Jun Nakanishi

About

Jun Nakanishi is a pioneering robotics and computational neuroscience researcher whose work has fundamentally shaped how robots learn and reproduce complex motor behaviors. His research sits at the intersection of nonlinear dynamical systems, imitation learning, and humanoid robotics, with particular emphasis on biologically inspired approaches to movement generation and control. Nakanishi's most transformative contribution is his foundational work on Dynamical Movement Primitives (DMPs), a framework for representing and learning motor behaviors as attractor dynamics — a paper that has garnered over 1,574 citations and become a cornerstone reference in robot learning. Building on this foundation, his earlier work on movement imitation in humanoid robots (842 citations) established practical methods for on-line trajectory modification and learning from demonstration that remain widely used today. His research extends into biped locomotion, exploring central pattern generator (CPG)-based walking controllers and policy gradient reinforcement learning methods applied to physical humanoid hardware. His brachiating robot work demonstrates an early and elegant use of target dynamical systems for complex whole-body motion. Across more than a decade of influential publications, Nakanishi has consistently bridged biological motor control principles and practical robotics engineering, producing work that has profoundly influenced the fields of robot learning, motion planning, and imitation learning.

Research Focus

Key Achievements

Top Papers

1. 1
   Dynamical Movement Primitives: Learning Attractor Models for Motor Behaviors

   1,574 citations · 2012
2. 2
   Movement imitation with nonlinear dynamical systems in humanoid robots

   842 citations · 2003
3. 3
   Learning from demonstration and adaptation of biped locomotion

   400 citations · 2004
4. 4
   Learning Movement Primitives

   346 citations · 2005
5. 5
   Learning CPG-based Biped Locomotion with a Policy Gradient Method: Application to a Humanoid Robot

   211 citations · 2008
6. 6
   A brachiating robot controller

   211 citations · 2000
7. 7
   Trajectory formation for imitation with nonlinear dynamical systems

   175 citations · 2002
8. 8
   A Biologically Inspired Biped Locomotion Strategy for Humanoid Robots: Modulation of Sinusoidal Patterns by a Coupled Oscillator Model

   119 citations · 2008
9. 9
   Learning CPG-based biped locomotion with a policy gradient method

   98 citations · 2006
10. 10
    Comparative experiments on task space control with redundancy resolution

    81 citations · 2005

Key Collaborators

Contact & Links