CN104299351B - Intelligent early-warning firefighting robot - Google Patents

Intelligent early-warning firefighting robot Download PDF

Info

Publication number
CN104299351B
CN104299351B CN201410565265.6A CN201410565265A CN104299351B CN 104299351 B CN104299351 B CN 104299351B CN 201410565265 A CN201410565265 A CN 201410565265A CN 104299351 B CN104299351 B CN 104299351B
Authority
CN
China
Prior art keywords
fire
flame
fire disaster
warning
disaster flame
Prior art date
Application number
CN201410565265.6A
Other languages
Chinese (zh)
Other versions
CN104299351A (en
Inventor
段锁林
周玉勤
郑剑锋
王雪
靳琪琳
Original Assignee
常州大学
江苏润仪仪表有限公司
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 常州大学, 江苏润仪仪表有限公司 filed Critical 常州大学
Priority to CN201410565265.6A priority Critical patent/CN104299351B/en
Publication of CN104299351A publication Critical patent/CN104299351A/en
Application granted granted Critical
Publication of CN104299351B publication Critical patent/CN104299351B/en

Links

Classifications

    • GPHYSICS
    • G08SIGNALLING
    • G08BSIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B17/00Fire alarms; Alarms responsive to explosion
    • G08B17/06Electric actuation of the alarm, e.g. using a thermally-operated switch
    • AHUMAN NECESSITIES
    • A62LIFE-SAVING; FIRE-FIGHTING
    • A62CFIRE-FIGHTING
    • A62C37/00Control of fire-fighting equipment
    • GPHYSICS
    • G08SIGNALLING
    • G08BSIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B17/00Fire alarms; Alarms responsive to explosion
    • G08B17/10Actuation by presence of smoke or gases automatic alarm devices for analysing flowing fluid materials by the use of optical means
    • GPHYSICS
    • G08SIGNALLING
    • G08BSIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B17/00Fire alarms; Alarms responsive to explosion
    • G08B17/12Actuation by presence of radiation or particles, e.g. of infra-red radiation or of ions
    • G08B17/125Actuation by presence of radiation or particles, e.g. of infra-red radiation or of ions by using a video camera to detect fire or smoke

Abstract

The invention discloses a kind of intelligent early-warning firefighting robot, it is made up of mounting temperature sensor, Smoke Sensor, flame sensor, monopod video camera, degree of freedom mechanical hand, fire extinguisher, controller on wheeled vehicle, two-freedom mechanical hand is encircled and is arranged on fire extinguisher side, and temperature sensor, Smoke Sensor, flame sensor, monopod video camera, degree of freedom mechanical hand connect controller according to electricity circuit type. Capacity of resisting disturbance of the present invention is strong, and the space suitability is good, it is possible to monitor the fire disaster flame video of the trickle burning things which may cause a fire disaster of Initial Stage of Fire, and fire alarm accuracy is high, and rate of false alarm is low.

Description

Intelligent early-warning firefighting robot

Technical field

The invention belongs to public safety field, relate to fire disaster flame detection early warning and Robot Extinguishing Fire system, be specifically related to the intelligent early-warning firefighting robot of a kind of automatic detecting.

Background technology

Fire is violent such as tiger! Fire is the major hidden danger threatening human survival, and along with the quickening of social development paces, especially in recent years, the fire taken place frequently causes great Loss of Life and property to people. According to statistics, China's breaking out of fire 125402 altogether in 2011, causes 1106 people dead, direct economic loss 19.8 hundred million yuan. The wildfire that in June, 2011, Shanghai occurred, causes 58 people dead, direct economic loss 1.58 hundred million yuan. Modern high tech method is utilized to find fire source in time and take effective fire protection to put out incipient flame in time, it is possible to greatly reducing various loss, firefighting robot is undoubtedly for we provide best selection.

Most of fire occur why can create greater casualties accident and the heavy losses of property, its main cause has: 1, the condition of a fire finds not in time, at some important places, as warehouse, laboratory, workshop etc. do not have the equipment of timing circling measurment flame, particularly easily cause the place such as the place of spontaneous combustion or the chemical workshop of fire-prone, if not being equipped with energy circling measurment equipment, the condition of a fire cannot be found in time, the little fire causing early stage quickly spreads into big fire, cause difficulty to fire extinguishing later, so that breeding disaster; Although 2, some has flame detecting device, as monitoring camera is first-class, but due to condition of a fire information propagation delay, it is impossible to take effective fire suppression measures at the initial stage that fire occurs, will also result in great fire damage.

It is known that to put out a fire, first obtaining fire detection, fire defector about experienced by the development course from temperature-sensitive detection → ionic smoke sensor detection → photoelectric smoke detection → air-sensitive detection → infrared detection → Video Detection. It is mainly used in the fire detection of indoor little scope as the temperature-sensitive of traditional fire detection, sense cigarette, air-sensitive class detector, has that the space suitability is limited, information reliability is poor, be subject to the shortcomings such as external interference. And Infrared Detectors is expensive, limit its use scope. Video detector is a kind of fire detector based on image processing techniques, utilize camera supervised monitoring point sight, by the video information caught being processed, analyzing, and then realize the detection of fire disaster flame and the early warning of fire, capacity of resisting disturbance is relatively strong, and the space suitability is strong. In video fire hazard early warning technology, fire disaster flame detection method is core, although video fire hazard flame detecting method has been carried out certain research and development by domestic expert in recent years, but the technology of maturation rarely has application, and corresponding product also seldom emerges.

