Interactive VR sand tables show robot
Technical field
The present invention relates to field in intelligent robotics, are related to domestic robot field and VR field of virtual reality.
Background technology
Robot was mainly divided into for 3 generations by people, and the robot of the first generation is straighforward operation machine, cannot leave the control of people
Move alone.The robot of the second generation is controlled robot by the program finished in advance, it is made to be repeatedly performed certain automatically
The operation of kind mode.Third generation robot rises to have from this and can perceive, identify, judging, learning and the function of emotion expression service, it can
Action intelligence is combined, therefore referred to as intelligent robot.
VR virtual realities are the emerging in recent years technologies with wide application prospect.Currently, virtual reality technology is
It is widely used in amusement game, architectural design, the fields such as medical treatment.Virtual reality is the modern times used using computer technology as core
The vision true to nature that advanced technology generates, the integrated virtual environment such as sense of hearing, user can be set by necessary input and output
The standby object with virtual environment interacts, and influences each other, and then obtains impression and experience on the spot in person.It is this by calculating
The virtual environment that machine generates can be the reproduction in a certain particular objective world.Advanced PCs VR is due to there is high performance video cards host
Support, the experience effect of immersion can be described as very outstanding, but can not carried in real time for client in the prior art
For explaining and showing the robot of sand table.
Invention content
1, the purpose of the present invention.
The present invention explains sand table for client in real time in order to realize, interaction is carried out with client, and provides one
Interactive VR sand tables show robot.
2, the technical solution adopted in the present invention.
A kind of interactive mode VR sand table show robots proposed by the present invention, including:
Voice reads locating module, obtains the direction that target speaker is determined by auditory localization, the sky of microphone array
Domain filters, and the angle to waking up people positions, and forms orientation pickup wave beam, and inhibit to the noise other than wave beam, is promoted remote
Field pickup quality ensures recognition effect, and carries out echo cancellor;
Voice wake-up module is waken up robot from dormant state by keyword;
Authority recognition module acquires face by camera module and carries out authority recognition;
Obstacle identification module detects whether moving direction has obstacle by infrared sensor, and is judged by system,
Stop mobile or detours;
Emotional characteristics reflect module, and the design and color by switching the full-color glue drop light bar of LED goes to be shown different
Emotional characteristics;
Mobile execution module, the preset mobile execution module of routing instruction control acts or external voice passes through
In Android system after acquisition instructions, chassis action is controlled by socket communication for five wheat ring battle array of HKUST News;
Speech exchange module establishes the mapping of voice answer-back, by calling the language in cloud server or local data base
Answer is identified in the voice that sound data obtain outside;
Module is followed, drives mobile execution module to follow action by camera tracker's movement locus;
Power source charges position locating module obtains information about power, is obtained comprising charging station location letter according to laser sensor
The pulse signal of breath, master board find charge power supply according to pulse signal;
The instruction of VR display modules, the command information obtained by speech exchange module or input carries out VR displayings.
In further specific implementation mode, sand table interactive module establishes voice and building position mapping relations in sand table
It is stored in local data base, the speech exchange module obtains the building recommended information corresponding to speech sound eeplaining, in local number
According to the corresponding position information for the explanation building found in library in sand table, the phase of explanation building is transmitted by wired or wireless mode
The indicating module in location information to sand table is answered, the indicating module in sand table is driven to be acted.
In further specific implementation mode, the sand table to being stored in local data base or cloud server explains voice
Identification information is classified, and is divided into building and house type classification information, is established corresponding grammer packet respectively:
The sand table for receiving input explains voice signal, respectively to receiving in the local data base or cloud server
Voice signal carries out speech recognition lookup,
After can identify classification information from the voice signal for receive sand table, from pre-stored classification information
It transfers corresponding sand table and shows and execute instruction;
Sand table is shown to execute instruction and is sent to sand table indicating module and acts;
Such as the explanation speech recognition failure of this sand table, and according to this voice recognition processing failure cause, from preassigning
The corresponding mark of each voice recognition processing failure cause in, select this sand table explanation voice recognition processing failure cause pair
Voice recognition result of the mark answered as the grammer packet to the voice signal.
In further specific implementation mode, the sand table, which shows to execute instruction, is sent to sand table indicating module into action
Corresponding building or house type are shown as by VR display modules.
In further specific implementation mode, the camera module uses 120 frame USB camera modules.
In further specific implementation mode, the voice wake-up module, speech exchange module pass through HKUST News
XFM10521 is realized.
3, beneficial effects of the present invention.
