Research
Intelligent robotic systems, control theory, Reinforcement Learning and machine learning.
Stats updated manually. View live on Google Scholar →
Research Statement
My research focuses on intelligent robotic systems that operate reliably in real-world environments. My primary MSc research is on electric wheelchair robots, addressing control challenges that arise from real-world mechanical constraints such as non-configurable caster wheel drift, using reinforcement learning with residual networks and sim-to-real transfer. I also worked on shape-aware path planning algorithms for the Smorphi reconfigurable modular robot, and on augmented reality systems for immersive indoor robot navigation using computer vision and depth estimation. My research broadly spans mobile robotics, reinforcement learning, motion control, and visual perception. I am currently targeting submission to ICRA 2027.
Research Interests
Reinforcement Learning for Robot Control
Training RL policies with residual networks for mobile robot control, learning to compensate for mechanical imperfections such as caster wheel drift, with sim-to-real transfer on physical hardware.
Mobile & Reconfigurable Robotics
Control and path planning for mobile robotic platforms, including wheelchair robots with non-ideal caster wheel mechanics and shape-aware navigation for the Smorphi reconfigurable modular robot. Includes sim-to-real transfer on physical hardware.
Motion Control & Dynamic Modeling
Kinematic and dynamic modeling of wheeled robotic systems, trajectory planning, and control architectures for reliable autonomous navigation in unstructured environments.
Computer Vision & Augmented Reality
Visual perception for robotic systems including depth estimation, semantic segmentation, 3D mapping, and AR-based interfaces for immersive indoor robot monitoring and human-robot interaction.
Enabling Immersive Indoor Navigation and Control Through Augmented Reality With Computer Vision
Integrates computer vision, AR, and cloud-based communication to create a real-time 3D map of a robot's surroundings for an immersive indoor robot control experience, with 1 cm accuracy depth estimation using Microsoft Kinect V2 and semantic segmentation.
CasteriX: A Wheel Configurable Caster Wheel-Based Prototype Design for Electric Wheelchair Motion Dynamics Research — Best Poster Award
A wheel-configurable modular research platform for electric wheelchair caster wheel motion dynamics analysis, supporting 212 wheel configurations across three major wheelchair types, recognized with the 🏆 Best Poster Award at ICIPRoB 2026.
Adaptive Navigation of a Transformer Robot in Warehouse Environments
A adaptive navigation framework for transformer robots in warehouse environments, integrating overhead camera perception, binary segmentation, and a shape-aware A* algorithm to determine optimal robot configurations through narrow passages.
Writing conference paper for the International Conference on Robotics & Automation(ICRA) 2027
Currently writing a paper for submission to the International Conference on Robotics and Automation (ICRA) 2027, focusing on advanced robotics research. Full citation will be available upon IEEE publication.
Research Projects
Ongoing and completed research projects
AR Indoor Navigation & Robot Control
AR Indoor Navigation & Robot Control
Integrates computer vision, Augmented Reality (AR), and AWS cloud services to create a real-time 3D map of a robot's surroundings, enabling immersive indoor robot monitoring and control with 1 cm depth accuracy using Microsoft Kinect V2.
Academic Services
Collaborators & Labs
References
Available upon request
Prof. A.G.B.P. Jayasekara
Professor & Head, Dept. of Electrical Engineering
University of Moratuwa, Sri Lanka
jayasekara@uom.lkProf. Chinthaka Premachandra
Professor, Dept. of Electronic Engineering
Shibaura Institute of Technology, Tokyo, Japan
premachandra@sic.shibaura-it.ac.jp