Summary of the invention

It is an object of the invention to: providing a kind of intelligent early-warning firefighting robot, capacity of resisting disturbance is strong, and the space suitability is good, it is possible to monitor the fire disaster flame video of the trickle burning things which may cause a fire disaster of Initial Stage of Fire, and fire alarm accuracy is high, and rate of false alarm is low.

The technical solution of the present invention is: this intelligent early-warning firefighting robot is made up of mounting temperature sensor, Smoke Sensor, flame sensor, monopod video camera, two-freedom mechanical hand, fire extinguisher, controller on wheeled vehicle; Utilize the clapped fire disaster flame image of monopod video camera, realize the early warning firefighting robot location to burning things which may cause a fire disaster by calculating; Fire extinguisher is arranged on two-freedom mechanical hand platform, utilizes joined camera review on mechanical hand to realize fire extinguisher nozzle alignment burning things which may cause a fire disaster, accurately to complete fire extinguishing; Temperature sensor, Smoke Sensor, flame sensor, monopod video camera, two-freedom mechanical hand connect controller according to electricity circuit type.

During work, wheeled vehicle moves, by the place that temperature sensor, Smoke Sensor, flame sensor and monopod video camera circling measurment are to be detected, instant fire condition of a fire state is provided through multi-data fusion and intelligent analysis, the analysis of recycling picture signal and process obtain the concrete orientation of burning things which may cause a fire disaster, provide corresponding early warning signal according to the condition of a fire, and realized transmission and the transmission of early warning signal by wireless network, start Portable extinguisher by two-freedom mechanical hand simultaneously and put out a fire in real time.

The invention have the advantage that

1, compared with detecting system with conventional fire, the early warning firefighting robot of the present invention can detect the discovery condition of a fire in time, the instant initial stage condition of a fire that eliminates, and fire is eliminated bud in the early stage by timely transmission condition of a fire information.

2, the space suitability is strong, both may be used for warehouse, the fire monitoring of the indoor little scope of laboratory class and early warning, it is also possible to for the monitoring of forest class large-range fire disaster and early warning.

3, adopting Internet video CCD camera to obtain monitoring point image, utilize digital image processing techniques to make early warning have higher capacity of resisting disturbance and networked Characteristics, cost performance is higher compared with infrared fire hazard monitoring system.

4, the static state of fire disaster flame and behavioral characteristics being carried out information fusion, provide a kind of fire distinguished number based on neutral net with the judgement of D-S evidence, energy Quick early warning, False Rate is relatively low.

5, three grades of fire disaster flame early warning are adopted, it is possible to realize early warning intellectuality according to the real-time flame in scene, scientific and reasonable and manpower when scheduling fire occurs in an orderly manner, financial resources.

6, the early warning firefighting robot of the present invention is more intelligent than traditional firefighting robot, it is possible to avoid fire damage or the heavy losses that are likely to result in by fire to be reduced to minimum.

7, early warning firefighting robot be integrate flame circling measurment, intelligent extinguishing device that fire alarm, startup are put out a fire, the fire disaster flame merged based on multicharacteristic information differentiates, by timing circling measurment, find the condition of a fire in time, and effectively fire extinguishing implemented by the airborne fire extinguisher of OnNow while issuing fire alarm information in time by network, fire is eliminated bud in the early stage, it is prevented effectively from spreading of initial fire disaster, it is ensured that find the condition of a fire, the instant elimination initial stage little condition of a fire, in time transmission condition of a fire information in time.

8, four layers of intelligence are adopted to think over-reaction system structure: sensing layer, layer of thinking over (intelligent decision layer), conversion zone (key-course) and execution level, total system adopts the distributed system architecture based on Internet of Things to carry out deployment facility, enhance intelligence degree and the suitability of system, three grades of modes of warnings of fire can be realized in conjunction with condition of a fire threshold value of warning.

9, this early warning firefighting robot is intended to the condition of a fire of monitoring discovery in time, immediate reaction eliminates the condition of a fire, send transmission condition of a fire information in time, as far as possible fire is eliminated in bud, this is maximum feature and the advantage of early warning firefighting robot of the present invention, is also that the present invention is different from tradition fire-fighting, fire extinguishing the indicating the most significantly of concept.

10, the present invention make use of the integration technology of the information such as temperature, smog, flame and image in fire disaster flame detection identification, burning things which may cause a fire disaster location, improves fire and differentiates accuracy, and False Rate is relatively low.

Accompanying drawing explanation

Fig. 1 is early warning firefighting robot population structure schematic diagram.

Fig. 2 is early warning firefighting robot four coating systems Organization Chart.

Fig. 3 is early warning firefighting robot hardware block diagram.

Fig. 4 judges flame object schematic diagram based on The Cloud Terrace angle.

Fig. 5 is mobile phone alarm system Technology Selection schematic diagram.

Fig. 6 is SMS alarm subsystem software structure flow chart.

Detailed description of the invention

Further illustrating the technical solution of the present invention below in conjunction with concrete accompanying drawing, it is impossible to be not understood as the restriction to technical scheme, the reasonable change on the basis of the technical program broadly falls into protection scope of the present invention.

