Randal C. Nelson

About

Randal C. Nelson is a robotics and computer vision researcher whose work has made significant contributions to the fields of visual navigation, robot manipulation, and adaptive motor control. His early and most influential work, "Visual homing using an associative memory" (1991, 58 citations), demonstrated how robots could navigate using biologically inspired memory structures, establishing him as a pioneer in vision-based autonomous systems. Much of Nelson's subsequent research tackled one of robotics' most persistent challenges: enabling robots to operate effectively in unstructured environments. Through a series of influential papers on dextrous manipulation, he developed methods allowing robotic hands to perform complex tasks without requiring prior object models or precise camera calibration. His work on adaptive visual feedback control (1994) and the acquisition of visual-motor models (1996) pushed the boundaries of uncalibrated hand-eye coordination, while his application of evolutionary strategies to learn manipulation primitives offered a compelling approach to autonomous skill acquisition. Nelson's concept of "virtual tools" provided an elegant framework for harnessing high-degree-of-freedom robotic systems without prohibitive computational overhead. With cumulative citations across a focused body of work, Nelson's research has meaningfully shaped how roboticists think about flexible, learning-capable manipulation systems that bridge biological inspiration and practical engineering.

Research Focus

Key Achievements

Top Papers

1. 1
   Visual homing using an associative memory

   58 citations · 1991
2. 2
   Visual space task specification, planning and control

   34 citations · 2002
3. 3
   Adaptive Differential Visual Feedback for Uncalibrated Hand-Eye Coordination and Motor Control

   24 citations · 1994
4. 4
   Acquiring visual-motor models for precision manipulation with robot hands

   23 citations · 1996
5. 5
   Learning Dextrous Manipulation Skills for Multifingered Robot Hands Using the Evolution Strategy

   15 citations · 1998
6. 6
   Learning Dextrous Manipulation Skills for Multifingered Robot Hands Using the Evolution Strategy

   12 citations · 1998
7. 7
   Experiments on dextrous manipulation without prior object models

   12 citations · 2002
8. 8
   The virtual tool approach to dextrous telemanipulation

   10 citations · 2002
9. 9
   Morphing Hands and Virtual Tools (or What Good is an Extra Degree of Freedom

   8 citations · 1994
10. 10
    Learning dextrous manipulation skills using the evolution strategy

    7 citations · 2002

Key Collaborators

Contact & Links