Present device is the completely new intelligent robot being combined with high end main frame VR equipment, and self-assisted voice may be implemented and say
Building is solved, and by speech identifying function, corresponding building or house type on sand table are shown, real-time interactive corresponds to,
Make house purchaser's more intuitive understanding building design conditions, promotes the experience of building viewing.
Description of the drawings
Fig. 1 is the building-block of logic of robot of the present invention.
Fig. 2 is the overall structure figure of robot of the present invention.
Fig. 3 is the detailed accessory introduction figure of the present invention.
Figure of description 2 is the overall structure figure of robot of the present invention, and the part respectively indicated in figure is disassembled:Head
A, body part B, chassis portion C.
Figure of description 3 is the whole accessory figure in detail of robot, and the component respectively indicated in figure is respectively:1. display PC
2. 3. 4. 5. 6. the full-color glue of LED drips lamp to mainboard to speech recognition interactive module to camera module to interconnecting module for display and touch screen
7. loudspeaker are 8. high with video card 9. external interface plate 10. high dispensing computer mainboard etc. for bandEquipment lockerChassis return pulleyBottom
Disk control moduleRobot external power supply and switchInfrared induction sensor.
Specific implementation mode
Embodiment
As shown in Figure 1, a kind of interactive mode VR sand table show robots proposed by the present invention, including:
Voice reads locating module, obtains the direction that target speaker is determined by auditory localization, the sky of microphone array
Domain filters, and the angle to waking up people positions, and forms orientation pickup wave beam, and inhibit to the noise other than wave beam, is promoted remote
Field pickup quality ensures recognition effect, and carries out echo cancellor;
Voice wake-up module is waken up robot from dormant state by keyword;
Authority recognition module acquires face by camera module and carries out authority recognition;
Obstacle identification module detects whether moving direction has obstacle by infrared sensor, and is judged by system,
Stop mobile or detours;
Emotional characteristics reflect module, and the design and color by switching the full-color glue drop light bar of LED goes to be shown different
Emotional characteristics;
Mobile execution module, the preset mobile execution module of routing instruction control acts or external voice passes through
In Android system after acquisition instructions, chassis action is controlled by socket communication for five wheat ring battle array of HKUST News;
Speech exchange module establishes the mapping of voice answer-back, by calling the language in cloud server or local data base
Answer is identified in the voice that sound data obtain outside;
Module is followed, drives mobile execution module to follow action by camera tracker's movement locus;
Power source charges position locating module obtains information about power, is obtained comprising charging station location letter according to laser sensor
The pulse signal of breath, master board find charge power supply according to pulse signal;
The instruction of VR display modules, the command information obtained by speech exchange module or input carries out VR displayings.
In further specific implementation mode, sand table interactive module establishes voice and building position mapping relations in sand table
It is stored in local data base, the speech exchange module obtains the building recommended information corresponding to speech sound eeplaining, in local number
According to the corresponding position information for the explanation building found in library in sand table, the phase of explanation building is transmitted by wired or wireless mode
The indicating module in location information to sand table is answered, the indicating module in sand table is driven to be acted.
In further specific implementation mode, the sand table to being stored in local data base or cloud server explains voice
Identification information is classified, and is divided into building and house type classification information, is established corresponding grammer packet respectively:
The sand table for receiving input explains voice signal, respectively to receiving in the local data base or cloud server
Voice signal carries out speech recognition lookup,
After can identify classification information from the voice signal for receive sand table, from pre-stored classification information
It transfers corresponding sand table and shows and execute instruction;
Sand table is shown to execute instruction and is sent to sand table indicating module and acts;
Such as the explanation speech recognition failure of this sand table, and according to this voice recognition processing failure cause, from preassigning
The corresponding mark of each voice recognition processing failure cause in, select this sand table explanation voice recognition processing failure cause pair
Voice recognition result of the mark answered as the grammer packet to the voice signal.
In further specific implementation mode, the sand table, which shows to execute instruction, is sent to sand table indicating module into action
Corresponding building or house type are shown as by VR display modules.
In further specific implementation mode, the camera module uses 120 frame USB camera modules.
In further specific implementation mode, the voice wake-up module, speech exchange module pass through HKUST News
XFM10521 is realized.