As shown in Figure 1, this intelligent early-warning firefighting robot is made up of mounting temperature sensor, Smoke Sensor, flame sensor, monopod video camera, degree of freedom mechanical hand, fire extinguisher, controller on wheeled vehicle, two-freedom mechanical hand is encircled and is arranged on fire extinguisher side, and temperature sensor, Smoke Sensor, flame sensor, monopod video camera, degree of freedom mechanical hand connect controller according to electricity circuit type.

As shown in Figure 2,3, the early warning firefighting robot of the present invention includes following system:

1, fire disaster flame data logger system: include wheeled vehicle and temperature sensor mounted thereto, Smoke Sensor, flame sensor and CCD monopod video camera, infrared distance sensor, utilizes these sensors and data message to merge the Real Time Obstacle Avoiding of the accurately identifying of fire disaster flame, burning things which may cause a fire disaster location and wheeled vehicle; Wherein CCD cradle head camera supports that WiFi is wirelessly transferred, support TCP/IP network remote image transmitting simultaneously, and have and dynamically detect recording function, it is responsible for completing fire disaster flame image acquisition in real time, according to slewing area (level 300 degree, vertical 90 degree) the timing scan monitoring site pre-set;

2, based on the data fusion decision system of DSP: with TI company's T MS320C6713 for acp chip, utilize the information such as fire disaster flame temperature that ARM system transmits, smog, image, real-time flame, utilize intelligent information processing technology to realize the functions such as the feature identification of fire disaster flame, information fusion, warning level judgement, intelligent decision;

3, based on the fire alarm data radio transmission system of LAN: adopt the Http agreement of standard, related data is transmitted by corresponding port, utilizing the advantage of this protocol transmission is can firewall-penetrating and NAT easily, use the occupancy volume that can reduce bandwidth without connection Http agreement simultaneously, the packet structure body of communications is as shown in the table, alert data in transmission is carried out DES symmetric cryptography process simultaneously, Base64 is utilized to be encrypted, VC and Java is general, it is decrypted at receiving terminal and verifies, it is ensured that the safety of alert data transmission and integrity;

Packet structure table

Type Title Explanation Time_t Create_time Send packet time Comm._addr dest_addr, src_addr Source address and destination address Short Priority Data transport priority Short Len Data packet length Short Type Transmission data program class Short Sub_type Transmission data function class Char Buf[MAX_MESSAGE_BUF_LEN] Detailed data information

4, two-freedom mechanical hand fire extinguishing system: fire extinguisher (dry-type) is arranged on a two-freedom platform, utilize image trace technology, adjust orientation in real time, and reach nozzle of fire extinguisher alignment fire's point of origin, utilize relay to open fire extinguisher, and then realize fire extinguishing;

5, monopod video camera controls module: monopod video camera is for realizing the location to target flame and tracking, once there be fire disaster flame to occur, first pass through position and direction that monopod video camera detects that flame occurs, then driven machine people convergence fire disaster flame, and start Robot actions fire extinguisher to complete fire extinguishing task;Ccd video camera colored for the EVI-D100P adopted has bigger field range, this video camera also has at a high speed, rotates within the scope of wide visual field, tilts running simultaneously, 40 times of zooms, fast and stable ground autozoom and automatic exposure control BLC, suitable in light bad when, it both can adopt infrared control zoom, focusing, orientation, brightness etc., it is also possible to is carried out RS-232 series connection from PC by VISCATM code and controls, the same EVI-D30/EVI-D31 of communications protocol, it is proper that direct drive motor makes video camera each several part assembly coordinate, it is achieved the high speed rotating of The Cloud Terrace, in addition, EVI-D100P is also equipped with having any different in the intelligent feature of other video camera products: from motion tracking (AT) and mobile detection (MD), autotrack mode plays a part particular importance in mobile apparatus people's target following, when there is relative movement with mobile apparatus people in the object in scene, automatic tracking system limits a camera picture region by controller, this region contains this mobile object, then the pixel that color is similar to this object is found out in the zone, The Cloud Terrace is controlled by pan-shot and rotating function, video camera is made automatically to follow this mobile object,

6, multisensor fire information Fusion Module: in order to find the video fire hazard fire defector technology of practicality, strengthen the robustness of video fire hazard flame detecting method, improve detectivity simultaneously and reduce rate of false alarm, after analyzing fire disaster flame identification differentiation feature, on the basis of D-S evidence decision theory, adopt Kalman's Weighted Fusion algorithm to combine and carry out fire disaster flame judgement, but wherein many experiments judgement is played auxiliary and is asked for weights effect, to eliminate individually a kind of or the less produced False Rate of feature judgement is high, be disturbed the problems such as impact is big; First the fire disaster flame data sample of indoor environment is gathered, by D-S evidence decision theory and neutral net, sample data is trained obtaining weights and the threshold value of four kinds of fire disaster flame features, it is normalized the final weights of acquisition, then utilizes multiple features fusion algorithm to obtain final judging result; Under other outdoors or complex environment, it is necessary to re-start the study of sample collection and neutral net, it is determined that more suitable multiple features fusion weights.

