The use of drones is officially legal in India, which is a benefit to several industries including for safety, agriculture, construction, energy, and others. For example, drones can be used to survey land, buildings, or transmission wires. They can also be used for search and rescue, or to map roads or traffic, drop objects (such as medicine or supplies) in remote locations, spray pesticides at farmlands, and more.
The legality of drone operation is proving excellent timing for a team of students of the IT engineering program at Pune Institute of Computer Technology in India. The students — which have nicknamed themselves Team Dronebolt — have created two drones and an algorithm that’s capable of coordinating eight or nine drones while in flight together.
The advantage of multiple drones working together is that if one fails, the other ones can fill any gaps in the data being attained, for example. Of course, with higher levels of operational complexity (such as multiple drone use), there are increased levels of risk.
Although the idea of synchronized drone use is not entirely new, the team says the algorithm they have produced is unique. “The existing drones comprise of a master, but there are numerous limitations, and the chances of failure are high,” explains Siddhant Nikumbh (19), leader of Team Dronebolt. “If the master fails, the complete system gets delayed. But we have created a sort of virtual node, which acts as the master.”
Nikumbh worked with Mohit Arora (19), Japjyot Gulati (20), Pallavi Dadape (19), Ashwin Kotgire (19), and Prajakta Lanje (19).
According to Team Dronebolt, they’ve ensured that the chances of failure are zero because there’s a virtual connection that, essentially, serves as the master drone. It seems others agree. Judges at the grand finale of the Smart India Hackathon-hardware edition, organized by REVA University in Bengaluru, awarded the team a win and Rs 1 lakh for their work.
“All the errors were the best teachers,” says Nikumbh. So, what’s in their drones? A recording camera that temporarily stores data on a single board computer device, Raspberry Pi. The students also created an Apache web server where all the data gets transferred.
To ensure proper mapping of the flight area, the team implemented a simultaneous localization and mapping (SLAM) algorithm. Thanks to the Raspberry Pi camera model V2, it can take photos of up to two acres away. With upgrades, even greater distances and accuracies are possible. Additionally, the drones can be modified based on a variety of use cases.
Determined to succeed, the students each contributed Rs 25,000 to finance the project, which cost them Rs 1.5 lakh in total. They sourced several journals and online sources for information on drones, including YouTuber Dennis Baldwin. He regularly showcases his work with robotics and drones on Painless360, as well as other YouTube channels.
“For the APIs used in developing the device…we somehow went through the full API operations and figured out how to use it,” says Nikumbh.
Since winning the Hackathon, Team Dronebolt has been preparing to take part in another competition at the Indian Institute of Crafts and Design, coordinated by AICTE. AICTE is the All India Council for Technical Education, statutory body and a national-level council for technical education, under the Department of Higher Education.
The students are also planning to work on more advanced drones, including object delay and manned drones. They’ve also heard from the government of India.
“We are in touch with a government official and are working with security. As of now, we hold three to four proposals, which they will shortlist,” Nikumbh says.
Filed Under: Blog entry, Tech Articles