Sensor Image and UAV based Platform for disaster Management

Project Title

Sensor Image and UAV based Platform for disaster Management

Project Area of Specialization Internet of ThingsProject Summary

Nowadays various types of disaster are occurring in the whole world. With the passage of time new ways are being introduced to tackle these situations .Science is playing vital role to minimize the effect of these disaster with the advancement in technology and research studies. When the buildings are demolished and the disaster area becomes completely destructed then its extremely difficult to locate the disaster location because of blockage of paths and approximately accurate statistics of the persons inside the building due to demolish of building because of disaster. The communication network become disabled so there exist a loss of human lives because of late rescue process. In this research project, we proposed building a management system consisting of sensor, image and UAV for efficient disaster management. The project is equipped with 4IR (fourth Industrial revolution) technologies involving next generation of IOT devices, Big Data and Machine learning algorithm. The essential features of the project is the communication linkage establishment from control center to UAV then UAV to ground station and then transfer of information back to control center.

The implementations of the project consist of the following module:

1. Manually deploy the UAV and then send the real-time blockage detection information to control center.
2. Person counting data through kinect depth sensor.
3. Sending person counting data from ground station raspberry pi to UAV raspberry pi.
4. Sending all the data of blockages and persons counting back to control center.

Project Objectives

       1.    Person counting through Kinect depth sensor.

The person counting is done through kinect depth sensor and then the persons data will be sent to raspberry pi. After all the data of person counting is sent to raspberry pi of ground station then that raspberry pi will be sent to the raspberry pi of UAV in the form of c.s.v file.

        2.   Blockage detection in real-time through UAV.

Real-time blockage detection will be done through the through UAV. The data of blockage detection in case of 1,0 (blockage detected=1,no blockage=0) and GPS co-ordinates will be sent to control center in the form of c.s.v file.

        3.  Communication link establishment.

Communication link establishment between control center and UAV raspberry pi,  then UAV raspberry pi to ground station raspberry pi (containing person counting data) and then sending back all the information(including blockage information and person counting data) to control center. All the data will be send in the form of c.s.v file.

   

                                                             

Figure 1: Overview of the Objectives of  proposed Project Sensor, Image and UAV based platform for Disaster Management system.

Project Implementation Method

    

Figure 2, Figure 3, Figure 4 and Figure 5 shows the over view of the proposed system. The system is equipped with kinect depth sensor, raspberry pi’s and UAV. The UAV is equipped with  APM 2.8 flight controller, 450 quad copter frame,  Radio link T8FB ,  Quad copter legs,  Gens 4000 45c 3s LiPo battery, 30A Escs ,A2212 1000kv motors ,10 X 4.5 props set, ,Rf7 simulator , B6 Charger, Telemetry (TX, RX),  GPS. Kinect depth sensor is connected to raspberry pi .The persons counting are done through the kinect depth sensor and then all the data will be stored in raspberry pi. In case of disaster the UAV will be manually deployed by the control center and then the UAV will detect blockages by taking images in real-time and after comparing with Yolo v3 trained model images  send the information of whether the blockages are detected or not  along with GPS co-ordinates to the control center in case of c.s.v file. When the UAV raspberry pi comes in the range of ground station raspberry pi (having person counting data) then the person counting data is sent to UAV raspberry pi and that’s how raspberry pi to raspberry pi communication is done. After that the UAV sends both person counting data and blockage detection data to control center and that’s how raspberry pi to pc communication is done.

Figure 2: Overview of the Architectural Diagram of person counting process.

Figure 3: Overview of the Architectural Diagram of blockage detection process.

Figure 4: Flowchart Overview of the communication between PC to UAV raspberry pi process.

Figure 5: Flow chart Overview of the communication link establishment of Raspberry pi of UAV to Ground station raspberry pi.

Benefits of the Project

Easy to implement: This project is easy to implement in any environment within specific communication ranges.

Convenience for rescue processes: the project will provide convenience for rescue processes as manual detection of path blockages is difficult .Through blockage information the rescue operation can be implemented more fastly in comparison of manual detection.

Approximately accurate statistics of persons inside the building at disaster point: Through this project almost accurate statistics of person at disaster point inside the building will be gathered. So, by doing this the person detection process will become convenient and many of human lives will be saved.

Flexibility of the project:

The project is made flexible to adopt amendments if needed. This project can be utilized for any type of disaster in which paths are blocked and buildings are demolished. Also where person counting contains significant importance.

Low Cost: An efficient management system is built which is low cost and will be able to do save human lives with less delay compared to the manual processes of rescue.

Efficient Management System:

An efficient management is made to detect blockages and person counting in real-time.This project can be further utilized for

• Saving lives
• Save time
• Better quality of management and life.
• Improvement of safety factor.

Technical Details of Final Deliverable

• Technology development for establishing communication link between UAV on board raspberry pi to control center PC ,UAV on board raspberry pi to ground station raspberry pi.
  • Kinect depth sensor interfaced with ground station raspberry pi
  • Client server configuration is done while maintaining communication among UAV, Ground station and control center.
• Software technological developmental algorithm for image classification in case of person counting and blockage
  • YOLO V3.

• Technical reports of implemented systems.
  • Finalized achieved results in form of tables, graphs and charts.

• Specification of equipment/tools used in the management system.
  • Person detected within 2 to 4m of the coverage area with an angle of 35 degree.
  • Data set of blockage detection is pre-processes at 15 FPS.
  • The communication link is established when the UAV is within 40 m range of control center .
  • The communication link between onboard raspberry pi to ground station raspberry pi is established when the UAV is within 2m range of ground station raspberry pi.

Final Deliverable of the Project HW/SW integrated systemCore Industry SecurityOther Industries Manufacturing , Others , Telecommunication Core Technology Internet of Things (IoT)Other Technologies Artificial Intelligence(AI), RoboticsSustainable Development Goals Industry, Innovation and Infrastructure, Sustainable Cities and Communities, Climate ActionRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	70000
Raspberry pi 4	Equipment	2	22500	45000
Pi camera	Equipment	1	1000	1000
UAV battery	Equipment	1	4000	4000
GPS	Equipment	1	1000	1000
UAV battery charger	Equipment	1	5500	5500
Kinect Sensor	Equipment	1	3500	3500
Miscellaneous equipment for assembling	Miscellaneous	2	5000	10000