R. Eckmiller
Papers
10
Total Citations
51
H-Index
4
About
R. Eckmiller is a pioneering researcher in neural robotics, whose work bridges the gap between biological motor control and artificial systems. His primary research areas include neural network architectures for robot control, inverse kinematics, and force/position control. Eckmiller’s major contributions center on developing adaptive neural structures that enable robots to learn complex tasks, such as the CATE (Connection Assignment and Topographical Encoding) framework, which allows rapid learning of inverse kinematics through a novel topographical encoding scheme. He also introduced Neural Force Control (NFC), a hybrid approach that significantly expands manipulator capabilities by integrating neural networks for inverse dynamics and kinematics. His work on quasi-local solutions for redundant robot arms and globally stable neural control with payload adaptation further underscores his impact. With papers accumulating up to 10 citations each, Eckmiller’s research has influenced both theoretical and applied robotics. Notably, his NFC concept for 6DOF industrial robots achieved a cycle time of just 2 milliseconds, demonstrating real-time feasibility. For students and researchers, Eckmiller’s work offers a compelling vision of how neural networks can endow robots with adaptive, human-like motor skills.
Research Focus
Key Achievements
Top Papers
- 1
- 2Quasi-local solution for inverse kinematics of a redundant robot arm10 citations · 1990
- 3
- 4
- 5Neural force control (NFC) for complex manipulator tasks4 citations · 1997
- 6
- 7
- 8A Recurrent Trajectory Storage Network with Parceling of the Workspace2 citations · 1993
- 9Improvement of robot control by neural computers2 citations · 1991
- 10Globally stable neural robot control capable of payload adaptation2 citations · 2005