Companion Website

Nature Machine Intelligence - March 2019

Article (open-access):

Arxiv Preprint

A tendon-driven robotic limb learns movements autonomously from sparse experience, by a short period of ‘motor babbling’ (that is, repeated exploratory movements), followed by a phase of reinforcement learning. In this video, the limb is learning to make cyclic movements to propel the treadmill. The approach is a step towards designing robots with the versatility and robustness of vertebrates, which can adapt quickly to everyday environments.


Marjaninejad, Ali, et al. “Autonomous functional movements in a tendon-driven limb via limited experience.” Nature Machine Intelligence 1.3 (2019): 144.

Citation (.ris)
Citation (BibTex)


Research reported in this publication was supported in part by the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health under award numbers R01 AR-050520, R01 AR-052345, the Department of Defense CDMRP Grant MR150091, and Award W911NF1820264 from the DARPA’s Lifelong Learning Machines (L2M) program.

We’d be happy to help you apply these methods to your work. Send us a message: [email protected]

Feature in Wired Magazine, 2018

News Coverage

Nature: Editorial
Defense Advanced Research Projects Agency (DARPA)
TechXplore: Robotics
Neuroscience News
USC Viterbi School of Engineering
EurekAlert-AAAS Photo
EurekAlert-AAAS Article
TuniseSoir News
NeuroLogica Blog

Supplemental site prepared by: Brian A. Cohn and Ali Marjaninejad