7, three grades of intelligent early-warning modules of fire fuzzy reasoning: early warning system adopts three grades of modes of warnings, on-the-spot acousto-optic early warning, remote short messages breath early warning and Network Warning; When there is the one-level condition of a fire, being arranged on the voice guard in the monitoring point in monitored district and warning light is responsible for realizing on-the-spot acousto-optic early warning, alarm command is controlled module by monitoring site and independently assigns; Android phone server sends early warning short message to realize remote short messages breath early warning by JavaServlet real time access warning data storehouse and when monitoring point two grades of condition of a fire of discovery to appointment person liable, it is wirelessly transferred and utilizes WiFi technology, the transmission of warning data adopts encrypted transmission, it is ensured that the safety of data; When the condition of a fire that fire disaster flame detects system reception and registration reaches the highest three grades, jump out warning window and picture, it is achieved network alarming, and it is live to transmit on-the-spot fire;

8, burning things which may cause a fire disaster identification locating module based on image: follow the tracks of in the process of target flame at The Cloud Terrace, real-time feeds back to host computer by the angle information of own rotation, the relative position of target flame and mobile apparatus people is calculated based on the angle information of The Cloud Terrace, as shown in Figure 4, dotted line position represents cradle head camera orientation in an initial condition, h represents the difference in height between camera center C point and ground, when monopod video camera searches target fire disaster flame p, the optical axis of video camera is cp, wherein Φ is the pitch angle of monopod video camera, wherein Ψ represents horizontal side angle, the distance of video camera and impact point;Because The Cloud Terrace is fixed on mobile apparatus people, its basis coordinates system and the relativeness of robot coordinate system are also changeless, by measuring the relation obtaining between Two coordinate system;

9, based on the Real Time Obstacle Avoiding module of range finding and path planning: based on the infrared sensor and the laser range sensor signal that are arranged on the mobile apparatus people left and right sides and front, robot real-time implementation barrier avoiding function, and respective paths planning and navigation are realized;

10, based on the radio network information transport module of WiFi: needed immediately to be sent to the person liable of server and relevant departments through the fire alarm information that information fusion algorithm obtains by dsp system, the wireless network transmissions function utilizing WiFi technology is needed to realize, wireless network communication technique has with low cost, the advantage such as adaptability good, favorable expandability, maintenance are easy, being used widely in fields such as industry, oil, electric power energy, robots, wireless network transmission module is mainly completed by communication protocol;

11: based on the mobile phone fire alarm information transmission modular of Android technology: short message alarm module is that the handset server of an indoor Android is by JavaServlet real time access alarm database, fire alarm information is had in this region when detecting, then warning message and position are sent to the director specified by handset server automatically, notify that director takes measures to put out a fire; Android client application uses Java technology, and network service uses ApacheHTTP protocol client, and middle Web server adopts the request of Servlet customer in response; Background data base adopts database connection pool to access data base, the part storage of client adopts SQLite data base, because referring here to the data synchronization problems of Android server and database server, employing heartbeat mechanism solves, and the Technology Selection of detailed alarm mobile phone server is as shown in Figure 5; SMS alarm subsystem operation platform is handset server, as shown in Figure 6, is shown as mobile phone alarm subsystem and arranges the IP address of server database place machine in figure, and after being provided with, mobile phone alarm subsystem starts to start;

12, audible alarm subsystem: when firefighting robot is by Timer timer access remote-end alarm data base, if detecting, the alarm database corresponding to arbitrary CCD camera of the interior space changes, the then corresponding interior space produces fire alarm information, robot will produce audible alarm in the very first time, warning personnel around have fire dangerous situation, prompting is withdrawn as early as possible, and robot starts search fire extinguishing simultaneously; Audible alarm subsystem is to be provided simultaneously with at firefighting robot and detection terminal, calls the VC operation code of alarm sound;

13, two-freedom robot arm fire extinguishing system controls module: utilize image information to complete the orientation recognition of burning things which may cause a fire disaster target, controls two-freedom platform in time, it is achieved the shower nozzle of dry powder extinguisher is accurately aimed at fire's point of origin, and unlatching fire extinguisher completes fire extinguishing immediately;

14, video management module: adopt generic video management module, image pick-up card obtains the real time imaging of described Internet video CCD camera, process after obtaining image by digital image processing techniques, call fire disaster flame discriminating program, provide the judgement conclusion whether having the condition of a fire, if there being the condition of a fire to provide the judgement conclusion of condition of a fire rank.

Wherein, the method for discrimination of the fire disaster flame data fusion of the present invention and the acquisition methods of real-time early warning threshold value comprise the following steps:

Step one: described Internet video CCD camera completes field imaging, image pick-up card obtains the on-the-spot digital picture that described Internet video CCD camera transmits, and utilizes digital image processing techniques and this digital picture is carried out pretreatment;

Step 2: utilize gray scale thresholding technology by comparing the motion feature of the change-detection fire disaster flame of the gray scale thresholding of front and back two two field picture; The area change feature of growth detection fire disaster flame whether is had by comparing front and back two two field picture area of flame; Detection flame color feature in HIS color space; The differentiation to ambient interferences such as naturally light and daylight lamp monochrome informations is realized by extracting the circularity feature of fire disaster flame; Extract the wedge angle number of fire disaster flame, utilize the wedge angle number of fire disaster flame along with presenting the feature of irregular rule burning time using the change of the wedge angle number of fire disaster flame as one of Fire Criterion; The basic frequency at fire disaster flame burning initial stage is within 12Hz, and the change of surrounding is not notable to frequency influence, therefore extracts the blinking characteristics of fire disaster flame as one of Fire Criterion;

