ProDapt: Proprioceptive Adaptation using Long-term Memory Diffusion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bejarano, Federico Pizarro, Jones, Bryson, Moreno, Daniel Pastor, Bowkett, Joseph, Backes, Paul G., Schoellig, Angela P.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915177661726720
author Bejarano, Federico Pizarro
Jones, Bryson
Moreno, Daniel Pastor
Bowkett, Joseph
Backes, Paul G.
Schoellig, Angela P.
author_facet Bejarano, Federico Pizarro
Jones, Bryson
Moreno, Daniel Pastor
Bowkett, Joseph
Backes, Paul G.
Schoellig, Angela P.
contents Diffusion models have revolutionized imitation learning, allowing robots to replicate complex behaviours. However, diffusion often relies on cameras and other exteroceptive sensors to observe the environment and lacks long-term memory. In space, military, and underwater applications, robots must be highly robust to failures in exteroceptive sensors, operating using only proprioceptive information. In this paper, we propose ProDapt, a method of incorporating long-term memory of previous contacts between the robot and the environment in the diffusion process, allowing it to complete tasks using only proprioceptive data. This is achieved by identifying "keypoints", essential past observations maintained as inputs to the policy. We test our approach using a UR10e robotic arm in both simulation and real experiments and demonstrate the necessity of this long-term memory for task completion.
format Preprint
id arxiv_https___arxiv_org_abs_2503_00193
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ProDapt: Proprioceptive Adaptation using Long-term Memory Diffusion
Bejarano, Federico Pizarro
Jones, Bryson
Moreno, Daniel Pastor
Bowkett, Joseph
Backes, Paul G.
Schoellig, Angela P.
Robotics
Machine Learning
Systems and Control
Diffusion models have revolutionized imitation learning, allowing robots to replicate complex behaviours. However, diffusion often relies on cameras and other exteroceptive sensors to observe the environment and lacks long-term memory. In space, military, and underwater applications, robots must be highly robust to failures in exteroceptive sensors, operating using only proprioceptive information. In this paper, we propose ProDapt, a method of incorporating long-term memory of previous contacts between the robot and the environment in the diffusion process, allowing it to complete tasks using only proprioceptive data. This is achieved by identifying "keypoints", essential past observations maintained as inputs to the policy. We test our approach using a UR10e robotic arm in both simulation and real experiments and demonstrate the necessity of this long-term memory for task completion.
title ProDapt: Proprioceptive Adaptation using Long-term Memory Diffusion
topic Robotics
Machine Learning
Systems and Control
url https://arxiv.org/abs/2503.00193