CN108053824A - A kind of patrol rescue unmanned plane and its method of work based on voice signal - Google Patents

A kind of patrol rescue unmanned plane and its method of work based on voice signal Download PDF

Info

Publication number
CN108053824A
CN108053824A CN201711307151.1A CN201711307151A CN108053824A CN 108053824 A CN108053824 A CN 108053824A CN 201711307151 A CN201711307151 A CN 201711307151A CN 108053824 A CN108053824 A CN 108053824A
Authority
CN
China
Prior art keywords
module
voice
unmanned plane
signal
processor
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.)
Pending
Application number
CN201711307151.1A
Other languages
Chinese (zh)
Inventor
王泓博
朱昌平
汪先成贤
缪立波
邱俊
刘志远
吕雪芬
王飞
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.)
HEHAI UNIVERSITY WENTIAN COLLEGE
Original Assignee
HEHAI UNIVERSITY WENTIAN COLLEGE
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 HEHAI UNIVERSITY WENTIAN COLLEGE filed Critical HEHAI UNIVERSITY WENTIAN COLLEGE
Priority to CN201711307151.1A priority Critical patent/CN108053824A/en
Publication of CN108053824A publication Critical patent/CN108053824A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00Speech recognition
    • G10L15/08Speech classification or search
    • G10L15/18Speech classification or search using natural language modelling
    • G10L15/1815Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S5/00Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
    • G01S5/18Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using ultrasonic, sonic, or infrasonic waves
    • G01S5/22Position of source determined by co-ordinating a plurality of position lines defined by path-difference measurements
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/12Target-seeking control
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10KSOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
    • G10K11/00Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
    • G10K11/16Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L25/00Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/48Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
    • G10L25/51Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00Television systems
    • H04N7/18Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/183Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a single remote source
    • H04N7/185Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a single remote source from a mobile camera, e.g. for remote control

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Acoustics & Sound (AREA)
  • Health & Medical Sciences (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Signal Processing (AREA)
  • Computational Linguistics (AREA)
  • General Physics & Mathematics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Automation & Control Theory (AREA)
  • Artificial Intelligence (AREA)
  • Alarm Systems (AREA)

Abstract

The invention discloses a kind of patrol rescue unmanned planes and its method of work based on voice signal, belong to unmanned plane technical field of security and protection.The noise reduction cover of patrol rescue unmanned plane is mounted below unmanned plane housing, wireless data transfer module, real-time locating module, flying power module are placed in unmanned plane enclosure interior, acoustical signal of seeking help identification module is mounted in noise reduction cover, wireless data transfer module, real-time locating module, acoustical signal of seeking help identification module, flying power module are electrically connected respectively with processor, and wireless data transfer module is used to communicate with the PC end datas of unmanned plane and security control center;Weak acoustic signal is analyzed and processed by voice identification module, acoustical signal coordinate calculates to seeking help.The present invention can work in the environment of the visions such as the region with a varied topography such as mountainous region, forest land and night, haze, cloudy day have obstacle, find person for help position in time, notify security control center staff rescue in time.

