Robotic Exploration for Learning Human Motion Patterns

Molina Mellado, Sergio, Cielniak, Grzegorz and Duckett, Tom (2021) Robotic Exploration for Learning Human Motion Patterns. IEEE Transaction on Robotics . ISSN 1552-3098

Full content URL: https://doi.org/10.1109/TRO.2021.3101358

Documents
Robotic Exploration for Learning Human Motion Patterns

Request a copy
Robotic Exploration for Learning Human Motion Patterns
Authors' Accepted Manuscript
[img]
[Download]
[img] PDF
Robotic_Exploration_for_Learning_Human_Motion_Patterns.pdf - Whole Document
Restricted to Repository staff only

5MB
[img]
Preview
PDF
molina2021tro.pdf - Whole Document

4MB
Item Type:Article
Item Status:Live Archive

Abstract

Understanding how people are likely to move is key to efficient and safe robot navigation in human environments. However, mobile robots can only observe a fraction of the environment at a time, while the activity patterns of people may also change at different times. This paper introduces a new methodology for mobile robot exploration to maximise the knowledge of human activity patterns by deciding where and when to collect observations. We introduce an exploration policy driven by the entropy levels in a spatio-temporal map of pedestrian flows, and compare multiple spatio-temporal exploration strategies including both informed and uninformed approaches. The evaluation is performed by simulating mobile robot exploration using real sensory data from three long-term pedestrian datasets. The results show that for certain scenarios the models built with proposed exploration system can better predict the flow patterns than uninformed strategies, allowing the robot to move in a more socially compliant way, and that the exploration ratio is a key factor when it comes to the model prediction accuracy.

Keywords:human motion model, long-term data, mobile robots, Spatio-temporal Exploration
Subjects:G Mathematical and Computer Sciences > G400 Computer Science
Divisions:College of Science > School of Computer Science
ID Code:46497
Deposited On:08 Oct 2021 08:21

Repository Staff Only: item control page