Flame monitoring trolley based on visual identification

Abstract

The utility model discloses a flame monitoring trolley based on visual identification, which belongs to the field of fire fighting equipment and comprises an upper-layer acrylic plate, a lower-layer acrylic plate, a Raspberry Pi 4B, a camera, a chip module, a power supply battery, a driving mechanism, an identification mechanism, a fire extinguishing mechanism, a shell and a supporting copper column arranged between the upper-layer acrylic plate and the lower-layer acrylic plate. According to the utility model, through the Raspberry Pi 4B, the chip module, the camera and the identification mechanism, the flame monitoring dolly can carry out continuous tour inspection according to a pre-designed route and autonomous navigation, discover and position a fire source in the first time and reduce fire response time, and meanwhile, the fire extinguishing mechanism is equipped to extinguish the fire source, so that the fire extinguishing efficiency is improved. The investment of human resources is reduced, and a large amount of financial cost is saved; meanwhile, the system is low in design cost, can replace manpower-intensive fire alarm monitoring operation, and improves the intelligent level of fire prevention.

Description

Technical Field

[0001] The utility model belongs to the field of fire-fighting equipment, and in particular relates to a flame monitoring vehicle based on visual recognition. Background Art

[0002] At present, in large warehouses and other places without real-time supervision, traditional monitoring has met the requirements of daily monitoring of most warehouses with its advantages of low cost and wide coverage. However, traditional full-coverage monitoring relies on human eye recognition for fire detection, which is inefficient and easy to miss and cause huge accidents. Existing fire-fighting robots are not intelligent enough, bulky in size, and their functions are limited to firefighting without detection capabilities. Their usage scenarios are relatively simple and require human control. They cannot act according to the on-site situation on their own, so their use is limited and not universal. Utility Model Content

[0003] In view of the above-mentioned deficiencies in the prior art, the utility model provides a flame monitoring vehicle based on visual recognition, which can solve the common problems of low efficiency, easy missed judgment, and large consumption of manpower and material resources in the current fire monitoring.

[0004] In order to achieve the above-mentioned utility model purpose, the technical solution adopted by the utility model is: a flame monitoring vehicle based on visual recognition, comprising: an upper acrylic plate, a lower acrylic plate, a Raspberry Pi 4B, a camera, a chip module, a power supply battery, a driving mechanism, an identification mechanism and a fire extinguishing mechanism, a shell, and a supporting copper column arranged between the upper acrylic plate and the lower acrylic plate;

[0005] The Raspberry Pi 4B is respectively connected to the camera, the chip module, the power supply battery, the driving mechanism, the identification mechanism and the fire extinguishing mechanism, and the power supply battery is respectively connected to the Raspberry Pi 4B, the chip module, the driving mechanism and the fire extinguishing mechanism.

[0006] Furthermore, the Raspberry Pi 4B is arranged in the middle of the upper acrylic plate, the chip module is arranged between the upper acrylic plate and the Raspberry Pi 4B, and the camera is arranged in the front of the upper acrylic plate; the Raspberry Pi 4B is connected to the camera via a DuPont line.

[0007] Furthermore: the power supply battery is arranged at the rear part of the lower acrylic plate, the driving mechanism is arranged at the middle part of the bottom plate of the lower acrylic plate, and the identification mechanism is arranged at the front part of the lower acrylic plate.

[0008] Furthermore: the chip module is an integrated PCB board, which is connected to the Raspberry Pi 4B through a uart serial port.

[0009] Furthermore: the identification mechanism includes: an ultrasonic obstacle avoidance module and four flame identification modules, the ultrasonic obstacle avoidance module is located in front of the lower acrylic plate, facing the forward direction of the vehicle; each flame identification module is respectively arranged on both sides of the ultrasonic obstacle avoidance module, and a flame identification module is arranged at a fixed distance.

[0010] Further: the driving mechanism includes: two stepping motors, two power wheels and a universal wheel, the two stepping motors are respectively arranged on both sides of the middle part of the bottom plate of the lower acrylic plate, and are respectively connected to a power wheel, and the universal wheel is arranged at the rear part of the bottom plate of the lower acrylic plate;

[0011] The stepper motor is connected to the IO port of the Raspberry Pi 4B via a DuPont line.

