Design and Development of Drone Swarm for monitoring and surveillance

Using drones for monitoring and surveillance is much more efficient as compared to our conventional methods. Drones are significantly in use for the purposes of disaster preparedness and management all around the world as they can take on the tasks where manned vehicles and relief workers

Project Title

Design and Development of Drone Swarm for monitoring and surveillance

Project Area of Specialization

Robotics

Project Summary

Using drones for monitoring and surveillance is much more efficient as compared to our conventional methods. Drones are significantly in use for the purposes of disaster preparedness and management all around the world as they can take on the tasks where manned vehicles and relief workers fall short. This project aims to design and develop a workable swarm system which can serve the purpose of surveillance by observing research and development in this field. Introduction of swarm technology for surveillance can help reduce damage in case of disaster to a great extent and serve a wide range of other purposes.Drone swarms can consist of 3 to 100 drones based on the user demands and scale of application. The project starts off with selection of appropriate components for assembly of drones. Selection of motors is based on the takeoff weight of the drone. CFD of drone frame and propeller is performed to calculate drag, pressure, and wake region of the drone and determine minimum distance between the drones. For the flight automation and coordination, virtual drones which match best with F-450 are spawned and controlled on ROS and the code is then fed to the microcontrollers for each drone. ROS code burned on Raspberry pi is used for programming of physical system. Testing includes straight line path following for the leader drone and follower drone following the leader at an appropriate distance.

Project Objectives

Drones can reduce the overall response time by 44.46%. The aim of this project is to design and develop outdoor swarm network for monitoring and surveillance which can aid the disaster management processes as well as make the security systems more efficient by providing faster and easier access to multiple areas at once. Our solution also aims to make industrial surveillance easy and less expensive This multi-modular project will provide real-time communication between the disaster site and the planning department through multiple unmanned aerial vehicles that scout the disaster site and simultaneously transmit data about people in distress. For the first module that we will be working on, hardware and software integration for simple multi robot system for outdoor autonomous flight will be provided.

Project Implementation Method

We implemented our plan in following steps: 

• Part Weight Estimation
• Static Analysis
• Motor and Propeller Selection
• Weight Analysis
• Assembly of a Single Drone
• Callibration of Flight Controller
• Mission Path Planning
• Troubleshooting
• Assembly of Second Drone
• Programming for Swarm
• Data Reception
• Testing and Troubleshooting

Initially we planned to use a total of three drones for swarm but due to time and cost constraints, we will be providing two fabricated drones [coordinated in swarm].

Benefits of the Project

The application of swarm system can revolutionize Pakistan’s disaster management methods to a great extent and can be of substantial importance in saving lives and economy. Surveillance with drones can help us detect anomalies in a much less time than other practices. It is approximately 5 times faster than manned surveillance.  For Pre-DRR(disaster risk redution), drones can be utilized for surveys, mapping risk profiles and identification purposes. For DRR, drones come in handy in risk mitigation, disaster preparedness etc. Additionally, drones can instigate on-ground action in case of disasters/other anomalies much faster than the manual process. Strategical application of drones can also contribute to efficient consumption of resources and saving more lives. Moreover, drones can be utilized in medical emergencies for surveillance and then aiding on-ground teams. UAVs are much more beneficial than manned vehicles in case of rescue and search operations as they can cover a much larger area at once and even cover hard to reach areas, thus reducing staff number and costs to some extent. 

Technical Details of Final Deliverable

This project has both mechanical as well as programming parts. Off the shelf products were taken to assemble the drone and programming for swarm communication and formation was worked on. 

• Quadcopter was chosen for this project as it has a more robust controls system than fixed-wing controls system. In case of any calamity, quadcopter can hover, tilt at 90° to pass through tight spaces, take sharp turns and be more resilient to cross winds in comparison to fixed wing. 
• Quadcopter configuration "x" was chosen over "+" configuration because it is a more stable configuration. "+" configuration has easier maneuverability for the purpose of acrobatics, but a project in which next phase shall be mounting of camera and real-time data collection and processing through the camera, a more stable drone is required for the job. 

• The drones are programmed on ROS  instead of MATLAB. This is because ROS is open source and more used in the robotics industry itself. 

• For swarm surveillance and reconnaissance, simulations have shown that Voronoi Partition Algorithm is best suited as it divides the AOI and sends each drone at a designated node. 

• For basic quadcopter outdoor flying, DJI FlameWheel 450 Quadcopter Frame was selected. Its material is PA66+30GF with frame weight of 282g and takeoff weight carrying capacity is 1600g. Drag calculations were made on the frame in ANSYS to verify the aerodynamic proficiency at our required speed.

• For an F450 drone of takeoff weight around 1kg, motors of 920-1500 KV are generally used. For this project, DJI 2212 920 KV motor was used. 

• Screw Sizes for DJI 450 are M3x8 and M2.5x6. Different parts of frame have different thread depth. for motors, M3x8 screws were used. These are recommended in the DJI F450 User Manual.

• 1045 Multiaxial CW CCW ABS Blade propellers (diameter:10 in, thread pitch:4.5inch) were used as they are most compatible with 920 KV motor.

• To control the rpm of each motor, a DJI specified ESC was used. LiPo battery (11.1 V) 3S 5500 mAh is most compatible with ESCs and 920 KV motor.

• 0m50 PDM is connected to the Power Distribution Board of the frame to provide stable Voltage throughout the circuit.

• For flight stability Pixhawk 2.4.8 is used. 

• Since it is an outdoor drone system, for Localization and Navigation of each drone, a GPS Module M8N was mounted on it. It couples with IMU Sensor of Pixhawk to minimize localization errors. Raspberry Pi 4 and Raspebrry pi 3b+ were used as microcontrollers in leader and follower drone respectively.

• Drones use the difference in their GPS coordinates to keep a minimum distance between them and avoid collision.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Security

Other Industries

Media

Core Technology

Robotics

Other Technologies

Sustainable Development Goals

Industry, Innovation and Infrastructure, Partnerships to achieve the Goal

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com