Hello! I’m Ishaan Mehta, a final-year Ph.D. candidate at Toronto Metropolitan University (TMU), working with the Robotics and Computer Vision Lab (RCVL).

My research spans robot learning, planning, perception, and multi-robot systems. I have 8+ years of experience developing autonomous robotic systems, including drones, self-driving cars, mobile robots, and robotic manipulators. My work covers the robotics stack from SLAM, 3D perception, and spatial mapping to manipulation planning, reinforcement learning, and multi-robot coordination. I am particularly interested in how deep learning can be combined with well-established classical and physics-informed methods to build robotic systems that are both adaptable and reliable.

My broader research interests include:

Reinforcement learning for planning and decision-making
Manipulation, task sequencing, and motion planning
Multi-robot task allocation and distributed planning
SLAM, 3D Gaussian Splatting, and neural scene representations
Embodied and agentic AI for real-world robotic systems

I have contributed to 11+ publications (including work under review), collaborated with 20+ researchers, and mentored students across robotics and AI projects. A recurring theme in my work is translating research ideas into practical systems through prototyping, rigorous evaluation, and real-robot validation.

Most recently, I worked as a Machine Learning Associate at the Vector Institute, where I collaborated with Geomate on an agentic simulation pipeline for generating diverse, safety-critical autonomous-driving scenarios. Previously, I helped develop UV-disinfection mobile robots at RCVL for real-world deployment during the COVID-19 pandemic.

I earned my Master of Engineering in Autonomous Robotics from the University of Toronto Institute for Aerospace Studies (UTIAS) and my Bachelor of Engineering in Mechatronics from Thapar University.

You can find more details about my research, publications, and projects below. Download Resume →

Selected Projects

RTSP-Net: Reinforcement Learning-Based Robotic Task Sequencing for Robotic Manipulators in Obstacle-Rich Environments
Submitted to IEEE Robotics and Automation Letters
Paper. Project.

SplatUnity: A Collaborative Online SLAM and Learning Framework with 3D Gaussian Splatting and Inter-Agent Pose Alignment
Submitted to IEEE Robotics and Automation Letters
Paper. Project.

Morphological Detection and Classification of Microplastics and Nanoplastics Emerged from Consumer Products by Deep Learning
Submitted to Science of the Total Environment
Paper. Project.

M^3RS: Multi-robot, Multi-objective, and Multi-mode Routing and Scheduling
Submitted to Journal of Field Robotics
Paper. Project.

Pareto Frontier Approximation Network (PA-Net) to Solve Bi-objective TSP
IEEE International Conference on Automation Science and Engineering 2022
Mexico City, Mexico
Paper. Project.

Far-UVC disinfection with robotic mobile manipulator
International Symposium on Medical Robotics 2022
Atlanta, USA
Paper. Project.

Volunteer Work

I’m an active volunteer with the Art of Living Foundation, dedicated to promoting breathwork and meditation for stress relief and inner peace. Amidst the COVID-19 pandemic, I initiated daily virtual sessions offering free breathwork and meditation to help individuals combat stress and exhaustion. Further, I actively engage with SKY Campus Canada to promote well being in student community through breathwork and meditation. Here is some research about the impact of breathwork on stress and wellness.

Contact

Feel free to reach out to me for collaborations, guidance for aspiring researchers, or to learn more about breathwork.