Step 3: gather the fire disaster flame data sample of monitoring point, selects weights and threshold value that the fire disaster flame data sample of described monitoring point is trained in advance obtaining the flicker frequency feature these four fire disaster flame feature including fire disaster flame color characteristic, the circularity feature of fire disaster flame, the wedge angle number of fire disaster flame and fire disaster flame;

Step 4: utilize formula (1) to obtain fire disaster flame threshold value of warning:

(1)

In formula:Export for i-th fire disaster flame feature,, H0Indicate without fire disaster flame, H1Indicate fire disaster flame;, mi(H0) and mi(H1) for the detection probability of i-th fire disaster flame feature and probability of miscarriage of justice, utilize above-mentioned detected real time threshold to compare with empirical value and realize fire disaster flame and detect in real time and early warning.

Intelligent early-warning firefighting robot of the present invention, adopts the distributed system architecture of Internet of Things to come deployment facility and system, and system deployment is divided into perception terminal, data collection and analysis layer, decision-making level and four parts of terminal application platform (execution level); Perception terminal includes high resolution CCD video camera, temperature, smog, range finding, infrared sensor, flame detection sensor and the voice guard and the warning light that are arranged in the monitoring point in monitored district by The Cloud Terrace; Data collection and analysis layer includes the signal processing and analysis device of various kinds of sensors, dsp system, installs video frequency collection card in a computer, Mysql data base and Android phone server; Terminal application platform includes mobile apparatus people, two-freedom Camera Platform and fire extinguisher platform; Also include having data processing function, be connected to the Internet and have the computer of public network IP address; Ccd video camera has been responsible for sight imaging in indoor monitoring point, image pick-up card gathers the become image of ccd video camera, by video management module, the digital picture read is sequentially carried out pretreatment, each feature extraction, fire disaster flame characteristics of image, in conjunction with information such as temperature, smog, flame sensors, utilizes multicharacteristic information blending algorithm output early warning signal; Three grades of modes of warnings are adopted, including on-the-spot acousto-optic early warning, long-range short message early warning and Network Warning according to condition of a fire size; Initial Stage of Fire, alarm command is controlled module by monitoring site and independently assigns or assigned by Surveillance center, being arranged on the voice guard in the monitoring point in monitored district and warning light is responsible for realizing on-the-spot acousto-optic early warning, meanwhile, early warning firefighting robot starts dry powder extinguisher and extinguishes the condition of a fire; Can also by being positioned near monitoring point and can hear that the person liable of alarm song adopts manual type to put out a fire; Also starting remote short messages modes of warning, Android phone server sends early warning short message by JavaServlet real time access warning data storehouse and when the monitoring point discovery condition of a fire to appointment person liable, and now on-the-spot sound and light alarm is still effective simultaneously;Specify person liable to the site disposal condition of a fire, according to the on-the-spot real-time monitoring result of fire disaster flame early warning system, if fire spreading degree arrives certain experiences threshold value, then start network alarming, jump out warning window and picture, transmit far-end video information simultaneously, select to be reported to the police by the staff 110 of Surveillance center or automatic 110 warnings of system, employ the fire extinguishing of public fire protection facility.

Claims (2)