[0012] Furthermore: the fire extinguishing mechanism includes: a water storage bottle and a water spray nozzle, the water storage bottle is arranged at the rear of the upper acrylic plate, the water spray nozzle is arranged above the shell, and a water suction pump is arranged inside the water storage bottle, and the water suction pump is connected to the chip module through a DuPont line.

[0013] Further: the housing comprises: a movable cover plate, a retractable connecting rod, a side wall through hole, a rotating shaft motor, a bottom through hole, a rotating shaft and a connecting rod;

[0014] The movable cover is arranged in front of the housing at a fixed distance from the top, the telescopic connecting rod is arranged at the bottom edge of the through hole of the movable cover, and the telescopic connecting rod is connected to the movable cover;

[0015] The side wall through hole is arranged on the right side wall of the shell near the top, and the bottom through hole is arranged on the bottom of both sides of the shell to provide space for the driving mechanism;

[0016] The shaft motor is arranged on the top of the housing, passes through the housing and is connected to the Raspberry Pi 4B via a Dupont line, and is also mechanically connected to the shaft, the shaft is connected to one end of a connecting rod, and the other end of the connecting rod is connected to a water spray nozzle;

[0017] The bottom through holes are arranged at the bottoms of both sides of the shell to accommodate the power wheels of the driving mechanism.

[0018] The beneficial effects of the utility model are as follows: the utility model provides a flame monitoring vehicle based on visual recognition, which can monitor unmanned areas or areas that are difficult to monitor manually in real time, and perform efficient fire warning and prevention and control; the utility model can perform continuous and uninterrupted patrol capabilities according to a pre-designed route and autonomous navigation, can discover and locate the fire source in the first time, reduce fire response time, reduce human resource investment, and save a lot of financial costs; at the same time, the utility model has a low design cost, can replace labor-intensive fire alarm monitoring operations, and improve the intelligent level of fire prevention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is the overall structure diagram of a flame monitoring vehicle based on visual recognition;

[0020] Figure 2 This is the main structural diagram of a flame monitoring vehicle based on visual recognition;

[0021] Figure 3 This is a bottom structure diagram of a flame monitoring vehicle based on visual recognition;

[0022] Figure 4 This is a shell structure diagram of a flame monitoring vehicle based on visual recognition;

[0023] Among them: 1. Upper acrylic plate; 2. Lower acrylic plate; 3. Support copper column; 4. Raspberry Pi 4B; 5. Camera; 6. Chip module; 7. Ultrasonic obstacle avoidance module; 8. Flame recognition module; 9. Power wheel; 10. Universal wheel; 11. Water storage bottle; 12. Shell; 13. Movable cover; 14. Retractable connecting rod; 15. Side wall through hole; 16. Motor; 17. Connecting rod; 18. Rotating shaft; 19. Water spray nozzle; 20. Bottom through hole; 21. Power supply battery; 22. Stepper motor. DETAILED DESCRIPTION

[0024] The specific implementation methods of the utility model are described below to facilitate technical personnel in this technical field to understand the utility model, but it should be clear that the utility model is not limited to the scope of the specific implementation methods. For ordinary technical personnel in this technical field, as long as various changes are within the spirit and scope of the utility model defined and determined by the attached claims, these changes are obvious, and all utility model creations using the concept of the utility model are protected.

[0025] like Figure 1 As shown in FIG. 1 , it is the overall structure diagram of the flame monitoring vehicle of the utility model, such as Figure 2 As shown in FIG. 1 , it is a main structural diagram of the flame monitoring vehicle of the utility model, such as Figure 3 As shown in FIG. 1 , it is a bottom structure diagram of the flame monitoring vehicle of the utility model, such as Figure 4As shown, it is a shell structure diagram of the flame monitoring vehicle of the utility model; Figure 1 , Figure 2 , Figure 3 and Figure 4 The embodiments of the present invention are described as follows:

