Constrained geodesic trajectory generation on learnt skill manifolds

Ioannis Havoutis, Subramanian Ramamoorthy

Research output: Chapter in Book/Report/Conference proceedingConference contribution


This paper addresses the problem of compactly encoding a continuous family of trajectories corresponding to a robotic skill, and using this representation for the purpose of constrained trajectory generation in an environment with many (possibly dynamic) obstacles. With a skill manifold that is learnt from data, we show that constraints can be naturally handled within an iterative process of minimizing the total geodesic path length and curvature over the manifold. We demonstrate the utility of this process with two examples. Firstly, a three-link arm whose joint space and corresponding skill manifold can be explicitly visualized. Then, we demonstrate how this procedure can be used to generate constrained walking motions in a humanoid robot.
Original languageEnglish
Title of host publicationIEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010)
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Number of pages6
ISBN (Print)978-1-4244-6675-7
Publication statusPublished - Oct 2010


  • constrained geodesic trajectory generation
  • constrained walking motion
  • geodesic path length
  • humanoid robot
  • manifold curvature
  • obstacle avoidance
  • robotic skill
  • skill manifold
  • collision avoidance
  • differential geometry
  • humanoid robots
  • mobile robots


Dive into the research topics of 'Constrained geodesic trajectory generation on learnt skill manifolds'. Together they form a unique fingerprint.

Cite this