Small Aerial Vehicles' Collaboration Using CoSLAM

Project Title

Small Aerial Vehicles' Collaboration Using CoSLAM

Project Area of Specialization Artificial IntelligenceProject Summary

Unmanned aerial vehicles (UAV) are receiving popularity to use them as moving surveillance machine.  The are usually tele-operated to achieve the desire goals such guarding or investigating an unknown area.  We use multiple UAVs to perform a complex task of surveillance.

Usually the coordination or communication among multiple UAVs is either lacking or not present.  The collaboration among multiple UAV enhances their functionality and efficiency.

This project aims to build a coordination system using the simultaneous localization and Mapping (SLAM) among multiple drones to coordinate in an unknown environment efficiently.  Aerial Vehicles collaboration using CoSLAM

The SLAM algorithm enables Unmanned Aerial Vehicle (UAV) to map an unknown and unstructured building from different dimensions.  It can also follow the pre-programmed path autonomously. in our project the data is collected by three quadcopters. The quadcopters will navigate autonomously with the help of CoSlam and collaborate with each other and collect the images of a particular building from different dimensions using Cameras. The data will be transmitted to our server.

Project Objectives

Our focus is to use multi aerial vehicles using CoSlam methodology for the indoor environments such as manufacturing environment monitoring tasks ,3d imagery data collection it is specially used in construction environment .it will help to improve.

The experimental evaluation of the developed methods and algorithms is an important objective.

 manufacturing and construction areas the specific aims of this project are as follow

1.  Multiple drone robots movement in unstructured environment.

2.  Localization and Mapping of each drone in multiple drone environment additionally GPS tracking of individual drone

3.  Building a global shared map for all drones to use and move in the environment.

4.  Collaboration in the shared and local map build by the drones.

5. Incorporating visual camera to the quadcopter to enable the computer vision algorithm to work with quadcopter

6. Design the Simultaneous localization and Mapping (SLAM) for the quadcopter to build a map for autonomously in the environment

Project Implementation Method

We will use computer vision and machine learning techniques to develop to develop map of the unstructured area and localize the robot in the unknown environment.

We build a simultaneous localization and Mapping (SLAM) and coordination of aerial vehicles in that map to move smartly in the known environment.

The proposed methodology includes pixel classification and detecting objects through sensor and make the map of indoor building. At the initial step, acquired indoor image of building are preprocessed to remove noise and prepare them for further steps.

(Figure 1) A robot in the indoor environment to do localization and collaboration with other drones.

Benefits of the Project

1. A step towards the producing intelligent machines and collaboration among them.

2. Investigate the area by multiple machines together as the work done by multiple humans.

3. Search a suspicious object by the multiple drones.

4. A smart and optimized moving surveillance by multiple drones. 

5. Reduce human efforts in construction development areas.

6. Rescue operations and healthcare..

7. Archeological surveys. 

8. Geographic Mapping.

Technical Details of Final Deliverable

We build a multi-drone environment in which drone collaborate.

1. The quadcopter will map the environment by use of Visual  Algorithms such as SLAM.

2. We build a global shared map among all drone. There one map used by all drone to coordinate.

3. Comparing the local and global map.

4. Sharing the drone location in the map build by the drone.

Final Deliverable of the Project HW/SW integrated systemType of Industry IT Technologies Artificial Intelligence(AI), RoboticsSustainable Development Goals Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	80000
quad-copter Bebop 2	Equipment	2	35000	70000
stationery, travel, wiring	Miscellaneous	1	10000	10000