Description

A kind of patrol rescue unmanned plane and its method of work based on voice signal
Technical field
The invention belongs to unmanned plane technical field of security and protection, and in particular to a kind of patrol based on voice signal rescues nobody Machine, voice localization method and method of work.
Background technology
The many scenic spots in China due to taking up a large area, it is with a varied topography, staff can not will patrol and salvaging accomplish Comprehensive inspection, running into some emergencies can not implement to rescue in time.The security protection measure at scenic spot at present is by staff mostly Walking is gone on patrol, and work efficiency is low, and when emergency occurs, staff can not go and find out what's going on rapidly and take rescue in time Measure;And employ the expense of patrol staff also higher.Therefore, because unmanned plane miniaturization, quick landing and without airborne The features such as personnel, has obtained quick development and has been widely applied in the wild in search field.Existing unmanned plane search system, Mostly taken photo by plane using airborne high-definition camera, infrared imaging, image identification etc. equipment as way of search, search effect is by four chow rings Border, accuracy of instrument, formula are affected, but unmanned plane is using taking photo by plane or infrared imaging is there are vision dead in patrol is searched and rescued Angle, search efficiency are low.
It is 201710001046.9 through retrieval, such as Chinese Patent Application No., application is public to solve the technical issues of above-mentioned Cloth day discloses a kind of rescue system for the patent application document on March 29th, 2017, which includes:It can be by using The wearable device of family body worn, the wearable device have sign probe unit and/or environment detection unit, sign probe unit To obtain user's vital sign information, environment detection unit is obtaining environmental information;Be arranged in wearable device or with Rescue identifying unit in the associated server of wearable device, to be obtained according to sign probe unit and/or environment detection unit The information taken judges to wear whether user needs to rescue, and sends rescue signal in the case where judging to need;Unmanned plane is rescued Device in response to foregoing rescue signal, flies to wearing user and is rescued.The rescue system needs to coordinate wearable device that could reach To rescue function, many times, accident is accident, can not it is anticipated that and be easily lost in the wild, using being limited System.
For another example Chinese Patent Application No. is 201610047601.7, and data of publication of application is the patent Shen on June 29th, 2016 Please file disclose a kind of unmanned plane rescue system and its rescue mode, system includes unmanned plane, for collection site video Environment information acquisition module, rescue execution module, control process module and rescue method memory module;Control process module includes For searching for the scene Recognition list of the wounded by the progress image scene identification of secondary cascade model from the live video of acquisition Member is found from rescue method memory module and is rescued with the matching unit of the matched rescue method of scene determination result and control The control unit that execution module succours the wounded.The unmanned plane rescue system is still combined with image recognition technology to realize On the one hand field rescue can not be broken away from during patrol is searched and rescued there are visual dead angle, search efficiency is low;On the other hand this rescue is base In modular processing, the desired result of people's rescue is unable to reach.
The content of the invention
1. it to solve the problems, such as
There are visual dead angle in being gone on patrol for existing unmanned plane, Quick rescue can not be carried out and treat that rescue personnel can not The problem of carrying rescue auxiliary device, the present invention provides a kind of combination voice signal, can position person for help position, can be The rescue unmanned plane of the patrol based on voice signal and its work side of intercommunion platform are gone on patrol, built under visual environment harsh conditions Method,.
2. technical solution
To solve the above-mentioned problems, the technical solution adopted in the present invention is as follows:
The rescue unmanned plane of the patrol based on voice signal of the present invention includes processor, wireless data transfer module and seeks help Acoustical signal identification module, the wireless data transfer module, acoustical signal of seeking help identification module are electrically connected with processor, sound of seeking help letter Number identification module feeds back to processor by picking up emergency sound, and processor passes through wireless data transfer module and security control The PC end datas communication at center.
In a kind of possible embodiment of the present invention, the acoustical signal identification module of seeking help is received including weak acoustic signal Device, voice locating module, voice identification module and semantics recognition module, these modules integrate an entirety, wherein faint sound is believed Number receiver receives the weak acoustic signal below unmanned plane, and weak acoustic signal is transmitted to voice identification module, by amplifying, filtering The emergency sound in preliminary identification weak acoustic signal is carried out, existing voice recognition technology may be employed in the present invention, and emergency sound is by language After adopted identification module recognition and verification, voice locating module is positioned by real-time voice, and definite emergency sound orientation is transmitted to place Manage device.
In a kind of possible embodiment of the present invention, which further includes unmanned plane housing, in real time noise reduction cover, positioning Module and flying power module;The noise reduction cover is mounted below unmanned plane housing, wireless data transfer module, real-time positioning mould Block, flying power module are placed in unmanned plane enclosure interior, and acoustical signal of seeking help identification module is mounted in noise reduction cover, real-time positioning mould Block, flying power module are electrically connected respectively with processor, real-time locating module for determine unmanned plane real time position and by nobody The relative position of machine is positioned and is transmitted to processor.
In a kind of possible embodiment of the present invention, the processor uses Artik5, supports that Wi-Fi, low-power consumption are blue Tooth supports 802.11b/g/n wireless protocols.
In a kind of possible embodiment of the present invention, the wireless data transfer module uses IEEE802.11b, uses The ISM band of 2.4GHz is modulated demodulation using direct sequence spreading DSSS technologies;The real-time locating module, using north Struggle against satellite navigation system, and establishes correspondence with the system.
In a kind of possible embodiment of the present invention, the voice locating module uses MSP near fields moving sound status System including two panels MSP430F149 processing chips, builds sound circuit and detection circuit, and sound circuit uses two panels NE555 Chip self-oscillation driving buzzer transmitting voice signal, detection circuit form array received by least three microphone sound pick-ups With processing voice signal, pass through CC1101 wireless communication modes between circuit.It is used to receiving and handling sound in the voice locating module The microphone sound pick-up array of sound signal is determined using a kind of three stage voices of the time delay estimation thought based on voice signal Position formula is removed the potential noise in voice signal by a wave filter, obtains voice as pure as possible first;Then Time delay estimation is done by a GCC-PHAT formula;The last spatial position for building geometrical relationship estimation voice again.