As shown in Figures 2 and 3, the intelligent robot of interactive VR includes head A, body part B, chassis portion C;Head
A includes display PC interconnecting modules 1, display and touch screen 2, camera module 3, speech recognition interactive module 4, mainboard 5;Body
Body portion B includes loudspeaker 7, height with video card 8, external interface 9, high dispensing computer mainboard 10, VR equipment locker 11, chassis portion C
Including chassis return pulley 12, chassis control module 13;Display and touch screen 2 are mounted on the top on head, and camera module 3 is in head
The top in portion is mounted laterally, and lower end installation display PC interconnecting modules 1, the mainboard 5 of display and touch screen 2 are mounted on camera
The lower end of module 3;Height is located at the middle part of body part B with video card 8, and loudspeaker 7 are located at the top of body, high dispensing computer mainboard 10 and
Height is connected with video card 8;Chassis control module 13 is connected with chassis return pulley 12 drives chassis return pulley moving in rotation.Speech recognition interacts
Module 4 uses HKUST News XFM10521 modules.Mainboard 5 uses the complete eight core mainboards of will A83T of Android.Head A further includes LED complete
Color glue drop light bar 6 shows different colours and pattern.Further include that infrared induction sensor 15 is detected mounted on the front end of chassis portion C
Surroundings outside is connected with chassis control module 13.The loudspeaker 7 are 3 cun of Full-frequency trumpets.It further include the external electricity of robot
Source and switch 14 are mounted on body part B or chassis portion C.
Voice read module uses HKUST News XFM10521 modules, carries 4+1 multi-microphones, realizes 360 ° of voice signals
Acquisition, and the direction of target speaker can be determined by auditory localization, module utilizes the airspace filter characteristic of microphone array,
By being positioned to the angle for waking up people, orientation pickup wave beam is formed, and inhibit to the noise other than wave beam, promote far field and pick up
Sound quality ensures recognition effect, and carries out echo cancellor.By using the five rings battle array retrievable voice messaging of wheat, far field is picked up
Sound Technology application far field identifies and noise reduction technology, and pickup distance is made to reach 5 meters;Voice awakening technology, can be by keyword by mould
Block wakes up from dormant state;Collect the intelligent sound chat feature of cloud server data support and customizable;Image is known
Do not use and in terms of Emotion identification 120 frame USB camera modules, progress real-time photography acquisition that can also pass through by this equipment and be
System locking follows owner;Whether obstacle function has obstacle using infrared sensor detection moving direction, and is made and being sentenced by system
It is disconnected, stop movement.Present invention uses the full-color glue of LED to drip light bar for emotional characteristics, can carry out design and color and go to be shown not
Same effect.Instruction control is by five wheat ring battle array of HKUST News, logical by socket in Android system after acquisition instructions
The action such as news control chassis.
Carrying out practically flow is in Fig. 3:It is opened by robot external power supply and switch after robot booting, is then passed through
The camera of 120 frame USB camera modules carries out head portrait acquisition, enters corresponding user by carrying out authority distribution after identification
In, it can start voice wake-up intelligent sound function, while the mould in HKUST News XFM10521 modules by saying keyword
Block may determine that owner, position transformation direction simultaneously coordinate after openingThe control modules such as chassis steering gear electric-machine have automatically with
With owner's function, colleague bottom, which has infrared induction sensor, can carry out obstacle avoidance, stop moving ahead.3. 120 frame USB simultaneously
Camera module can be in Android in conjunction with the voice collecting of HKUST News XFM10521 modules by acquiring user's face expression
It is judged in the system of complete eight core mainboards of will A83T, real-time judge goes out the expression of owner, with semantic understanding, passes through 3 cun of full ranges
Loudspeaker make the expression of feedback language, and can drip light bar by the full-color glue of LED and make the expression feedback of oneself, and displaying mood is special
Sign, while can also be judged by system, the phonetic order to come for acquisition can also make the actions such as corresponding action.By red
Outer inductive pick-up.Electricity it is low or instruction when can automatic decision cradle seat find power source charges.It is put in VR equipment lockers
For VR equipment after powering by operating 2. display and touch screen, recalling VR programs can be practical, while can also be by 9.
External interface plate go it is aobvious outside other external displays or projecting apparatus, also can the input equipments such as external access of mouse, keyboard, game paddle
Preferably to experience VR.
This money can experience the intelligent robot of VR in combination with current mainstream VR equipment, in conjunction with the smooth video card master of Thailand top at present
Machine, the experience fluency to ensure VR virtual realities carry out relevant interface in inside and carry out portable disposable wiring in experience,
Refuse cumbersome butted line.It is carried with excessive bulk-breaking.It can allow the arrangement VR equipment that user is more convenient, and carry VR
Equipment, using VR, while as a robot, language is most natural, most convenient the means of Human communication, therefore also inevitable
As one of most natural, most convenient means are exchanged between people and computer, robot can carry out free friendship with our mankind
It talks, also having thinking, learning ability, also more tool is interesting and recreational compared to host in the past, is brought more for our life
Convenient and enjoyment is more rich and varied.