A5089188394- Surgical Simulation and Training
- Advanced Image Processing Techniques
- Reinforcement Learning in Robotics
- Soft Robotics and Applications
- Generative Adversarial Networks and Image Synthesis
- Scoliosis diagnosis and treatment
- Anatomy and Medical Technology
- Assembly Line Balancing Optimization
- Augmented Reality Applications
- 3D Shape Modeling and Analysis
- Robot Manipulation and Learning
- Digital Games and Media
- Manufacturing Process and Optimization
- Computer Graphics and Visualization Techniques
- Engineering Technology and Methodologies
- Advanced Manufacturing and Logistics Optimization
- Innovations in Concrete and Construction Materials
- Digital Media Forensic Detection
University of Toronto
2022-2024
SickKids Foundation
2022
University of Ottawa
2019-2020
We introduce MuJoCo Playground, a fully open-source framework for robot learning built with MJX, the express goal of streamlining simulation, training, and sim-to-real transfer onto robots. With simple "pip install playground", researchers can train policies in minutes on single GPU. Playground supports diverse robotic platforms, including quadrupeds, humanoids, dexterous hands, arms, enabling zero-shot from both state pixel inputs. This is achieved through an integrated stack comprising...
Research in surgical robotics and automation has made remarkable advancements recent years thanks to new methods computer vision, control, deep learning. Autonomous end-effector manipulation is a challenging task robotics, cutting with scissor tools largely unexplored. A concurrent work explored path trajectory generation for deformable materials using the da Vinci Kit (dVRK) [1]. However, an efficient realistic simulation necessary such as reinforcement learning (RL) or learned planning....
Autonomous surgical robotics is a growing area of research, with advances being made in the areas vision and control. Central to this research need for simulations facilitate data collection simulate learning environments Reinforcement Learning (RL) agents. Recent simulators have facilitated RL policy generation, but lack robust sim2real pipeline proven vision-based that can use any type camera including da Vinci Surgical System (dVSS) Endoscope. To solve this, we build ROS-based...
Simulating medical images such as X-rays is of key interest to reduce radiation in non-diagnostic visualization scenarios. Past state the art methods utilize ray tracing, which reliant on 3D models. To our knowledge, no approach exists for cases where point clouds from depth cameras and other sensors are only input modality. We propose a method estimating an X-ray image generic cloud using conditional generative adversarial network (CGAN). train CGAN pix2pix translate into dataset created...
Surgical tasks such as tissue retraction, exposure, and needle suturing remain challenging in autonomous surgical robotics. One challenge these is nonprehensile manipulation pushing tissue, pressing cloth, threading. In this work, we isolate the problem of by implementing a vision-based reinforcement learning agent for rolling block, task that has complex dynamics interactions, small scale objects, narrow field view. We train agents simulation with reward formulation encourages efficient...