Dr Awadhesh Dixit, Dr Ravi Kant Kumar, Assistant Professors and M.Tech. student Ms Meesala Sai Divya from the Department of Computer Science and Engineering have published a new patent titled “An Adaptive Anti-Drone Defense System for Protection of Critical Infrastructure Facilities” with the application no – in the Patent Office Journal. This patent presents a drone defense system designed to autonomously detect and mitigate rogue drones near critical sites like airports and government buildings, utilizing a combination of advanced sensors and AI technology.
Abstract
An adaptive anti-drone system to achieve the protection of critical sites. This model incorporates LiDAR, RF scanning, and computer vision to identify, track, and categorize aerial threats. An onboard AI fusion engine evaluates the threat level on the basis of the trajectory, speed, and proximity, and chooses the most suitable countermeasures, ranging between GPS spoofing and RF jamming, all the way to net launchers and kinetic interceptors. Engagements make the system learn constantly to minimize false positives, enhance the performance, and ensure reliable deployment in sensitive and high-risk settings.
Explanation of the Research in Layman’s Terms:
This study establishes a drone defence mechanism that is capable of using autonomous detection and prevention of rogue drones around strategic locations such as airports, power plants, and government buildings. The use of drones is very universal nowadays; however, they may cause a significant threat, like spying, drug smuggling, or interference with operations. The system is a smart security screen for the sky. It relies on multiple types of sensors, such as the LIDAR (laser light mapping) sensors, radio-frequency (RF) sensors, and cameras powered by AI to detect and recognize flying objects. It is then programmed by an artificial intelligence engine to determine whether it is a drone, how harmful it can be, and the next step to take. The system may react differently, depending on the circumstances: it may interfere with the signal of the drone, lure it into a landing, or net other interceptor drones. Everything can also be monitored by the security operators on a central dashboard and intervene where necessary. One of its strengths is that the system can even identify the so-called silent drones that do not transmit radio signals, and it will become more precise with time as it learns through experience. The success rates of testing were high, with a detection rate of more than 98% and a rapid response time; as a result, this makes it a strong solution to counter the threat of the evolving drones in the sensitive areas.
Practical Implementation /Social Implications of the Research
The steps for the practical implementation of the drone defense system are as follows:
- Prepare environment: Create a Python 3.11+ virtual environment and activate it.
- Implement the sensor simulation module: generate realistic synthetic sensor inputs (LIDAR, RF, camera metadata, location).
- Implement threat analysis/fusion module: fuse sensor streams, run classification, and compute threat score and confidence.
- Implement countermeasure recommendation module (modules/countermeasures.py): map threat score + context -> recommended action.
- Implement data storage & logging (modules/data_storage.py): persist detections, metrics, and allow analytics.
Deployment & demo
- For on-site demo: run app on an edge device (small server/NUC) near sensors; use HTTPS for remote access.
- For cloud demo: provision VM, expose Streamlit via reverse proxy, but remember sensor data latency and legal constraints.
- Backup: scheduled exports of detection DB and model snapshots.
Future Research Plans
The research team will focus on developing a security-enabled anti-drone communication and coordination management system to enhance UAV network performance and prevent critical infrastructure in a more secure and optimal way through advance blockchain techniques.