1. intelligent early-warning firefighting robot, it is made up of mounting temperature sensor, Smoke Sensor, flame sensor, CCD monopod video camera, degree of freedom mechanical hand, fire extinguisher, controller on wheeled vehicle, two-freedom mechanical hand is encircled and is arranged on fire extinguisher side, and temperature sensor, Smoke Sensor, flame sensor, CCD monopod video camera, degree of freedom mechanical hand connect controller according to electricity circuit type; It is characterized in that this early warning firefighting robot includes following system:
(1) fire disaster flame data logger system: include wheeled vehicle and temperature sensor mounted thereto, Smoke Sensor, flame sensor and CCD monopod video camera, infrared distance sensor, utilizes these sensors and data message to merge the Real Time Obstacle Avoiding of the accurately identifying of fire disaster flame, burning things which may cause a fire disaster location and wheeled vehicle; Wherein CCD monopod video camera supports that WiFi is wirelessly transferred, support TCP/IP network remote image transmitting simultaneously, and have and dynamically detect recording function, it is responsible for completing fire disaster flame image acquisition in real time, according to the level pre-set 300 degree, vertical 90 degree of slewing area timing scan monitoring sites;
(2) based on the data fusion decision system of DSP: with TI company's T MS320C6713 for acp chip, utilize fire disaster flame temperature that ARM system transmits, smog, image, real-time flame information, utilize intelligent information processing technology to realize the feature identification of fire disaster flame, information fusion, warning level judge, intelligent decision function;
(3) based on the fire alarm data radio transmission system of LAN: adopt the Http agreement of standard, related data is transmitted by corresponding port, utilizing the advantage of this protocol transmission is firewall-penetrating and NAT easily, use without the occupancy volume connecting Http agreement reduction bandwidth simultaneously, alert data in transmission is carried out DES symmetric cryptography process, utilizing Base64 to be encrypted, VC and Java is general, is decrypted at receiving terminal and verifies;
(4) two-freedom mechanical hand fire extinguishing system: dry powder type fire extinguisher is arranged on a two-freedom platform, utilize image trace technology, adjust orientation in real time, and reach nozzle of fire extinguisher alignment fire's point of origin, utilize relay to open fire extinguisher, and then realize fire extinguishing;
(5) CCD monopod video camera controls module: CCD monopod video camera is for realizing the location to target flame and tracking, once there be fire disaster flame to occur, first pass through position and direction that CCD monopod video camera detects that flame occurs, then driven machine people convergence fire disaster flame, and start Robot actions fire extinguisher to complete fire extinguishing task; Adopt CCD monopod video camera colored for EVI-D100P, carried out RS-232 series connection from PC by VISCATM code and control, the same EVI-D30/EVI-D31 of communications protocol;
(6) multisensor fire information Fusion Module: after analyzing fire disaster flame identification differentiation feature, on the basis of D-S evidence decision theory, adopt Kalman's Weighted Fusion algorithm to combine and carry out fire disaster flame judgement; First the fire disaster flame data sample of indoor environment is gathered, by D-S evidence decision theory and neutral net, sample data is trained obtaining weights and the threshold value of four kinds of fire disaster flame features, it is normalized the final weights of acquisition, then utilizes multiple features fusion algorithm to obtain final judging result;Under other outdoors or complex environment, it is necessary to re-start the study of sample collection and neutral net, it is determined that more suitable multiple features fusion weights;
(7) three grades of intelligent early-warning modules of fire fuzzy reasoning: early warning system adopts three grades of modes of warnings, on-the-spot acousto-optic early warning, remote short messages breath early warning and Network Warning; When there is the one-level condition of a fire, being arranged on the voice guard in the monitoring point in monitored district and warning light is responsible for realizing on-the-spot acousto-optic early warning, alarm command is controlled module by monitoring site and independently assigns; Android phone server sends early warning short message to realize remote short messages breath early warning by JavaServlet real time access warning data storehouse and when monitoring point two grades of condition of a fire of discovery to appointment person liable, it is wirelessly transferred and utilizes WiFi technology, the transmission of warning data adopts encrypted transmission, it is ensured that the safety of data; When the condition of a fire that fire disaster flame detects system reception and registration reaches the highest three grades, jump out warning window and picture, it is achieved network alarming, and it is live to transmit on-the-spot fire;
(8) based on the burning things which may cause a fire disaster identification locating module of image: following the tracks of in the process of target flame at The Cloud Terrace, real-time feeds back to host computer by the angle information of own rotation, the relative position of target flame and mobile apparatus people is calculated based on the angle information of The Cloud Terrace;
(9) based on the Real Time Obstacle Avoiding module of range finding and path planning: based on the infrared sensor and the laser range sensor signal that are arranged on the mobile apparatus people left and right sides and front, robot real-time implementation barrier avoiding function, and respective paths planning and navigation are realized;
(10) based on the radio network information transport module of WiFi: needed immediately to be sent to the person liable of server and relevant departments through the fire alarm information that information fusion algorithm obtains by dsp system, the wireless network transmissions function utilizing WiFi technology realizes, and wireless network transmission module is mainly completed by communication protocol;
(11) based on the mobile phone fire alarm information transmission modular of Android technology: short message alarm module is that the handset server of an indoor Android is by JavaServlet real time access alarm database, fire alarm information is had in this region when detecting, then warning message and position are sent to the director specified by handset server automatically, notify that director takes measures to put out a fire; Android client application uses Java technology, and network service uses ApacheHTTP protocol client, and middle Web server adopts the request of Servlet customer in response; Background data base adopts database connection pool to access data base, and the part storage of client adopts SQLite data base;
(12) audible alarm subsystem: when firefighting robot is by Timer timer access remote-end alarm data base, if detecting, the alarm database corresponding to arbitrary CCD monopod video camera of the interior space changes, the then corresponding interior space produces fire alarm information, robot will produce audible alarm in the very first time, warning personnel around have fire dangerous situation, prompting is withdrawn as early as possible, and robot starts search fire extinguishing simultaneously; Audible alarm subsystem is to be provided simultaneously with at firefighting robot and detection terminal, calls the VC operation code of alarm sound;
(13) two-freedom robot arm fire extinguishing system controls module: utilize image information to complete the orientation recognition of burning things which may cause a fire disaster target, controls two-freedom platform in time, it is achieved the shower nozzle of dry powder extinguisher is accurately aimed at fire's point of origin, and unlatching fire extinguisher completes fire extinguishing immediately;
(14) video management module: adopt generic video management module, image pick-up card obtains the real time imaging of described CCD monopod video camera, process after obtaining image by digital image processing techniques, call fire disaster flame discriminating program, provide the judgement conclusion whether having the condition of a fire, if there being the condition of a fire to provide the judgement conclusion of condition of a fire rank.
2. intelligent early-warning firefighting robot according to claim 1, is characterized in that the acquisition methods of the method for discrimination of fire disaster flame data fusion and real-time early warning threshold value comprises the following steps:
Step one: described CCD monopod video camera completes field imaging, image pick-up card obtains the on-the-spot digital picture that described CCD monopod video camera transmits, and utilizes digital image processing techniques that this digital picture is carried out pretreatment;
Step 2: utilize gray scale thresholding technology by comparing the motion feature of the change-detection fire disaster flame of the gray scale thresholding of front and back two two field picture; The area change feature of growth detection fire disaster flame whether is had by comparing front and back two two field picture area of flame; Detection flame color feature in HIS color space; The differentiation to two kinds of ambient interferences such as naturally light and daylight lamp monochrome informations is realized by extracting the circularity feature of fire disaster flame; Extract the wedge angle number of fire disaster flame, utilize the wedge angle number of fire disaster flame along with presenting the feature of irregular rule burning time using the change of the wedge angle number of fire disaster flame as one of Fire Criterion; The basic frequency at fire disaster flame burning initial stage, within 12Hz, therefore extracts the blinking characteristics of fire disaster flame as one of Fire Criterion;
Step 3: gather the fire disaster flame data sample of monitoring point, selects weights and threshold value that the fire disaster flame data sample of described monitoring point is trained in advance obtaining the flicker frequency feature these four fire disaster flame feature including fire disaster flame color characteristic, the circularity feature of fire disaster flame, the wedge angle number of fire disaster flame and fire disaster flame;
Step 4: utilize formula (1) to obtain fire disaster flame threshold value of warning:
(1)
In formula:Export for i-th fire disaster flame feature,, H0Indicate without fire disaster flame, H1Indicate fire disaster flame;, mi(H0) and mi(H1) for the detection probability of i-th fire disaster flame feature and probability of miscarriage of justice, utilize above-mentioned detected real time threshold to compare with empirical value and realize fire disaster flame and detect in real time and early warning.
CN201410565265.6A 2014-10-22 2014-10-22 Intelligent early-warning firefighting robot CN104299351B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410565265.6A CN104299351B (en) 2014-10-22 2014-10-22 Intelligent early-warning firefighting robot

