Mohammadreza Sharif

About

Mohammadreza Sharif is a robotics researcher focused on advancing prosthetic hand control through multimodal sensing and reinforcement learning. His work centers on integrating electromyography (EMG) signals with vision and other sensor modalities to overcome the robustness issues that plague traditional EMG-only control systems. His most cited paper, "Multimodal fusion of EMG and vision for human grasp intent inference in prosthetic hand control" (2024, 44 citations), proposes a fusion framework that significantly improves inference accuracy for transradial amputees by mitigating motion artifacts and muscle fatigue. Sharif has also pioneered end-to-end deep reinforcement learning approaches for human-in-the-loop robot grasping, as demonstrated in his 2021 paper (4 citations), which trains grasping policies directly from multimodal inputs. His earlier work on particle filters versus hidden Markov models for grasp selection (2019, 3 citations) laid groundwork for probabilistic intent inference. Beyond prosthetics, he contributed to legged robotics with the design of a quadruped shuffling mobile robot (2021, 6 citations). Sharif’s research directly addresses the critical challenge of translating lab-grade EMG control into real-world reliability, making him a notable figure in assistive robotics and human-robot interaction.

Research Focus

Key Achievements

Top Papers

1. 1
   Multimodal fusion of EMG and vision for human grasp intent inference in prosthetic hand control

   44 citations · 2024
2. 2
   Towards End-to-End Control of a Robot Prosthetic Hand via Reinforcement Learning

   8 citations · 2020
3. 3
   Design and development of a quadruped shuffling mobile robot

   6 citations · 2021
4. 4
   End-to-end grasping policies for human-in-the-loop robots via deep reinforcement learning

   4 citations · 2021
5. 5
   Particle Filters vs Hidden Markov Models for Prosthetic Robot Hand Grasp Selection

   3 citations · 2019

Key Collaborators

Contact & Links