Papers
5
Total Citations
61
H-Index
5
About
Mohsen Imani is a leading researcher at the intersection of brain-inspired computing and robotics, pioneering the use of Hyperdimensional Computing (HDC) to create ultra-efficient, lightweight AI for real-world control systems. His work directly addresses the critical challenge of deploying intelligent decision-making on energy-constrained edge devices, where traditional deep neural networks are too resource-hungry. Imani’s major contributions include the development of novel HDC frameworks such as **ReactHD** for sensorimotor control of wheeled robots and **ScaleHD** for scalable cognition tasks, demonstrating that symbolic, brain-like models can achieve robust performance with minimal computational cost. He has also advanced robotic grasping perception through event-based HDC, enabling accurate object property inference from dynamic vision sensors. With his most cited work, "HDPG" (27 citations), applying HDC to continuous control tasks, and "DARL" (14 citations) pioneering HDC-powered reinforcement learning, Imani’s research has garnered significant attention for its potential to democratize intelligent robotics. His achievements showcase a compelling vision: replacing hand-crafted control and costly deep learning with self-learning, energy-efficient HDC systems that bring human-level adaptability to robots operating in the wild.