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410565265.6A CN104299351B (en) 2014-10-22 2014-10-22 Intelligent early-warning firefighting robot

Publications (2)

Publication Number Publication Date
CN104299351A CN104299351A (en) 2015-01-21
CN104299351B true CN104299351B (en) 2016-06-15

Family

ID=52319068

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410565265.6A CN104299351B (en) 2014-10-22 2014-10-22 Intelligent early-warning firefighting robot

Country Status (1)

Country Link
CN (1) CN104299351B (en)

Families Citing this family (31)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104732701A (en) * 2015-04-08 2015-06-24 鹰潭嘉坤云计算科技有限公司 Method and system for monitoring fire warning information
CN105169610A (en) * 2015-08-24 2015-12-23 铜陵学院 Dual-core servo control system based on two-wheeled high-speed fully automatic fire extinguishing robot
US9940820B2 (en) * 2015-10-29 2018-04-10 Honeywell International Inc. Systems and methods for verified threat detection
CN105261140B (en) * 2015-11-05 2017-03-15 四川雷克斯智慧科技股份有限公司 Multiple sensor integrated method for fire-fighting
CN105261029B (en) * 2015-11-20 2018-08-21 中国安全生产科学研究院 The robot of burning things which may cause a fire disaster positioning and fire extinguishing is carried out based on binocular vision
CN106887108A (en) * 2015-12-16 2017-06-23 天维尔信息科技股份有限公司 Early warning interlock method and system based on thermal imaging
CN105758450B (en) * 2015-12-23 2017-11-24 西安石油大学 Met an urgent need based on multisensor the fire-fighting early warning sensory perceptual system construction method of robot
CN105760824B (en) * 2016-02-02 2019-02-01 北京进化者机器人科技有限公司 A kind of moving human hand tracking method and system
CN105954718A (en) * 2016-02-18 2016-09-21 青岛克路德机器人有限公司 Fire scene fire source positioning method and system and fire-fighting robot
CN105931411A (en) * 2016-06-14 2016-09-07 广州东亚保安服务有限公司 Firefighting remote monitoring and early warning platform of multistoried building and realization method of platform
CN105913605A (en) * 2016-06-15 2016-08-31 宝鸡通茂电子传感设备研发有限公司 Intelligent identification method and system of flame sensor
CN106020052B (en) * 2016-07-14 2018-06-15 国网山东省电力公司平原县供电公司 A kind of converting station electric power equipment automatic temperature measurement prior-warning device
CN106274776A (en) * 2016-08-31 2017-01-04 宁波力匠机器人科技有限公司 A kind of robot arm
CN106411873A (en) * 2016-09-21 2017-02-15 北京光年无限科技有限公司 Data interaction method and system based on intelligent robot
CN106383515A (en) * 2016-09-21 2017-02-08 哈尔滨理工大学 Wheel-type moving robot obstacle-avoiding control system based on multi-sensor information fusion
CN106468717A (en) * 2016-10-18 2017-03-01 秦艳霞 Biochip for HUMAN HEAT SHOCK PROTEINS Hhsp HSP70 detection
CN106530571A (en) * 2016-10-28 2017-03-22 江苏金米智能科技有限责任公司 Intelligent firefighting system
CN106362338B (en) * 2016-11-30 2019-05-17 汕头大学 A kind of fire-fighting robot of motion sensing control, control system and control method
CN106683329B (en) * 2017-01-19 2019-01-01 浙江科技学院 For coping with the site disposal robot of security risk event within the scope of plane domain
CN107093302A (en) * 2017-03-30 2017-08-25 常州市武进区半导体照明应用技术研究院 Intelligent building fire video monitoring system based on CAN
CN106955443A (en) * 2017-04-18 2017-07-18 南京三宝弘正视觉科技有限公司 A kind of fire handling machine people and system
CN107368073A (en) * 2017-07-27 2017-11-21 上海工程技术大学 A kind of full ambient engine Multi-information acquisition intelligent detecting robot system
CN108053605A (en) * 2017-12-25 2018-05-18 南京南邮信息产业技术研究院有限公司 A kind of intelligent video analysis system and method based on wireless internet of things
CN108198371A (en) * 2018-02-28 2018-06-22 王婧瑶 A kind of fire hazard monitoring robot
CN108785942A (en) * 2018-05-10 2018-11-13 上海钧工智能技术有限公司 A kind of fire-fighting robot
CN109931101A (en) * 2018-10-21 2019-06-25 倪秀英 A kind of miniature automatic closing door in cable tunnel
CN109377709A (en) * 2018-12-05 2019-02-22 厦门泰消防科技开发有限公司 Intelligent domestic Yan Gan robot
CN109745643A (en) * 2018-12-17 2019-05-14 义乌市吉龙科技有限公司 A kind of visual identifying system for Intelligent fire-fighting robot
CN109908523B (en) * 2019-01-15 2020-01-31 特斯联(北京)科技有限公司 Mobile intelligent fire extinguishing device based on Internet of things communication and control system
CN110115819A (en) * 2019-05-14 2019-08-13 上海荷福人工智能科技(集团)有限公司 Fire water monitor orientation early warning and extinguishing method, storage medium and fire water monitor based on artificial intelligence
CN111445659A (en) * 2020-04-03 2020-07-24 西华大学 Intelligent fire-fighting measurement and control system for intelligent community based on Internet and use method thereof