In a kind of possible embodiment of the present invention, the voice identification module is one mode identifying system, including Feature extraction, pattern match, three, reference model storehouse elementary cell.
In a kind of possible embodiment of the present invention, the flying power module includes 4108 external rotor brushless power electrics Machine (KV480), poly-lithium battery (15C 20000mAh 14.8V) and brushless electricity adjust brushless electricity to adjust and are mounted on polymer lithium electricity Between pond, 4108 external rotor brushless power motors, the work of brushless electricity regulation and control 4108 external rotor brushless power motors of system, polymer Lithium battery provides power supply for 4108 external rotor brushless power motors.
In a kind of possible embodiment of the present invention, the camera module for real time image collection is further included, mould of photographing Block is electrically connected with processor.
In a kind of possible embodiment of the present invention, the communication module being made of loud speaker and microphone is further included, is led to Letter module is electrically connected with processor.
In a kind of possible embodiment of the present invention, highlighted lamp source is further included, highlighted lamp source is electrically connected with processor.
On the basis of the technical solution of above-mentioned patrol rescue unmanned plane, a kind of patrol based on voice signal rescues nobody The voice localization method of machine, including specific steps:It is removed first by the wave filter of a processor potential in voice signal Noise obtains voice as pure as possible;Then time delay estimation is done by a GCC-PHAT formula;Finally geometry is built again Relation estimates the spatial position of voice;Specifically GCC-PHAT formula are::Voice reaches microphone sound pick-up i and microphone picks up The range difference of sound device j=voice reaches the distance of distance-voice arrival microphone sound pick-up j of microphone sound pick-up i, i.e.,:
Definition voice reaches microphone sound pick-up i and the range difference of arrival microphone sound pick-up j is:
di,j=Di-Dj=c@tij (1)
Wherein tI, jRepresent that voice signal reaches the time difference of microphone sound pick-up i and microphone sound pick-up j, c is the velocity of sound;
The center that No. 0 microphone sound pick-up is array is made, then formula chooses it as reference, can then extend formula (1) For:
By N-1 equation merger into matrix form:
On the basis of the technical solution of above-mentioned patrol rescue unmanned plane, a kind of patrol based on voice signal rescues nobody The method of work of machine, including step in detail below:
Step S101:The weak acoustic signal receiver of patrol rescue unmanned plane receives weak acoustic signal;
Step S102:Voice identification module judges whether received signal is voice;
Step S103:If voice identification module judges signal for voice, semantics recognition module is started to work, and judges signal Whether it is distress signal;
Step S104:If signal is distress signal, voice locating module carries out the relative position of signal and unmanned plane It positions and this is transmitted to processor;
Step S105:Locating module positions the real time position of unmanned plane, and the real time position of unmanned plane is transmitted to processor;
Step S106:The real time position of the relative position and unmanned plane of signal and unmanned plane is used above-mentioned voice by processor Localization method carries out comprehensive analysis, and the coordinate of the signal analyzed is sent to security control center, and in security control It reminds at the PC ends of the heart;
Step S107:Processor controls flying power module, and unmanned plane is with flying to sound source;
Step S108:When flying to sound source, again perform step S101 to step S103, if distress signal is confirmed as to The PC ends of security control center send alarm;
Step S109:Open image capture module and by the PC ends of image scene real-time Transmission to security control center.
In a kind of possible embodiment of the present invention, a kind of work side of the patrol rescue unmanned plane based on voice signal Method, it is further comprising the steps of:
Step S110:The PC ends of security control center are by checking the live realtime graphic of unmanned plane transmission to field condition Judged and take corresponding rescue measure;
Step S111:If scene there are the objects such as trees to block, surface state is not seen, then unmanned plane hovering, security control Interim call mechanism is established in the PC ends at center by communication module and person for help;
Step S112:If person for help is hindered by animal, the PC ends of security control center pass through the loud speaker in communication module The audio for dispersing animal is played, animal is dispersed;
Step S113:If being rescued at night, person for help's specific location is indicated for rescue personnel using highlighted lamp source.
3. advantageous effect
Compared with the prior art, beneficial effects of the present invention are:
(1) the rescue unmanned plane of the patrol based on voice signal of the invention, sets noise reduction cover below unmanned plane housing, subtracts It makes an uproar and is covered with the influence that the noise and sound of the wind that unmanned plane is sent above effect decrease gather acoustical signal, pass through acoustical signal of seeking help and identify Module can carry out work in the environment of the visions such as the region with a varied topography such as mountainous region, forest land and night, haze, cloudy day have obstacle Make, find person for help position in time, notify security control center staff rescue in time, it is with a varied topography especially for scenic spot And some hidden regions, the energy of the tourism accident rescue of scenic spot management department can be strengthened by accurately, timely detecting Power, it is ensured that people's security of the lives and property;
(2) the rescue unmanned plane of the patrol based on voice signal of the invention, by adding taking the photograph for real time image collection The situation picture for waiting for rescue people is sent to the PC ends of security control center, is arranged convenient for making suing and labouring in time by shadow module in time It applies;
(3) the rescue unmanned plane of the patrol based on voice signal of the invention, the real-time locating module used use OptiTrack realizes picture catching and is accurately positioned, frame rate:100hz captures unmanned plane with the capture rate of hundreds of frames per second Upper fixed special mark point, can accurately construct mark point three-dimensional space position information, realize unmanned plane in real time Position and the capture of attitude information;
(4) the rescue unmanned plane of the patrol based on voice signal of the invention, by adding communication module, if sound source have barrier Object is hindered to block, hinders unmanned plane investigation surface state, the staff at PC ends is established by UAV Communication module and person for help Interim call mechanism, understands person for help's situation, conveys effective rescue information;
