CV

Research Intern

Multi Agent Robotic Motion Lab, National University of Singapore

• Developed an adaptive, reinforcement learning based method for Informative Path Planning in an obstacle-free 2D environment subject to budget constraints.
• Our work was accepted at the 6th Annual Conference on Robot Learning

Research Intern

Lab for Photogrammetry and Robotics, University of Bonn

• Proposed a novel dynamic graph approach for adaptive informative motion planning to map targets-of-interest in 3D environment in presence of unknown obstacles.
• First author work published in IEEE Robotics and Automation Letters

Cadet Corporal

National Cadet Corps

The National Cadet Corps (NCC) at the Indian Institute of Technology Kharagpur (IITKGP) is the college-level youth wing of the Indian Armed Forces, offering students basic military training to develop character, discipline, and leadership qualities.

• Completed the Winter Annual Training Camp (ATC) at IIT Kharagpur - a mandatory, intensive 12-day military-style training camp held on campus during the winter vacation.
• Cadet Corporal (CPL) is a selected Non-Commissioned Officer (NCO) leadership rank awarded to second-year NCC cadets who have demonstrated superior discipline, military proficiency, and the capability to command a section of junior cadets.

Governor

Communiqué

Communiqué at IIT Kharagpur is the institute’s Official Soft Skills and Personality Development Society that bridges the gap between professional communication experts and the student body.

• The position of Governor represents the highest tier of student leadership and executive decision-making within the society.

B.Tech + M.Tech

Indian Institute of Technology, Kharagpur

Mechanical Engineering

• Programming and Data Structures
• Partial Differential Equations
• Systems and Control
• Mechatronics
• B.Tech Thesis on “Modeling and Energy Optimization of 3-Wheeled Omnidirectional Mobile Robot (TOMR)”
• M.Tech Thesis on “Voronoi Partitioning Based Multi-agent Path Planning”

PhD

Purdue University

Computer Science

• Imitation Learning
• Artificial Intelligence
• Thesis focusing on “Temporal Context for Robust Robot Planning - Integrating History Awareness and Future Prediction”

Herbold Scholarship

The Herbold Foundation

The Herbold Scholarship is a financial award established by the Herbold Foundation—founded by former Microsoft COO Bob Herbold—to support outstanding graduate students in engineering, computer science, and data science.

Ross Fellowship

Purdue University

Ross Fellowship is a highly competitive, prestigious recruitment award offered by Purdue University to recruit outstanding Ph.D.-track students.

International Internship Scholarship Award

IITKGP Foundation USA

The International Internship Award IITKGP is a prestigious scholarship program offered by IIT Kharagpur in association with the IITKGP Foundation of USA. It provides financial support to meritorious undergraduate and graduate students pursuing high-quality research and professional internships at top global institutions.

NTSE Scholar

NCERT, Government of India

The National Talent Search Examination (NTSE) is a prestigious, government-funded scholarship program in India aiming to identify and financially support exceptionally talented 10th-grade students pursuing higher education in basic sciences, social sciences, medicine, and engineering.

CoReLIN: Constraint-based Reasoning for Zero-shot Lifelong Interactive Navigation

European Conference on Computer Vision

This work introduces an LLM-driven, constraint-based planning framework that enables robots to rearrange clutter, actively explore, and complete long-horizon navigation tasks with zero-shot generalization.

Scalable Multi-Robot Informative Path Planning for Target Mapping via Deep Reinforcement Learning

IEEE Robotics and Automation Letters

We present a scalable deep reinforcement learning framework that enables teams of robots to efficiently map targets in unknown 3D environments while coordinating and avoiding collisions without retraining for different team sizes.

Deep Reinforcement Learning With Dynamic Graphs for Adaptive Informative Path Planning

IEEE Robotics and Automation Letters

We develop a deep reinforcement learning planner that uses dynamic local graphs to efficiently discover and map targets in unknown, obstacle-filled 3D environments under limited resource budgets.

CAtNIPP: Context-Aware Attention-based Network for Informative Path Planning

Conference on Robot Learning

This work introduces a self-attention-based reinforcement learning planner that balances exploration and exploitation to perform fast, adaptive informative path planning under limited resource budgets.