Stavros Tsogkas
Centre for Social Innovation, University of Toronto, Supélec
Papers
3
Total Citations
95
H-Index
3
About
Stavros Tsogkas is a leading researcher in robotic manipulation and assistive technologies, whose work bridges the gap between simulation and real-world dexterity. His primary research areas include differentiable physics for robotics, multi-fingered grasp synthesis, and human-aware perception for mobility assistance. Tsogkas’s major contribution is the development of **Grasp’D**, a groundbreaking framework for differentiable contact-rich grasp synthesis that enables gradient-based optimization of multi-fingered grasps. This work, which has garnered 53 citations, overcomes the limitations of traditional data-driven methods by making contact dynamics amenable to optimization, thus producing high-quality grasps without relying on hard-to-obtain human data or simplifying assumptions. He further advanced this line of research with **Fast-Grasp’D** (24 citations), accelerating dexterous grasp generation through differentiable simulation. Earlier in his career, Tsogkas made notable contributions to assistive robotics with his work on accurate human-limb segmentation in RGB-D images (18 citations), a key component for intelligent mobility assistance robots designed to improve the independence and quality of life for elderly individuals. His work stands out for its elegant fusion of theory and application, offering practical solutions to some of the most challenging problems in robotic dexterity and human-robot interaction.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3