Sensorimotor Learning With Stability Guarantees via Autonomous Neural Dynamic Policies

State-of-the-art sensorimotor learning algorithms, either in the context of reinforcement learning or imitation learning, offer policies that can often produce unstable behaviors, damaging the robot and/or the environment. Moreover, it is very difficult to interpret the optimized controller and anal...

Full description

Saved in:
Bibliographic Details
Published in:IEEE robotics and automation letters Vol. 10; no. 2; pp. 1760 - 1767
Main Authors: Totsila, Dionis, Chatzilygeroudis, Konstantinos, Modugno, Valerio, Hadjivelichkov, Denis, Kanoulas, Dimitrios
Format: Journal Article
Language:English
Published: IEEE 01.02.2025
Subjects:
Aerospace electronics
Artificial neural networks
Asymptotic stability
Computer Science
Dynamical systems
Heuristic algorithms
Imitation learning
Learning from demonstration
machine learning for robot control
Robot learning
Robotics
Robots
sensorimotor learning
Stability analysis
Trajectory
Machine Learning for Robot Control
Sensorimotor Learning
Learning from Demonstration Sensorimotor Learning Machine Learning for Robot Control
Learning from Demonstration
ISSN:2377-3766, 2377-3766
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:State-of-the-art sensorimotor learning algorithms, either in the context of reinforcement learning or imitation learning, offer policies that can often produce unstable behaviors, damaging the robot and/or the environment. Moreover, it is very difficult to interpret the optimized controller and analyze its behavior and/or performance. Traditional robot learning, on the contrary, relies on dynamical system-based policies that can be analyzed for stability/safety. Such policies, however, are neither flexible nor generic and usually work only with proprioceptive sensor states. In this work, we bridge the gap between generic neural network policies and dynamical system-based policies, and we introduce Autonomous Neural Dynamic Policies (ANDPs) that: (a) are based on autonomous dynamical systems, (b) always produce asymptotically stable behaviors, and (c) are more flexible than traditional stable dynamical system-based policies. ANDPs are fully differentiable, flexible generic-policies that accept any observation input, while ensuring asymptotic stability. Through several experiments, we explore the flexibility and capacity of ANDPs in several imitation learning tasks including experiments with image observations. The results show that ANDPs combine the benefits of both neural network-based and dynamical system-based methods.
ISSN:2377-3766
2377-3766
DOI:10.1109/LRA.2024.3524878