(5) the rescue unmanned plane of the patrol based on voice signal of the invention, by adding communication module, if person for help is Object is hindered, and PC can play the audio for dispersing animal by the loud speaker in communication module, and animal is dispersed;
(6) the rescue unmanned plane of the patrol based on voice signal of the invention, by adding highlighted lamp source, if being carried out at night Highlighted lamp source can be used to indicate person for help's specific location for rescue personnel for rescue, this patrol rescue unmanned plane;
(7) the voice localization method of the patrol rescue unmanned plane based on voice signal of the invention, using at least eight wheats Gram wind, can introduce more parameters by introducing more microphones, so as to reduce evaluated error, improve the essence of positioning Degree;
(8) the rescue unmanned plane of the patrol based on voice signal of the invention is simple in structure, and design is reasonable, easily fabricated.
Description of the drawings
Technical scheme is described in further detail below with reference to drawings and examples, but should To know, these attached drawings only design for task of explanation, therefore not as the restriction of the scope of the invention.In addition, except non-specifically It points out, these attached drawings are meant only to conceptually illustrate structure construction described herein, without to be drawn to scale.
Fig. 1 is the structure diagram for the patrol rescue unmanned plane that the embodiment of the present invention is positioned based on voice;
Fig. 2 is the acoustical signal identification module schematic diagram of seeking help of the embodiment of the present invention;
Fig. 3 is the structure diagram of the noise reduction cover of the embodiment of the present invention.
Description of symbols in figure:
1st, real-time locating module;2nd, flying power module;3rd, processor;4th, camera module;5th, wireless data transfer module; 6th, lamp source is highlighted;7th, communication module;8th, acoustical signal of seeking help identification module;9th, weak acoustic signal receiver;10th, voice identification mould Block;11st, semantics recognition module;12nd, voice locating module;13rd, noise reduction cover.
Specific embodiment
Hereafter to the detailed description of exemplary embodiment of the present invention with reference to attached drawing, which forms one of description Point, in the figure the enforceable exemplary embodiment of the present invention is shown as example.Although these exemplary embodiment quilts Fully describe in detail so that those skilled in the art can implement the present invention, it is to be understood that can realize other embodiment and Can without departing from the spirit and scope of the present invention to the present invention various changes can be made.Hereafter to the embodiment of the present invention More detailed description is not limited to required the scope of the present invention, and just to be illustrated and do not limit pair The description of the features of the present invention and feature to propose to perform the best mode of the present invention, and is sufficient so that those skilled in the art It can implement the present invention.Therefore, the scope of the invention is only defined by the appended claims.
Hereafter detailed description of the present invention and example embodiment can be more fully understood with reference to attached drawing, wherein the present invention Element and feature are identified by reference numeral.
Embodiment 1
As shown in Figure 1 to Figure 3, in this embodiment, the patrol rescue unmanned plane based on voice signal includes unmanned plane Housing, noise reduction cover 13, processor 3, wireless data transfer module 5, real-time locating module 1, acoustical signal of seeking help identification module 8 and fly Action edge module 2, wireless data transfer module 5, real-time locating module 1, acoustical signal of seeking help identification module 8, flying power module 2 It is electrically connected respectively with processor 3;Noise reduction cover 13 is mounted below unmanned plane housing, wireless data transfer module 5, real-time positioning mould Block 1, flying power module 2 are placed in unmanned plane enclosure interior, and acoustical signal of seeking help identification module 8 is mounted in noise reduction cover 13.
The processor 3 of the present embodiment uses Artik5, supports Wi-Fi, low-power consumption bluetooth, supports that 802.11b/g/n is wireless Agreement.Wireless data transfer module 5 uses IEEE802.11b, using the ISM band of 2.4GHz, using direct sequence spreading DSSS technologies are modulated demodulation, for unmanned plane and security control center data communication;Real-time locating module 1, using north Struggle against satellite navigation system, and establishes correspondence with the system, for determining unmanned plane real time position.
The acoustical signal identification module 8 of seeking help of the present embodiment includes weak acoustic signal receiver 9, voice locating module 12, people Sound identification module 10 and semantics recognition module 11, voice locating module 12 are positioned using the real-time voice based on microphone array, Weak acoustic signal receiver 9 receives weak acoustic signal, and wherein weak acoustic signal receiver receives the faint sound letter below unmanned plane Number, weak acoustic signal is transmitted to voice identification module, by amplifying, filtering the emergency carried out in preliminary identification weak acoustic signal Existing voice recognition technology, such as method disclosed in patent CN201510977077.9 may be employed in sound, the present embodiment, emergency For sound by 11 recognition and verification of semantics recognition module, semantics recognition module 11, which is known, confirms distress signal, and voice is sent to by sensor Locating module 12 positions the relative position of voice signal and unmanned plane and is transferred to processor 3;
On the one hand the processor of the present embodiment is used to control each device of unmanned plane;On the other hand to acoustical signal coordinate of seeking help It is calculated, the specific location of final definite deliverer, convenient for the expansion of rescue;Flying power module 2 includes 4108 outer rotors Brushless power motor (KV480), poly-lithium battery (15C 20000mAh 14.8V) and brushless electricity are adjusted, and brushless electricity, which is adjusted, to be mounted on Between poly-lithium battery, 4108 external rotor brushless power motors, brushless electricity regulation and control 4108 external rotor brushless power motors of system Work, poly-lithium battery provide power supply for 4108 external rotor brushless power motors.
The voice locating module 12 of the present embodiment is using MSP near fields moving sound status system, including two panels MSP430F149 processing chips, build sound circuit and detection circuit, and sound circuit is driven using the self-oscillation of two panels NE555 chips Dynamic buzzer transmitting voice signal, detection circuit form array received and processing voice signal, electricity by eight microphone sound pick-ups Pass through CC1101 wireless communication modes between road.It is used to receive and handle the microphone array of voice signal in the voice locating module 12 Row pass through a filter first using a kind of three stage voice ranging formulas of the time delay estimation thought based on voice signal Potential noise in ripple device removal voice signal, obtains voice as pure as possible;Then by a GCC-PHAT formula come Do time delay estimation;The last spatial position for building geometrical relationship estimation voice again.