[0026] This embodiment provides a flame monitoring vehicle based on visual recognition, comprising: an upper acrylic plate 1, a lower acrylic plate 2, a Raspberry Pi 4B 4, a camera 5, a chip module 6, a power supply battery 21, a driving mechanism, an identification mechanism, a fire extinguishing mechanism, a housing 12, and a supporting copper column 3 arranged between the upper acrylic plate 1 and the lower acrylic plate 2;

[0027] The Raspberry Pi 4B 4 is respectively connected to the camera 5, the chip module 6, the power supply battery 21, the driving mechanism, the identification mechanism and the fire extinguishing mechanism, and the power supply battery 21 is respectively connected to the Raspberry Pi 4B 4, the chip module 6, the driving mechanism and the fire extinguishing mechanism.

[0028] Among them, the Raspberry Pi 4B 4 is arranged in the middle of the upper acrylic plate 1, the chip module 6 is arranged between the upper acrylic plate 1 and the Raspberry Pi 4B 4, and the camera 5 is arranged in the front of the upper acrylic plate 1; the Raspberry Pi 4B4 is connected to the camera 5 through a DuPont line; a plurality of through holes are opened on the upper acrylic plate 1, and the through holes on the shell of the Raspberry Pi 4B 4 are aligned with the through holes, and the Raspberry Pi 4B 4 is fixed to the acrylic panel with screws and nuts, and the camera 5 is arranged in the front of the upper acrylic plate 1, and is fixed to the upper acrylic plate 1 with screws, so as to obtain an image of the moving direction of the trolley and transmit it back to the Raspberry Pi 4B 4, and the Raspberry Pi 4B 4 can perform flame monitoring through the image, thereby improving the accuracy of flame monitoring.

[0029] The power supply battery 21 is arranged at the rear part of the lower acrylic plate 2 , the driving mechanism is arranged at the middle part of the bottom plate of the lower acrylic plate 2 , and the identification mechanism is arranged at the front part of the lower acrylic plate 2 .

[0030] In this embodiment, the chip module 6 is an STM32F103C8T6 integrated PCB board, which is a model of existing integrated PCB boards and is connected to the Raspberry Pi 4B 4 through a uart serial port.

[0031] In this embodiment, the identification mechanism includes: an ultrasonic obstacle avoidance module 7 and four flame identification modules 8. The ultrasonic obstacle avoidance module 7 is located in front of the lower acrylic plate 2 and faces the forward direction of the vehicle. The flame identification modules 8 are respectively arranged on both sides of the ultrasonic obstacle avoidance module 7, and a flame identification module 8 is arranged every 5 cm. The ultrasonic obstacle avoidance module 7 and the four flame identification modules 8 are all connected to the IO port of the Raspberry Pi 4B 4 through DuPont lines. Among them, the ultrasonic obstacle avoidance module can adopt the HC-SR04 ultrasonic ranging module, and the flame identification module can adopt the FS24X flame detector. The ultrasonic obstacle avoidance module and the flame identification module are both existing models.

[0032] In this embodiment, the driving mechanism includes: two stepper motors 22, two power wheels 9 and a universal wheel 10. The two stepper motors 22 are respectively arranged on both sides of the middle part of the bottom plate of the lower acrylic plate 2, and are respectively connected to a power wheel 9. The universal wheel 10 is arranged at the front part of the bottom plate of the lower acrylic plate 2; the model of the stepper motor 22 is tb6600, and its ENA+, ENA-, PUL+, PUL-, DIR+ and DIR- interfaces are respectively connected to the IO port of the Raspberry Pi 4B 4 through DuPont lines.

[0033] In this embodiment, the fire extinguishing mechanism includes: a water storage bottle 11 and a water spray nozzle 19. The water storage bottle 11 is fixed to the rear of the upper acrylic plate 1 by hot melt adhesive. A water suction pump is arranged inside the water storage bottle 11, which can transport water to the water spray nozzle 19 through a water pipe. The water suction pump is connected to the STM32F103C8T6 integrated PCB board through a DuPont line.

