Improving Pedestrian Prediction Models with Self-Supervised Continual Learning

Research output: Contribution to journalArticleScientificpeer-review

Abstract

Autonomous mobile robots require accurate human motion predictions to safely and efficiently navigate among pedestrians, whose behavior may adapt to environmental changes. This paper introduces a self-supervised continual learning framework to improve data-driven pedestrian prediction models online across various scenarios continuously. In particular, we exploit online streams of pedestrian data, commonly available from the robot's detection and tracking pipelines, to refine the prediction model and its performance in unseen scenarios. To avoid the forgetting of previously learned concepts, a problem known as catastrophic forgetting, our framework includes a regularization loss to penalize changes of model parameters that are important for previous scenarios and retrains on a set of previous examples to retain past knowledge. Experimental results on real and simulation data show that our approach can improve prediction performance in unseen scenarios while retaining knowledge from seen scenarios when compared to naively training the prediction model online.

Original languageEnglish
Pages (from-to)4781-4788
JournalIEEE Robotics and Automation Letters
Volume7
Issue number2
DOIs
Publication statusPublished - 2022

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

  • Adaptation models
  • Continual Learning
  • Data models
  • Human-Aware Motion Planning
  • Predictive models
  • Robots
  • Service Robotics
  • Task analysis
  • Training
  • Trajectory
  • Trajectory Prediction

Fingerprint

Dive into the research topics of 'Improving Pedestrian Prediction Models with Self-Supervised Continual Learning'. Together they form a unique fingerprint.

Cite this