Developing a Proof of Concept of Search and Locate Drone Swarm for disaster hit areas in Pakistan.

The age of technology is evolving rapidly, and Pakistan has to keep up with the rest of the world. Today the world is dominated by technological marvels as the private companies continue to provide infrastructure for day-to-day solutions. One of the most emerging fields in the modern era is

Project Title

Developing a Proof of Concept of Search and Locate Drone Swarm for disaster hit areas in Pakistan.

Project Area of Specialization

Artificial Intelligence

Project Summary

The age of technology is evolving rapidly, and Pakistan has to keep up with the rest of the world. Today
the world is dominated by technological marvels as the private companies continue to provide
infrastructure for day-to-day solutions. One of the most emerging fields in the modern era is the use of
drones or Unmanned Aerial Vehicles. According to PwC report, the commercial drone market is already
worth 127 Billion dollars and this market is rapidly expanding, but the technology still demands maturity
for the commercial applications. One such technology promises to make the drones more efficient,
effective, scalable and a cost-effective solution, the Swarm Intelligence.
This new branch of artificial intelligence is mostly inspired by the collective behavior of swarms in
nature, such as colonies of ants, flocking of birds, honey bees. It involves distributed forms of control
(decentralized control) to increase performance and robustness of the applications. Swarm Intelligence
algorithms provide rapid solutions for complex problems and we want to incorporate this new
technology into a network of multirotor systems. By starting this initiative, we will serve to provide
foundations for cost-effective yet powerful applications with drones, ranging from smart surveillance to
search and rescue operations.

Project Objectives

Situational awareness is paramount in a disaster situation. For implementing the swarming behavior in
drones, we will be focusing on humanitarian assistance in times of a natural disaster. In the aftermath of
a disaster, one of the biggest impediments is to locate and reach out to the people needing vital help.
Since a disaster can cause different degrees of damage to multiple regions of an area, most affected
areas are often not prioritized for humanitarian aid; consequently causing more casualties, as there is no
way to measure, analyze and monitor the overall devastation.

Pakistan lies in the monsoon region and faces heavy rainfall, resulting in massive flooding. In 2010 alone,
one-fifth of Pakistan's total land area was hit by floods. Almost 20 million people were affected by
flooding and 2,000 people died. With sufficient use of scalable technology, the loss of human lives can
be minimized.

Project Implementation Method

Steps to initiate swarm intelligence on drones
Our algorithms will not heavily rely on GPS technology for the practical use of UAVs in the GPS dead
zones.

Test Phase: Implementation of Swarming Behavior
1: Research and algorithm analysis of Smarm Intelligence.
2: Software simulations using ROS (robot operating system).
3: General swarm Implementation on hardware.

Phase 1: Aerial Survey and Mapping with drones
The UAVs will scan an area autonomously while capturing multiple photographs by following an
algorithm given in the next pages. These photos will be sent to the server for Photogrammetry, which is
the creation of topographic maps by compiling/stitching multiple photographs taken in an area. This
process can be achieved using OpenDroneMap, an open source software, along with a modified python
script.

Phase 2: Remote Sensing for Geographic Information System (GIS) on Server
After the photogrammetry process, each and every structure will be examined to quantify the damage
done in a percentage. The neighboring buildings will also be inspected for damage, after which they will
ultimately become a part of a cluster; consisting of damaged houses, buildings, trees, and flood. These
clusters will highlight on a map, and then all of the information along with geospatial data will be shared
with the drones.

Phase 3: Swarm performing search and rescue operations
After the drones have acquired the necessary data, they will collectively decide, cooperate and
distribute themselves into multiple regions of an area for search and locate operation. For this purpose,
the following swarming algorithm (tentative) can be incorporated.

1. Attract to the nearest site of the disaster region (cluster) that has more damage.
2. If another drone is present in that cluster region, then swarm in a formation.
3. Cover the entire region of an area for live monitoring.
4. Perform Search and Locate operation in the devastated region while detecting SOS signals
generated by humans.
5. Switch between the cluster regions in case there are more clusters than the UAV units.

Benefits of the Project

Feasible regulations of drone usage for the public have opened a gateway to breakthroughs in
commercial technology. However, the applications for Smarm technology are yet to be implemented.
For achieving the swarm intelligence, we can take natural inspirations for specific purposes. For
instance, enhanced algorithms can be crafted from the behavior of vulture bird swarms and then these
algorithms can further be implemented on swarms of drones for a broad spectrum of applications such
as search rescue operations, distributed wide area surveillance, crowd monitoring, performing disaster
estimation on an area, environmental monitoring, and Cloud Seeding.

Technical Details of Final Deliverable

We purpose a cost-effective solution that would utilize multiple autonomous drones to quickly survey
an area plus analyze the damage by employing artificial intelligence and computer vision for remote
sensing an area. After which the drones would be able to decide where to perform search and rescue
operations.

Final Deliverable of the Project

HW/SW integrated system

Type of Industry

IT , Manufacturing , Security

Technologies

Artificial Intelligence(AI), Robotics

Sustainable Development Goals

Industry, Innovation and Infrastructure

Required Resources

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