The National Institute of Technology Rourkela has introduced BHU-MANACHITRA, an autonomous real-time land mapping drone system that can generate land maps without internet access, external computers, or manual processing. The innovation integrates Artificial Intelligence with Unmanned Aerial Vehicle technology to identify land features in real time, addressing long-standing limitations in traditional land mapping and drone imaging.

The patented system (Patent No. 573507) was developed by a research team led by Prof. Sambit Bakshi, Associate Professor, Department of Computer Science and Engineering, along with Dr. Tanmay Kumar Behera, Dr. Debbrota Paul Chowdhury, and Ms. Camellia Ray. Supported by the Vishlesan I-Hub Foundation at IIT Patna under the National Mission on Interdisciplinary Cyber Physical System of DST, Government of India, the project marks a major advancement in automated mapping.

In India, land surveys usually require weeks of manual work and generate inconsistent results. While drones are increasingly used for imaging, their outputs typically need extensive lab processing. Existing deep learning models also struggle to interpret crowded aerial scenes, resulting in unstable or inaccurate maps.

BHU-MANACHITRA overcomes these challenges by performing onboard AI-based processing to identify farmlands, forests, vegetation, urban zones, roads, and buildings instantly. With a lightweight AI model using only 2.48 million parameters, the system runs efficiently on low-power drone processors, eliminating the need for heavy hardware.

This technology has significant impact potential across multiple sectors. Government agencies can use it for land record modernization, urban planning, and smart city initiatives. Agriculture departments can evaluate crop conditions, soil health, and irrigation requirements in real time. During disasters such as floods or landslides, authorities can obtain immediate terrain information. Environmental bodies can use it to monitor deforestation, encroachments, and biodiversity changes.

According to Prof. Sambit Bakshi, the lightweight architecture makes the model ideal for drones that must remain compact to ensure long flight durations. The innovation aligns with the goals of Digital India and Atmanirbhar Bharat by supporting efficient governance, faster decision-making, and enhanced disaster management capabilities.

Click Here for More Institutional Activities