The voice identification module 10 of the present embodiment is one mode identifying system, and the structure of voice identification module 10 can adopt With presently disclosed technical solution, there is disclosure in Chinese patent CN201310429920.0, such as including feature extraction, pattern Three elementary cells such as matching, reference model storehouse, can be well realized the identification of voice.
With a varied topography for current scenic spot, it is accurate and clear to be collected for the traditional unmanned plane in some hidden regions Image, and traditional unmanned plane is difficult to effectively be worked in patrol by way of taking photo by plane.The present embodiment is based on voice The patrol unmanned plane of signal can have the ring of obstacle in the visions such as the region with a varied topography such as mountainous region, forest land and night, haze, cloudy day It works under border, the frequency of small-signal is 20Hz~1500Hz, can be with by unmanned plane patrol Technology application in tourist attractions By accurately, timely detecting, strengthen the ability of the tourism accident rescue of scenic spot management department, it is ensured that people's lives and properties Safety.
Embodiment 2
The substance of the present embodiment with embodiment 1, the difference is that:Further include the photography for real time image collection Module 4, the communication module 7 being made of loud speaker and microphone and highlighted lamp source 6, camera module 4, communication module 7 and high bright light Source 6 is electrically connected respectively with processor 3.
In actual use, after patrol rescue unmanned plane determines the position of person for help, controlled by camera module 4 for safety The PC ends transmission live view at center processed, it is then determined that rescue method or measure, if but scene has the objects such as trees to block, see Unclear surface state, then interim call is established in unmanned plane hovering, the PC ends of security control center by communication module 7 and person for help Mechanism;
In addition, if person for help is hindered by animal, the PC ends of security control center are played by the loud speaker in communication module 7 The audio of animal is dispersed, animal is dispersed.
Further, since rescue work may carry out at tangible night, therefore emergency is indicated for rescue personnel using highlighted lamp source 6 Person's specific location.
Embodiment 3
On the basis of embodiment 1, this patrol rescue unmanned plane can largely monitor emergency case in time, avoid Personnel death adjusts manually open rescue plan in time, substantially increases scenic spot patrol rescue ability safely.
A kind of method of work of the patrol rescue unmanned plane based on voice signal, including step in detail below:
Step S101:The weak acoustic signal receiver 9 of patrol rescue unmanned plane receives voice signal;
Step S102:Voice identification module 10 judges whether received signal is voice;
Step S103:If voice identification module 10 judges voice signal for voice, semantics recognition module 11 is started to work, Judge whether voice signal is distress signal;
Step S104:If voice signal is distress signal, voice locating module 12 is to the phase of voice signal and unmanned plane Position is positioned and this is transmitted to processor 3;
Step S105:Locating module positions the real time position of unmanned plane, and the real time position of unmanned plane is transmitted to processor 3;
Step S106:The real time position of the relative position and unmanned plane of voice signal and unmanned plane is used language by processor 3 Sound localization method carries out comprehensive analysis, and the coordinate of the voice signal analyzed is sent to security control center, and to safety It reminds at the PC ends of control centre;
Step S107:Processor 3 controls flying power module 2, and unmanned plane is with flying to sound source;
Step S108:When flying to sound source, again perform step S101 to step S103, if distress signal is confirmed as to The PC ends of security control center send alarm;
Step S109:Open image capture module and by the PC ends of image scene real-time Transmission to security control center;
Step S110:The PC ends of security control center are by checking the live realtime graphic of unmanned plane transmission to field condition Judged and take corresponding rescue measure;
Step S111:If scene there are the objects such as trees to block, surface state is not seen, then unmanned plane hovering, security control Interim call mechanism is established in the PC ends at center by communication module 7 and person for help;
Step S112:If person for help is hindered by animal, the PC ends of security control center pass through the loud speaker in communication module 7 The audio for dispersing animal is played, animal is dispersed;
Step S113:If being rescued at night, person for help's specific location is indicated for rescue personnel using highlighted lamp source 6.
Wherein, the voice localization method of the patrol rescue unmanned plane based on voice signal, specifically, by faint sound The processing of signal improves the precision of positioning.
The concrete structure of patrol rescue unmanned plane based on voice signal, can refer to foregoing teachings, not repeat to retouch herein It states.This method comprises the following specific steps that:It potential is made an uproar by what the wave filter of processor 3 removed in voice signal first Sound obtains voice as pure as possible, and the course of work of the wave filter is existing sound filter;Then a GCC- is passed through PHAT formula do time delay estimation;The last spatial position for building geometrical relationship estimation voice again;Specific GCC-PHAT formula For:The distance of the range difference of voice arrival microphone sound pick-up i and microphone sound pick-up j=voice arrival microphone sound pick-up i- Voice reaches the distance of microphone sound pick-up j, i.e.,:
Definition voice reaches microphone sound pick-up i and the range difference of arrival microphone sound pick-up j is:
di,j=Di-Dj=c@tij (1)
Wherein tI, jRepresent that voice signal reaches the time difference of microphone sound pick-up i and microphone sound pick-up j, c is the velocity of sound.
The center that No. 0 microphone sound pick-up is array is made, then formula chooses it as reference, can then extend formula (1) For:
By N-1 equation merger into matrix form:
1 symbol description of table
In formula (3), there are two r for free variablesAnd D0, wherein D0 can be by rsAnd r0It represents, so needs the change solved Amount just only has rsOne, therefore can be drawn from formula (3):If necessary to estimate the two-dimensional coordinate of voice, then including with reference to wheat Including gram wind sound pick-up, three microphone sound pick-ups are at least needed;And what is estimated if necessary is three-dimensional coordinate, then at least need Four microphone sound pick-ups are wanted, and this four microphone sound pick-ups cannot be on same straight line, by introducing more wheats Gram wind sound pick-up can introduce more parameters, so as to reduce evaluated error, improve the precision of positioning.
Schematically the present invention and embodiments thereof are described above, this describes no restricted, institute in attached drawing What is shown is also one of embodiments of the present invention, and actual structure is not limited thereto.So if common skill of this field Art personnel are enlightened by it, without departing from the spirit of the invention, are not inventively designed and the technical solution Similar frame mode and embodiment, are within the scope of protection of the invention.