[0034] In this embodiment, the shell 12 of the flame detection trolley is based on the main body of the flame detection trolley, and the design size is 50cm×50cm×30cm, including: a movable cover 13, a retractable connecting rod 14, a side wall through hole 15, a bottom through hole 20, a rotating shaft 18 and a connecting rod 17; the movable cover 13 is arranged in front of the shell 12 at 10cm from the top, and the retractable connecting rod 14 is arranged at the bottom edge of the through hole of the movable cover 13. The retractable connecting rod 14 is tightly connected to the movable cover 13, and the retractable connecting rod 14 can close the movable cover 13 to ensure the integrity of the shell of the flame detection trolley; the right side wall of the shell 12 of the flame detection trolley is provided with a side wall through hole 15, which is located 3cm from the top, and the size of the side wall through hole 15 is 8cm×5cm. The side wall through hole 15 is an optional design, and its function is to facilitate the Raspberry Pi 4B 4 Maintenance; The bottom through hole 20 is located 10 cm from the left side of the shell, and its size is 10 cm × 12 cm. The bottom through hole 20 corresponding to the left side is opened on the right side, which serves to provide space for the power wheel to facilitate the movement of the trolley;

[0035] A shaft motor 16 is arranged on the top of the shell 12 of the flame detection vehicle at a distance of 15 cm from the front and 5 cm from the side. The shaft motor 16 passes through the shell and is connected to the Raspberry Pi 4B 4 through the DuPont line, and is mechanically connected to the shaft 18. The shaft 18 is connected to one end of the connecting rod 17, and the other end of the connecting rod 17 is connected to the water spray nozzle 19. When the motor is started, the shaft can be driven to rotate, and the shaft drives the connecting rod to swing, thereby driving the water spray nozzle 19 to swing back and forth, so as to realize multi-angle water spraying. The water spray nozzle 19 can control the water spraying mode to be linear or spray type.

[0036] The working process and principle of the utility model are described below:

[0037] The utility model provides a flame monitoring vehicle based on visual recognition, which can patrol and detect flames under the control of Raspberry Pi 4B 4. Raspberry Pi 4B 4 uses PID algorithm to output control signals to a driving mechanism through three operation modes of proportional regulation, integral regulation and differential regulation, and controls the stepper motor of the driving mechanism, thereby controlling the action of the vehicle. The image is recognized by a camera 5 and returned to Raspberry Pi 4B 4. Raspberry Pi 4B 4 uses Yolo recognition algorithm to recognize and analyze the video image sent back by the camera, locate the size and position of the fire source, and use flame frequency characteristic analysis to avoid misjudgment of non-fire sources. If a flame is detected, the Raspberry Pi 4B 4 outputs a signal to the water pump in the water storage bottle 11 in the fire extinguishing mechanism, and extinguishes the fire through the water spray nozzle 19; the Raspberry Pi 4B 4 is connected to the identification mechanism to identify whether there is an obstacle in front of the car to avoid damage to the car; before the flame monitoring car is used, the pytorch environment is used to train the flame monitoring car, and a data set including flame photos and non-flame photos is used as a training set, and flame recognition training of camera images is performed based on the Yolo algorithm, so that the flame detection car can identify and analyze the flame through the camera image, and locate the size and position of the fire source.

[0038] The beneficial effects of the utility model are as follows: the utility model provides a flame monitoring vehicle based on visual recognition, which can monitor unmanned areas or areas that are difficult to monitor manually in real time, and perform efficient fire warning and prevention and control; the utility model can perform continuous and uninterrupted patrol capabilities according to a pre-designed route and autonomous navigation, can discover and locate the fire source in the first time, reduce fire response time, reduce human resource investment, and save a lot of financial costs; at the same time, the utility model has a low design cost, can replace labor-intensive fire alarm monitoring operations, and improve the intelligent level of fire prevention.

