A Real-Time Affordable Predictive Trajectory Tracking Controller for Differential Drive Robots via Imitation Learning
IEEE International Conference on Robotics and Automation (ICRA) 2026
About Me
I am a PhD researcher in control systems and mobile robotics, with a strong focus on combining model-based control with data-driven learning methods for navigation and control. My work bridges classical nonlinear control, optimal control, and modern machine learning to design systems that move through the world with greater accuracy, stability, and adaptability.
My research spans learning-based trajectory tracking, set-theoretic safety methods, and predictive control for multi-agent systems. I develop controllers capable of real-time execution on affordable robotic hardware, while maintaining formal guarantees on constraint satisfaction and collision avoidance.
I work extensively with ground robots and multi-vehicle platforms, investigating how robot models and action models can complement or enhance robust controllers. Across all my work, I aim to bridge theoretical safety guarantees with practical, on-hardware implementation. Ultimately, my goal is to develop control and learning frameworks that make autonomous robotic systems both smarter and more trustworthy, enabling them to navigate, coordinate, and interact safely in complex, real-world environments.
Research Interests
Mobile & Aerial Robotics
Optimal Control Methods
Imitation Learning
Education
PhD in Information Systems Engineering
Sept 2024 - PresentConcordia University
Montreal, Canada
- Supervisor: Prof. Walter Lucia
- Group: Predictive Cyber-Physical System Security Group (PreCySe)
- Courses: Multi-agent Systems, Nonlinear Systems
MASc in Quality Systems Engineering
Sept 2022 - Aug 2024Concordia University
Montreal, Canada
- Supervisor: Prof. Walter Lucia
- Group: Predictive Cyber-Physical System Security Group (PreCySe)
- Courses: Linear System, TQM, Advanced Statistics, Cyber-physical Systems
B.Tech in Mechanical Engineering
Sept 2018 - Aug 2022Vellore Institute of Technology
Chennai, India
- Supervisor: Prof. Arockia Selvakumar
- Courses: Control Systems, Mechatronics, Mechanics, Thermodynamics, System Design
Publications
A Set-Theoretic Control Strategy for a Platoon of Constrained Differential-Drive Robots with Inter-Vehicle Collision Avoidance
IEEE Transactions on Automation Science and Engineering, 2025
Guaranteed Set-Theoretic Collision Avoidance in Multi-Agent Mobile Robotic Systems
Manuscript under preparation
Collision-Free Platooning of Mobile Robots Through a Set-Theoretic Predictive Control Approach
2024 American Control Conference (ACC), pp. 608-613, July 2024
ME-CapsNet: A Multi-Enhanced Capsule Networks with Routing Mechanism
2022 IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT), pp. 1-6, July 2022
Optimization of Quadcopter Frame Using Generative Design and Comparison with DJI F450 Drone Frame
International Conference of Robotics, Intelligent Automation and Control Technologies, Chennai, India, 2021
Robust Set Theoretic MPC-based Guidance for Guaranteed Collision Avoidance in Multi-Agent Robotic System
Research Poster, Bi-Annual Meeting on Systems and Control Theory (MSCT), Waterloo, Canada, 2023
Event-based Dynamic Obstacle Avoidance for Outdoor Environments
Research Poster, International Conference on Intelligent Robots and Systems (IROS), Prague, Czech Republic, 2021
Experience
Research Intern
Nov 2021 - Apr 2022ARTPARK, Artificial Intelligence and Robotics Laboratory, IISc
Bangalore, India
- Worked on event vision based end-to-end agile navigation of a quadrotor in a cluttered environment
- Developed and deployed a module for event-to-depth prediction using a spiking neural network for efficient monocular depth prediction for dynamic obstacle avoidance intended to be used on neuromorphic chip (Intel Loihi)
Research Intern
Jul 2021 - Nov 2021Robotics Innovations Lab, IISc
Bangalore, India
- Used neuromorphic/event cameras to perform dynamic obstacle avoidance in a constrained outdoor space using an autonomous ground vehicle and aerial platforms
- Conceptualized and developed human-UAV collaboration solutions to perform autonomous collaborative asset inspection in an industrial warehouse/assembly line setup
Summer Research Intern
May 2021 - Jul 2021Arizona State University
Phoenix, USA
- Digitized construction spaces using laser scanning and photogrammetry via sensor fusion
- Deep learning algorithms were used for automated analysis
- The digital representations made was processed to provide insights to builders and stewards
Autonomous System Developer - Intern
Oct 2020 - Apr 2021Aero2Astro
Chennai, India
- Developed ROS-based autonomous navigation firmware using Visual Inertial SLAM capable of working in indoor, outdoor, and GPS-denied environments
- Experiments were conceptualized and conducted for monocular vision SLAM using ORB3 SLAM specifically designed for aerial vehicles
Project Research Intern
Apr 2020 - Jun 2020Yuan-Ze University
Taoyuan City, Taiwan
- Deployed multiple ROS packages for KUKA-based manipulators and simulated various ROS Industrial packages using MoveIt
- Implemented forward and inverse kinematics for a 6 DOF custom-designed manipulator in ROS Industrial
Research Projects
Naval Research - Canadian Armed Forces
Setpoint Attack Detection on a Networked Control System
Designed a command governor cyber-attack detection solution to detect attacks on setpoints/references shared through a network which was tested on a high-fidelity simulation of a fleet of naval vessels.
Naval Research - Canadian Armed Forces
GAN-based Optimal Attack Detection and Generation on GNSS Measurement
Architected and implemented an LSTM-GAN network to detect attacks on GNSS through a discriminator network and generate optimal attack vectors through a generator network that cannot be detected through classical statistical-based cyber-attack detection mechanisms.
Curriculum Vitae
Download my complete curriculum vitae for detailed information about my education, research experience, publications, skills, and achievements.