Claims (10)

1. a kind of patrol rescue unmanned plane based on voice signal, which is characterized in that including processor (3), wireless data transmission Module (5) and acoustical signal identification module (8) of seeking help, the wireless data transfer module (5), acoustical signal of seeking help identification module (8) It being electrically connected with the processor (3), acoustical signal of seeking help identification module (8) feeds back to processor (3) by picking up emergency sound, Processor (3) is communicated by wireless data transfer module (5) with the PC end datas of security control center.
2. the patrol rescue unmanned plane according to claim 1 based on voice signal, which is characterized in that the sound letter of seeking help Number identification module (8) includes weak acoustic signal receiver (9), voice locating module (12), voice identification module (10) and semanteme Identification module (11), weak acoustic signal receiver (9) receive weak acoustic signal, and wherein weak acoustic signal receiver (9) receives nothing The weak acoustic signal of man-machine lower section, weak acoustic signal are transmitted to voice identification module (10), are tentatively known by amplifying, filtering Emergency sound in other weak acoustic signal, after emergency sound is by semantics recognition module (11) recognition and verification, voice locating module (12) is logical Real-time voice positioning is crossed, definite emergency sound orientation is transmitted to processor (3).
3. the patrol rescue unmanned plane according to claim 1 or 2 based on voice signal, which is characterized in that further include nothing Man-machine housing, noise reduction cover (13), real-time locating module (1) and flying power module (2);Noise reduction cover (13) is mounted on unmanned casing Below body, wireless data transfer module (5), real-time locating module (1), flying power module (2) are placed in unmanned plane enclosure interior, Acoustical signal of seeking help identification module (8) be mounted on noise reduction cover (13) on, real-time locating module (1), flying power module (2) respectively with Processor (3) is electrically connected, and real-time locating module (1) is used to determine unmanned plane real time position and carries out the relative position of unmanned plane It positions and is transmitted to processor (3).
4. the patrol rescue unmanned plane according to claim 3 based on voice signal, which is characterized in that the voice positioning Module (12) uses MSP near fields moving sound status system, including two panels MSP430F149 processing chips, structure sound circuit and Detection circuit, sound circuit is using two panels NE555 chips self-oscillation driving buzzer transmitting voice signal, and detection circuit is by extremely Few three microphone sound pick-ups form array received and processing voice signal, pass through CC1101 wireless communication modes between circuit.
5. the patrol rescue unmanned plane according to claim 4 based on voice signal, which is characterized in that the flying power Module (2) includes 4108 external rotor brushless power motors, poly-lithium battery and brushless electricity and adjusts, and brushless electricity, which is adjusted, is mounted on polymer Between lithium battery, 4108 external rotor brushless power motors, the work of brushless electricity regulation and control 4108 external rotor brushless power motors of system gathers It closes object lithium battery and provides power supply for 4108 external rotor brushless power motors.
6. the patrol rescue unmanned plane according to claim 5 based on voice signal, which is characterized in that further include for real When Image Acquisition camera module (4), the camera module (4) is electrically connected with processor (3).
7. the patrol rescue unmanned plane according to claim 6 based on voice signal, which is characterized in that further include by raising one's voice The communication module (7) of device and microphone composition, the communication module (7) are electrically connected with processor (3);Further include highlighted lamp source (6), lamp source (6) is highlighted to be electrically connected with processor (3).
8. a kind of method of work of the patrol rescue unmanned plane based on voice signal, which is characterized in that including step in detail below:
Step S101:The weak acoustic signal receiver (9) of patrol rescue unmanned plane receives weak acoustic signal;
Step S102:Voice identification module (10) judges whether received signal is emergency sound;
Step S103:If voice identification module (10) judges signal for voice, semantics recognition module (11) is started to work, and is judged Whether signal is distress signal;
Step S104:If signal is distress signal, voice locating module (12) is to the relative position of voice signal and unmanned plane It is positioned and this is transmitted to processor (3);
Step S105:The real time position of real-time locating module (1) positioning unmanned plane, processing is transmitted to by the real time position of unmanned plane Device (3);
Step S106:Processor (3) positions the real time position of the relative position and unmanned plane of signal and unmanned plane with voice Method carries out comprehensive analysis, and the coordinate of the signal analyzed is sent to security control center, and to security control center It reminds at PC ends;
Step S107:Processor (3) control flying power module (2), unmanned plane is with flying to sound source;
Step S108:When flying to sound source, step S101 to step S103 is performed again, to safety if distress signal is confirmed as The PC ends of control centre send alarm;
Step S109:Open image capture module and by the PC ends of image scene real-time Transmission to security control center.
9. the method for work of the patrol rescue unmanned plane according to claim 8 based on voice signal, which is characterized in that institute The voice localization method stated, including specific steps:It is removed first by the wave filter of a processor (3) in weak acoustic signal Potential noise obtains voice as pure as possible;Then time delay estimation is done by a GCC-PHAT formula;Finally build again Geometrical relationship estimates the spatial position of voice;Specifically GCC-PHAT formula are:Voice reaches microphone sound pick-up i and microphone The range difference of sound pick-up j=voice reaches the distance of distance-voice arrival microphone sound pick-up j of microphone sound pick-up i.
10. the method for work of the patrol rescue unmanned plane according to claim 8 based on voice signal, which is characterized in that It is further comprising the steps of:
Step S110:The PC ends of security control center are by checking that the live realtime graphic that unmanned plane is sent carries out field condition Judge and take corresponding rescue measure;
Step S111:If scene there are the objects such as trees to block, surface state is not seen, then unmanned plane hovering, security control center PC ends pass through communication module (7) and person for help and establish interim call mechanism;
Step S112:If person for help is hindered by animal, the PC ends of security control center are broadcast by the loud speaker in communication module (7) The audio for dispersing animal is put, animal is dispersed;
Step S113:If being rescued at night, person for help's specific location is indicated for rescue personnel using highlighted lamp source (6).
CN201711307151.1A 2017-12-11 2017-12-11 A kind of patrol rescue unmanned plane and its method of work based on voice signal Pending CN108053824A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711307151.1A CN108053824A (en) 2017-12-11 2017-12-11 A kind of patrol rescue unmanned plane and its method of work based on voice signal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711307151.1A CN108053824A (en) 2017-12-11 2017-12-11 A kind of patrol rescue unmanned plane and its method of work based on voice signal

