-activation, present instruction can be sent into execution.If high-rise module output information, forbid that mechanism activates, present instruction is dropped, and forbids carrying out.
In the robot control system shown in Fig. 1, forbid that mechanism one co-exists in a place.When obstacle detection module output obstacle information, all instructions of sending to drive motor module all are dropped, and forbid carrying out.
The invention provides a kind of many scenes of all-purpose robot control method based on the containment type hierarchical structure; method and the approach of this technical scheme of specific implementation are a lot; the above is only the preferred embodiment of the present invention; should be understood that; for those skilled in the art; under the premise without departing from the principles of the invention, can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.In the present embodiment not clear and definite each ingredient all available prior art realized.

Claims (6)

1. many scenes of the all-purpose robot control method based on the containment type hierarchical structure, is characterized in that, comprises the steps:
(1) the people's platform that starts the machine, start external or Embedded computing machine, and camera, microphone are connected to computing machine;
(2) by the computing machine access network, and set up socket as server end and client and be connected;
(3) create four instruction queues, for depositing the steering order of keyboard, microphone, socket and mail;
(4) define the priority of four instruction queues, be followed successively by from high to low the steering order of keyboard, microphone, socket and mail;
(5) start four control modules for receiving instruction;
(6) four control modules are by the instruction the received corresponding instruction queue of packing into;
(7) take out the highest instruction of priority from four instruction queues;
(8) robot platform recognition instruction;
(9) the robot platform control is carried out instruction.
2. a kind of many scenes of all-purpose robot control method based on the containment type hierarchical structure according to claim 1, is characterized in that, step (8) is further comprising the steps:
(81) if the instruction None-identified provides prompting and abandons instruction at the control interface;
(82) if instruction can be identified, enter step (9).
3. a kind of many scenes of all-purpose robot control method based on the containment type hierarchical structure according to claim 2, is characterized in that, wherein step (9) is further comprising the steps:
(91) if instruction can't be carried out, at the control interface, provide prompting and abandon instruction;
(92) if instruction can be carried out, execution is returned to step (6) after finishing.
4. a kind of many scenes of all-purpose robot control method based on the containment type hierarchical structure according to claim 1, it is characterized in that, take out the highest instruction of priority and completed by priority mechanism, the priority of same instruction queue inside determines by time sequencing, and the instruction priority of first joining the team is high.
5. a kind of many scenes of all-purpose robot control method based on the containment type hierarchical structure according to claim 1, it is characterized in that, four control modules refer to respectively: keyboard receiver module, sound identification module, socket Socket communication module, mail reception module.
6. a kind of many scenes of all-purpose robot control method based on the containment type hierarchical structure according to claim 1, is characterized in that, the Starting mode of control module is: with form the concurrent or executed in parallel of separate threads.
CN201310461873.8A 2013-09-30 2013-09-30 A kind of many scenes of all-purpose robot control method based on containment type hierarchical structure Active CN103499972B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310461873.8A CN103499972B (en) 2013-09-30 2013-09-30 A kind of many scenes of all-purpose robot control method based on containment type hierarchical structure

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310461873.8A CN103499972B (en) 2013-09-30 2013-09-30 A kind of many scenes of all-purpose robot control method based on containment type hierarchical structure

Publications (2)

Publication Number Publication Date
CN103499972A true CN103499972A (en) 2014-01-08
CN103499972B CN103499972B (en) 2016-05-25

Family

ID=49865192

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310461873.8A Active CN103499972B (en) 2013-09-30 2013-09-30 A kind of many scenes of all-purpose robot control method based on containment type hierarchical structure

Country Status (1)