Claims

1. A flame monitoring vehicle based on visual recognition, characterized in that: include: An upper acrylic plate (1), a lower acrylic plate (2), a Raspberry Pi 4B (4), a camera (5), a chip module (6), a power supply battery (21), a driving mechanism, an identification mechanism and a fire extinguishing mechanism, a housing (12), and a supporting copper column (3) arranged between the upper acrylic plate (1) and the lower acrylic plate (2); The Raspberry Pi 4B (4) is respectively connected to the camera (5), the chip module (6), the power supply battery (21), the driving mechanism, the identification mechanism and the fire extinguishing mechanism, and the power supply battery (21) is respectively connected to the Raspberry Pi 4B (4), the chip module (6), the driving mechanism and the fire extinguishing mechanism.

2. The flame monitoring vehicle based on visual recognition according to claim 1 is characterized in that: The Raspberry Pi 4B (4) is arranged in the middle of the upper acrylic plate (1), the chip module (6) is arranged between the upper acrylic plate (1) and the Raspberry Pi 4B (4), and the camera (5) is arranged in the front of the upper acrylic plate (1); the Raspberry Pi 4B (4) is connected to the camera (5) via a DuPont line.

3. The flame monitoring vehicle based on visual recognition according to claim 1 is characterized in that: The power supply battery (21) is arranged at the rear of the lower acrylic plate (2), the drive mechanism is arranged at the middle of the bottom plate of the lower acrylic plate (2), and the identification mechanism is arranged at the front of the lower acrylic plate (2).

4. The flame monitoring vehicle based on visual recognition according to claim 1 is characterized in that: The chip module (6) is an integrated PCB board, which is connected to the Raspberry Pi 4B (4) through a UART serial port.

5. The flame monitoring vehicle based on visual recognition according to claim 1 is characterized in that: The identification mechanism comprises: an ultrasonic obstacle avoidance module (7) and four flame identification modules (8); the ultrasonic obstacle avoidance module (7) is located directly in front of the lower acrylic plate (2) and faces the direction of travel of the vehicle; the flame identification modules (8) are respectively arranged on both sides of the ultrasonic obstacle avoidance module (7), and a flame identification module (8) is arranged at a fixed distance.

6. The flame monitoring vehicle based on visual recognition according to claim 1 is characterized in that: The driving mechanism comprises: two stepping motors (22), two power wheels (9) and a universal wheel (10); the two stepping motors (22) are respectively arranged on both sides of the middle of the bottom plate of the lower acrylic plate (2) and are respectively connected to a power wheel (9); the universal wheel (10) is arranged at the rear of the bottom plate of the lower acrylic plate (2); The stepper motor (22) is connected to the IO port of the Raspberry Pi 4B (4) via a DuPont line.

7. The flame monitoring vehicle based on visual recognition according to claim 1 is characterized in that: The fire extinguishing mechanism comprises: a water storage bottle (11) and a water spray nozzle (19); the water storage bottle (11) is arranged at the rear of the upper acrylic plate (1); the water spray nozzle (19) is arranged above the shell (12); a water suction pump is arranged inside the water storage bottle (11); and the water suction pump is connected to the chip module (6) via a DuPont line.

8. The flame monitoring vehicle based on visual recognition according to claim 1 is characterized in that: The housing (12) comprises: a movable cover plate (13), a retractable connecting rod (14), a side wall through hole (15), a rotating shaft motor (16), a bottom through hole (20), a rotating shaft (18) and a connecting rod (17); The movable cover plate (13) is arranged in front of the housing (12) at a fixed distance from the top, the telescopic connecting rod (14) is arranged at the bottom edge of the through hole of the movable cover plate (13), and the telescopic connecting rod (14) is connected to the movable cover plate (13); The side wall through hole (15) is arranged on the right side wall of the housing (12) near the top, and the bottom through hole (20) is arranged on the bottom of both sides of the housing (12) to provide space for the driving mechanism; The rotating shaft motor (16) is arranged on the top of the housing (12), and the rotating shaft motor (16) passes through the housing and is connected to the Raspberry Pi 4B (4) via a DuPont line, and is mechanically connected to the rotating shaft (18), the rotating shaft (18) is connected to one end of the connecting rod (17), and the other end of the connecting rod (17) is connected to the water spray nozzle (19); The bottom through holes (20) are arranged at the bottoms of both sides of the housing (12) to accommodate the power wheels (9) of the driving mechanism.

Images