Hyderabad: A professor and a student from the International Institute of Information Technology, Hyderabad (IIIT–H), have developed an automated solution to improve aerial surveillance, which has been deployed at Bharat Electronics Limited (BEL).
Although the existing tracking mechanism used by the Indian Air Force provided an accuracy of 91 percent, the AI-assisted tracker created by IIIT–Team H demonstrated 96 accuracy percentageaccording to a statement.
In the Defense Forces, all flying objects in Indian airspace, including potential threats and commercial aircraft, are detected and tracked by ground radars produced by BEL. The current system used by the The IAF relies on a multi-sensor tracking mechanism (MST) via radars located in different parts of the country.
With overlapping radars sometimes detecting the same aircraft and a delay in communication between sensors, two common errors emerge. One is the “merge” error — where multiple aircraft in close proximity to each other are incorrectly identified as one and “separated” — where a single aircraft is detected as multiple and incorrectly reported as a threatstates the press release.
To resolve these issues while retaining the original tracking mechanism, a team from BEL Ghaziabad contacted IIIT–H for the development of an automated solution.
IIIT–Research team H, led by Professor Paruchuri with Masters student Anoop Dasika, trained a machine learning model with 11 days of anonymized and labeled data collected from 17 million data points captured by various radars. The algorithm they developed solves problems with merging and splitting errors, resulting in a 5% increase in accuracy. Another advantage is that it helps in detailed radar analysis.
“By identifying which radars contribute to the maximum number of errors, and which error in particular, such a probe helps to prioritize the replacement or repair of the radars concerned”, noted Teacher. Paruchuri.
The software has been transferred to BEL and is currently being tested in their simulation environment.