Country Link
CN (1) CN103499972B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106569613A (en) * 2016-11-14 2017-04-19 中国电子科技集团公司第二十八研究所 Multi-modal man-machine interaction system and control method thereof
CN108304714A (en) * 2017-01-13 2018-07-20 埃森哲环球解决方案有限公司 The integrated robotics of goal systems and access manage
CN109976338A (en) * 2019-03-14 2019-07-05 山东大学 A kind of multi-modal quadruped robot man-machine interactive system and method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050103767A1 (en) * 2001-01-25 2005-05-19 Lincoln Global, Inc. System and method providing automated welding notification
US20070078566A1 (en) * 2005-09-30 2007-04-05 Yulun Wang Multi-camera mobile teleconferencing platform
CN201237738Y (en) * 2008-08-01 2009-05-13 邓伟雄 Golf cart with intelligent automatic searching and tracing function
CN103293968A (en) * 2012-02-29 2013-09-11 陈瑞 Intelligent robot system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050103767A1 (en) * 2001-01-25 2005-05-19 Lincoln Global, Inc. System and method providing automated welding notification
US20070078566A1 (en) * 2005-09-30 2007-04-05 Yulun Wang Multi-camera mobile teleconferencing platform
CN201237738Y (en) * 2008-08-01 2009-05-13 邓伟雄 Golf cart with intelligent automatic searching and tracing function
CN103293968A (en) * 2012-02-29 2013-09-11 陈瑞 Intelligent robot system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
李静等: "仿真机器人足球赛中团队协作方法的研究", 《计算机科学》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106569613A (en) * 2016-11-14 2017-04-19 中国电子科技集团公司第二十八研究所 Multi-modal man-machine interaction system and control method thereof
CN108304714A (en) * 2017-01-13 2018-07-20 埃森哲环球解决方案有限公司 The integrated robotics of goal systems and access manage
CN108304714B (en) * 2017-01-13 2021-10-26 埃森哲环球解决方案有限公司 Access management system, access management robot promotion system, and access management method
CN109976338A (en) * 2019-03-14 2019-07-05 山东大学 A kind of multi-modal quadruped robot man-machine interactive system and method

Also Published As

Publication number Publication date
CN103499972B (en) 2016-05-25

Similar Documents

Publication Publication Date Title
CN105810003B (en) Bicycle parking and pick-up control system for parking lot and method
CN103646565B (en) Positional information storage and the finding method thereof of car Quick Response Code sought by a kind of vehicle based on wechat
CN108664957A (en) Number-plate number matching process and device, character information matching process and device
EP1672510A3 (en) Fast platform independent inter-process communication
CN106250369A (en) voice interactive method, device and terminal
CN103916513A (en) Method and device for recording communication message at communication terminal
US20200118569A1 (en) Conference sound box and conference recording method, apparatus, system and computer storage medium
CN104538034A (en) Voice recognition method and system
CN107402754A (en) Information processing and model training method, device, electronic equipment and storage medium
CN103067557A (en) Method, device and mobile terminal for showing information
JP2010532132A (en) Automatic contact information input by position detection
CN106485476A (en) A kind of staff attendance system based on video
CN112597286A (en) Case dispatching handling method and device, electronic equipment and storage medium
CN103499972A (en) Subsumption hierarchical structure-based universal robot multi-scene control method
CN106782604A (en) The processing method of speech message, device and its electronic equipment
CN102811288B (en) Method and device for recording call information
CN102810184A (en) Method and device for dynamically executing workflow and enterprise system
CN106095051A (en) The processing method of a kind of application program for mobile terminal, device and mobile terminal
Strickland AI-human partnerships tackle" fake news": Machine learning can get you only so far-then human judgment is required-[News]
CN107122120A (en) A kind of processing method of dummy keyboard, device and terminal
CN107665592A (en) A kind of intelligent parking lot reverse vehicle searching system
US9807214B2 (en) Temporary contacts
CN107623620B (en) Processing method of random interaction data, network server and intelligent dialogue system
CN112132015A (en) Detection method, device, medium and electronic equipment for illegal driving posture
CN101656803A (en) Operator position system capable of recognizing voices and voice recognition method thereof

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