We propose a micro-data (< 10 trials) sensorimotor learning and adaptation (SEED) model for human-like arm inverse dynamics control. The SEED model consists of a feedforward Gaussian motor primitive (GATE) neural network and an adaptive feedback impedance (AIM) mechanism. Sensorimotor weights over trials are learned in the GATE network, while the AIM mechanism is used to online tune impedance gains in a trial. The model was validated by periodic and non-periodic tracking tasks on a two-joint robot arm. As a result, the proposed model enables the arm to stably learn the tasks within 10 trials, compared to thousands of trials required by state-of-art deep learning. This model facilitates the exploration of unknown arm dynamics, in which the elbow joint requires much less active control compared to the shoulder. This control goes below 3% of the overall effort. This finding complies with a proximal-distal control gradient in human arm control. Taken together, the proposed SEED model paves a way for implementing data-efficient sensorimotor learning and adaptation of human-like arm movement.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1016/j.neunet.2021.06.029 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Bridging animal models and humans: neuroimaging as intermediate phenotypes linking genetic or stress factors to anhedonia.
BMC Med
January 2025
Early Intervention Unit, Department of Psychiatry, Affiliated Nanjing Brain Hospital, Nanjing Medical University, 264 Guangzhou Street, Nanjing, China.
Background: Intermediate phenotypes, such as characteristic neuroimaging patterns, offer unique insights into the genetic and stress-related underpinnings of neuropsychiatric disorders like depression. This study aimed to identify neuroimaging intermediate phenotypes associated with depression, bridging etiological factors to behavioral manifestations and connecting insights from animal models to diverse clinical populations.
Methods: We analyzed datasets from both rodents and humans.
Humans excel at applying learned behavior to unlearned situations. A crucial component of this generalization behavior is our ability to compose/decompose a whole into reusable parts, an attribute known as compositionality. One of the fundamental questions in robotics concerns this characteristic: How can linguistic compositionality be developed concomitantly with sensorimotor skills through associative learning, particularly when individuals only learn partial linguistic compositions and their corresponding sensorimotor patterns? To address this question, we propose a brain-inspired neural network model that integrates vision, proprioception, and language into a framework of predictive coding and active inference on the basis of the free-energy principle.
View Article and Find Full Text PDFCommentary on "Prognostic value of longitudinal HbA1c variability in predicting the development of diabetic sensorimotor polyneuropathy among patients with type 2 diabetes mellitus: A prospective cohort observational study".
J Diabetes Investig
January 2025
Infectious Diseases and Tropical Medicine Research Center, Health Research Institute, Babol University of Medical Sciences, Babol, Iran.
Motor decision-making under uncertainty and time pressure.
J Neurophysiol
February 2025
Centre for Sensorimotor Performance, School of Human Movement and Nutrition SciencesThe University of QueenslandBrisbaneQueenslandAustralia.
Purposeful movement often requires selection of a particular action from a range of alternatives, but how does the brain represent potential actions so that they can be compared for selection, and how are motor commands generated if movement is initiated before the final goal is identified? According to one hypothesis, the brain averages partially prepared motor plans to generate movement when there is goal uncertainty. This is consistent with the idea that motor decision-making unfolds through competition between internal representations of alternative actions. An alternative hypothesis holds that only one movement, which is optimized for task performance, is prepared for execution at any time.
View Article and Find Full Text PDFAfter a Hand Was Lent: Sporadically Experiencing Multisensory Interference During the Rubber Hand Illusion Does Not Shield Against Disembodiment.
J Cogn
January 2025
General Psychology, Trier University, Germany.
Observations from multisensory body illusions indicate that the body representation can be adapted to changing task demands, e.g., it can be expanded to integrate external objects based on current sensorimotor experience (embodiment).
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!