Publications (1)

Publication Number Publication Date
CN108053824A true CN108053824A (en) 2018-05-18

Family

ID=62123858

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711307151.1A Pending CN108053824A (en) 2017-12-11 2017-12-11 A kind of patrol rescue unmanned plane and its method of work based on voice signal

Country Status (1)

Country Link
CN (1) CN108053824A (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109229303A (en) * 2018-09-25 2019-01-18 上海海事大学 A kind of Intelligent lifesaving circle based on speech recognition
CN109460063A (en) * 2018-12-25 2019-03-12 中南大学 A kind of unmanned plane search and rescue system and its method
CN110203410A (en) * 2019-06-12 2019-09-06 河海大学文天学院 Unmanned machine testing rail error system and its method
CN110907894A (en) * 2019-11-11 2020-03-24 北京信息科技大学 Remote control type life detection device and detection method thereof
WO2020135528A1 (en) * 2018-12-25 2020-07-02 深圳市道通智能航空技术有限公司 Unmanned aircraft, voice system thereof, and voice interaction method
CN111540166A (en) * 2020-05-09 2020-08-14 重庆工程学院 Unmanned aerial vehicle night search system and method based on deep learning
CN112607023A (en) * 2020-12-31 2021-04-06 三峡大学 Unmanned aerial vehicle rescue device and method based on sound wave positioning
CN112904885A (en) * 2021-03-19 2021-06-04 国网浙江省电力有限公司培训中心 Supporting rescue device of unmanned aerial vehicle based on multi-modal location
CN112995597A (en) * 2021-02-24 2021-06-18 四川腾盾科技有限公司 System and method for real-time target locking of high-speed unmanned aerial vehicle
CN115598594A (en) * 2022-10-13 2023-01-13 广州成至智能机器科技有限公司(Cn) Unmanned aerial vehicle sound source positioning method and device, unmanned aerial vehicle and readable storage medium

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102521945A (en) * 2011-12-02 2012-06-27 无锡奥盛信息科技有限公司 Calling detection alarming method and device
CN104851424A (en) * 2015-04-06 2015-08-19 何涛 Voice recognition caring and nursing method and system
CN106826846A (en) * 2017-01-06 2017-06-13 南京赫曼机器人自动化有限公司 The intellect service robot and method driven based on abnormal sound and image event
US20170177925A1 (en) * 2015-12-16 2017-06-22 Global Tel*Link Corp. Unmanned aerial vehicle with biometric verification
CN106997655A (en) * 2017-05-16 2017-08-01 肇庆学院 A kind of fly able sound monitoring warning device, method and application
CN206460335U (en) * 2016-12-13 2017-09-01 陕西奥翔网络科技有限公司 Search and rescue unmanned plane in a kind of field
CN206627830U (en) * 2017-04-12 2017-11-10 西南交通大学 A kind of indoor fire-fighting UAS
CN107358958A (en) * 2017-08-30 2017-11-17 长沙世邦通信技术有限公司 Intercommunication method, apparatus and system

