Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/134536
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Journal article |
Title: | Run-time monitoring of machine learning for robotic perception: a survey of emerging trends |
Author: | Rahman, Q.M. Corke, P. Dayoub, F. |
Citation: | IEEE Access, 2021; 9:20067-20075 |
Publisher: | Institute of Electrical and Electronics Engineers (IEEE) |
Issue Date: | 2021 |
ISSN: | 2169-3536 2169-3536 |
Statement of Responsibility: | Quazi Marufur Rahman, Peter Corke, Feras Dayoub |
Abstract: | As deep learning continues to dominate all state-of-the-art computer vision tasks, it is increasingly becoming an essential building block for robotic perception. This raises important questions concerning the safety and reliability of learning-based perception systems. There is an established field that studies safety certification and convergence guarantees of complex software systems at design-time. However, the unknown future deployment environments of an autonomous system and the complexity of learning-based perception make the generalization of design-time verification to run-time problematic. In the face of this challenge, more attention is starting to focus on run-time monitoring of performance and reliability of perception systems with several trends emerging in the literature in the face of this challenge. This paper attempts to identify these trends and summarize the various approaches to the topic. |
Keywords: | Machine learning; performance evaluation; reliability; robot learning |
Rights: | © This work is licensed under a Creative Commons Attribution 4.0 License. |
DOI: | 10.1109/access.2021.3055015 |
Grant ID: | http://purl.org/au-research/grants/arc/CE140100016 |
Published version: | http://dx.doi.org/10.1109/access.2021.3055015 |
Appears in Collections: | Computer Science publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
hdl_134536.pdf | Published version | 3.12 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.