Sensorimotor learning of dexterous manipulation

Qiushi Fu, Marco Santello

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Human sensorimotor learning has been extensively studied. However, most of this work has been focused on the adaptation of reaching movements against force fields or in response to visual perturbations of movement trajectories. As a result, little is known about the mechanisms underlying the learning of dexterous manipulation. To address this gap, we have designed several tasks and sensorized devices to quantify the processes underlying acquisition and retention of dexterous manipulation, as well as mechanisms that may favor or interfere with the generalization of learned manipulation tasks to other task contexts. As humans learned to perform manipulation tasks of the same or different objects across different contexts, we found evidence for multiple sensorimotor mechanisms characterized by different time scales and extents to which they can interfere with each other. We conclude this chapter by presenting potential applications of these biological findings to robotic dexterous manipulation.

Original languageEnglish (US)
Title of host publicationHuman Inspired Dexterity in Robotic Manipulation
PublisherElsevier
Pages27-52
Number of pages26
ISBN (Electronic)9780128133965
ISBN (Print)9780128133859
DOIs
StatePublished - Jun 29 2018

Keywords

  • Adaptation
  • Generalization
  • Internal representations
  • Memory
  • Retention

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'Sensorimotor learning of dexterous manipulation'. Together they form a unique fingerprint.

  • Cite this

    Fu, Q., & Santello, M. (2018). Sensorimotor learning of dexterous manipulation. In Human Inspired Dexterity in Robotic Manipulation (pp. 27-52). Elsevier. https://doi.org/10.1016/B978-0-12-813385-9.00003-0