Reda Kara

About

Reda Kara’s research lies at the intersection of robotics, neural networks, and visual servoing, with a focus on enabling robots to learn complex visuo-motor coordination. His most cited work introduces a novel architecture using multiple self-organizing maps (SOMs) to model the correlations between a robot’s motor commands and its visual feedback, allowing a pair of active cameras to autonomously track a manipulator in 3D space. This bi-directional neural modularity approach, published in 2003, has garnered 15 citations and remains a foundational reference for adaptive robotic vision systems. Kara further advanced the field by integrating hierarchical CMAC (Cerebellar Model Articulation Controller) neurocontrollers for efficient, real-time target tracking, demonstrating that neural networks can replace traditional control algorithms in dynamic environments. His 2002 and 2018 papers on CMAC-based visual servoing, though less cited, underscore a sustained commitment to bridging machine learning and robotics. By showing how self-organizing maps and cerebellar-inspired models can learn precise motor commands from visual input, Kara has contributed to making robots more autonomous and adaptable—key achievements for students and researchers exploring neural control in complex, real-world tasks.

Research Focus

Key Achievements

Top Papers

1. 1
   Multiple self-organizing maps to facilitate the learning of visuo-motor correlations

   15 citations · 2003
2. 2
   Robot vision tracking with a hierarchical CMAC controller

   4 citations · 2002
3. 3
   The CMAC Neurocontroller for efficient learning in visual servoing

   2 citations · 2018

Key Collaborators

Contact & Links