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102521945A (en) * 2011-12-02 2012-06-27 无锡奥盛信息科技有限公司 Calling detection alarming method and device
CN104851424A (en) * 2015-04-06 2015-08-19 何涛 Voice recognition caring and nursing method and system
US20170177925A1 (en) * 2015-12-16 2017-06-22 Global Tel*Link Corp. Unmanned aerial vehicle with biometric verification
CN206460335U (en) * 2016-12-13 2017-09-01 陕西奥翔网络科技有限公司 Search and rescue unmanned plane in a kind of field
CN106826846A (en) * 2017-01-06 2017-06-13 南京赫曼机器人自动化有限公司 The intellect service robot and method driven based on abnormal sound and image event
CN206627830U (en) * 2017-04-12 2017-11-10 西南交通大学 A kind of indoor fire-fighting UAS
CN106997655A (en) * 2017-05-16 2017-08-01 肇庆学院 A kind of fly able sound monitoring warning device, method and application
CN107358958A (en) * 2017-08-30 2017-11-17 长沙世邦通信技术有限公司 Intercommunication method, apparatus and system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
刘真旭: ""基于麦克风阵列的实时声源定位"", 《计算机应用与软件》 *
杨云等: ""MSP430近场移动声源定位系统的实现"", 《计算机工程与设计》 *

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109229303A (en) * 2018-09-25 2019-01-18 上海海事大学 A kind of Intelligent lifesaving circle based on speech recognition
CN109460063B (en) * 2018-12-25 2022-05-27 中南大学 Unmanned aerial vehicle search and rescue system and method thereof
CN109460063A (en) * 2018-12-25 2019-03-12 中南大学 A kind of unmanned plane search and rescue system and its method
WO2020135528A1 (en) * 2018-12-25 2020-07-02 深圳市道通智能航空技术有限公司 Unmanned aircraft, voice system thereof, and voice interaction method
CN110203410A (en) * 2019-06-12 2019-09-06 河海大学文天学院 Unmanned machine testing rail error system and its method
CN110203410B (en) * 2019-06-12 2022-11-15 河海大学文天学院 System and method for detecting rail error by unmanned aerial vehicle
CN110907894A (en) * 2019-11-11 2020-03-24 北京信息科技大学 Remote control type life detection device and detection method thereof
CN111540166A (en) * 2020-05-09 2020-08-14 重庆工程学院 Unmanned aerial vehicle night search system and method based on deep learning
CN112607023A (en) * 2020-12-31 2021-04-06 三峡大学 Unmanned aerial vehicle rescue device and method based on sound wave positioning
CN114706041A (en) * 2020-12-31 2022-07-05 三峡大学 Positioning method of unmanned aerial vehicle rescue device based on sound wave positioning
CN114706041B (en) * 2020-12-31 2024-04-19 三峡大学 Positioning method of unmanned aerial vehicle rescue device based on acoustic wave positioning
CN112995597A (en) * 2021-02-24 2021-06-18 四川腾盾科技有限公司 System and method for real-time target locking of high-speed unmanned aerial vehicle
CN112904885A (en) * 2021-03-19 2021-06-04 国网浙江省电力有限公司培训中心 Supporting rescue device of unmanned aerial vehicle based on multi-modal location
CN115598594A (en) * 2022-10-13 2023-01-13 广州成至智能机器科技有限公司(Cn) Unmanned aerial vehicle sound source positioning method and device, unmanned aerial vehicle and readable storage medium
CN115598594B (en) * 2022-10-13 2023-05-05 广州成至智能机器科技有限公司 Unmanned aerial vehicle sound source positioning method and device, unmanned aerial vehicle and readable storage medium

Similar Documents

Publication Publication Date Title
CN108053824A (en) A kind of patrol rescue unmanned plane and its method of work based on voice signal
CN109460063B (en) Unmanned aerial vehicle search and rescue system and method thereof
CN109481248A (en) A kind of smart guide glasses
CN106859929A (en) A kind of Multifunctional blind person guiding instrument based on binocular vision
CN106598067A (en) Combined medical disaster rescue system and method based on multi-UAVs
CN206460332U (en) Group medicine disaster assistance system based on multiple no-manned plane
US20040190753A1 (en) Image transmission system for a mobile robot
CN112429229A (en) Unmanned aerial vehicle search and rescue system based on Internet of things
CN105881549B (en) One kind industrialization robot real-time high-precision detection system
CN106989728A (en) A kind of building ground mapping system based on unmanned plane
CN105930804A (en) Power transmission equipment identification platform arranged on unmanned aerial vehicle
CN109521793A (en) A kind of intelligent food safety detection method and its system based on unmanned plane
CN203865010U (en) Navigation aircraft capable of achieving automatic tracking, video recognition and intelligent aerial photographing
CN105208346B (en) Transmission facility identification method based on unmanned plane
CN110162102A (en) Unmanned plane automatic identification tracking and system based on cloud platform and machine vision
CN106251604A (en) A kind of Landscape Construction information collecting and monitoring system
CN108125776A (en) A kind of glasses for guiding blind
CN108196588A (en) A kind of mountain tour emergency rescuing system based on unmanned plane
CN108270927B (en) Life saving instrument and saving method based on mobile phone
CN110470307A (en) A kind of visually impaired patient navigation system and method
CN112607023B (en) Unmanned aerial vehicle rescue device and method based on sound wave positioning
CN211281412U (en) Intelligent recognition tracks rescue unmanned aerial vehicle on water
CN106452557A (en) Unmanned aerial vehicle search system based on wireless communication technology
Xu et al. Multi-functional smart E-glasses for vision-based indoor navigation
CN104932370A (en) Environment detecting device, human identification method and emergency monitoring system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20180518