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101635078A (en) * 2009-05-27 2010-01-27 北京奥腾讯达科技有限公司 Intelligent fire scene detecting device
CN102513987A (en) * 2011-12-26 2012-06-27 浙江金刚汽车有限公司 Intelligent fire-fighting rescue robot
CN103550883A (en) * 2013-11-04 2014-02-05 国家电网公司 Fire-fighting robot for transformer substation
CN103845834A (en) * 2012-12-04 2014-06-11 青岛三利中德美水设备有限公司 Fire rescue robot
CN203861819U (en) * 2014-04-10 2014-10-08 李洁 Multifunctional fire fighting truck system

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100690658B1 (en) * 2004-12-29 2007-03-09 엘지전자 주식회사 Fire extinguish apparatus and method in using robot cleaner

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101635078A (en) * 2009-05-27 2010-01-27 北京奥腾讯达科技有限公司 Intelligent fire scene detecting device
CN102513987A (en) * 2011-12-26 2012-06-27 浙江金刚汽车有限公司 Intelligent fire-fighting rescue robot
CN103845834A (en) * 2012-12-04 2014-06-11 青岛三利中德美水设备有限公司 Fire rescue robot
CN103550883A (en) * 2013-11-04 2014-02-05 国家电网公司 Fire-fighting robot for transformer substation
CN203861819U (en) * 2014-04-10 2014-10-08 李洁 Multifunctional fire fighting truck system

Also Published As

Publication number Publication date
CN104299351A (en) 2015-01-21

Similar Documents

Publication Publication Date Title
CN104143248B (en) Forest fire detection based on unmanned plane and preventing control method
KR101725774B1 (en) Smart Fire Fighting Evacuation System
US9819911B2 (en) Home, office security, surveillance system using micro mobile drones and IP cameras
CA2851143C (en) Emergency services system and method
US9811990B2 (en) Virtual enhancement of security monitoring
CN102622861B (en) Security monitoring alarm system based on geographical position and method therefor
CN103285540B (en) Suspension type intelligent fire robot
EP3032508B1 (en) Integrated camera awareness and wireless sensor system
CN103726879B (en) Utilize camera automatic capturing mine ore deposit to shake and cave in and the method for record warning in time
TWI448993B (en) An environmental safety monitoring escape-guidance system and its method of warning
US20170092138A1 (en) Drone detection systems
CN103191533B (en) Intelligent fire monitoring and evacuation indicating system and method
US10467885B2 (en) Drone-augmented emergency response services
US10698403B2 (en) Robotic assistance in security monitoring
CN102360082B (en) Monitoring system for power transmission line
DE60220303T2 (en) Information collecting device, information collecting procedure, information collection program, recording medium, that the information collective program contains and information collection system
CN103273982B (en) A kind of Multifunctional all-terrain bio-robot
US9536410B2 (en) Digital fingerprint tracking
RU2401947C2 (en) Underground system of monitoring, annunciation and determination of mine worker location
CN101673448B (en) Method and system for detecting forest fire
RU2293998C2 (en) Lidar-based system provided with computer control for smoke identification, particularly, for revealing forest fire at early stage
CN104052151A (en) Intelligent auxiliary comprehensive monitoring and D5000 comprehensive linkage platform of transformer substation
DE69933056T2 (en) Monitoring system
CN104079874B (en) A kind of security protection integral system and method based on technology of Internet of things
KR101914228B1 (en) IOT Based Safety Management System for Construction Site Using the RTLS and Video Control Technology

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
GR01 Patent grant
C14 Grant of patent or utility model