usc-bbdl / usc-bbdl.github.io

Website for the Brain Body Dynamics Lab
http://valerolab.org
1 stars 4 forks source link

Publication info - Ali 1 #84

Open marjanin opened 4 years ago

marjanin commented 4 years ago

Delete all nonrelevant information from this template when submitting your issue request:

Publication


Title: Simple Kinematic Feedback Enhances Autonomous Learning in Bio-Inspired Tendon-Driven Systems
Link: https://arxiv.org/pdf/1907.04539
Abstract: Error feedback is known to improve performance by correcting control signals in response to perturbations. Here we show how adding simple error feedback can also accelerate and robustify autonomous learning in robots. We implemented two versions of the General-to-Particular (G2P) autonomous learning algorithm to produce multiple movement tasks using a tendon-driven leg with two joints and three tendons: one with and one without kinematic feedback. As expected, feedback improved performance in simulation and hardware. However, we see these improvements even in the presence of sensory delays of up to 100 ms and when experiencing substantial contact collisions. Importantly, feedback accelerates learning by enhancing G2P's continual refinement of the initial inverse map because every experience counts. This allows the system to perform well even after only 60 seconds of initial motor babbling.
Authors: Ali Marjaninejad, Darío Urbina-Meléndez, Francisco J Valero-Cuevas
BibTex: not ready yet
Year: 2020
conference: 42th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).
``
 ## NOTE:
The format of the Author section is very important to the design of the site. If there’s no supplemental material, just leave it blank.
bc commented 4 years ago

throw up a PDF if you can, or of the abstract