Preliminary Analysis of Collar Sensors for Guide Dog Training Using Convolutional Long Short-Term Memory, Kernel Principal Component Analysis and Multi-Sensor Data Fusion

Guide dogs play a crucial role in enhancing independence and mobility for people with visual impairment, offering invaluable assistance in navigating daily tasks and environments. However, the extensive training required for these dogs is costly, resulting in a limited availability that does not mee...

Full description

Saved in:
Bibliographic Details
Main Authors: Devon Martin, David L. Roberts, Alper Bozkurt
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Animals
Subjects:
Online Access:https://www.mdpi.com/2076-2615/14/23/3403
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846124575825330176
author Devon Martin
David L. Roberts
Alper Bozkurt
author_facet Devon Martin
David L. Roberts
Alper Bozkurt
author_sort Devon Martin
collection DOAJ
description Guide dogs play a crucial role in enhancing independence and mobility for people with visual impairment, offering invaluable assistance in navigating daily tasks and environments. However, the extensive training required for these dogs is costly, resulting in a limited availability that does not meet the high demand for such skilled working animals. Towards optimizing the training process and to better understand the challenges these guide dogs may be experiencing in the field, we have created a multi-sensor smart collar system. In this study, we developed and compared two supervised machine learning methods to analyze the data acquired from these sensors. We found that the Convolutional Long Short-Term Memory (Conv-LSTM) network worked much more efficiently on subsampled data and Kernel Principal Component Analysis (KPCA) on interpolated data. Each attained approximately 40% accuracy on a 10-state system. Not needing training, KPCA is a much faster method, but not as efficient with larger datasets. Among various sensors on the collar system, we observed that the inertial measurement units account for the vast majority of predictability, and that the addition of environmental acoustic sensing data slightly improved performance in most datasets. We also created a lexicon of data patterns using an unsupervised autoencoder. We present several regions of relatively higher density in the latent variable space that correspond to more common patterns and our attempt to visualize these patterns. In this preliminary effort, we found that several test states could be combined into larger superstates to simplify the testing procedures. Additionally, environmental sensor data did not carry much weight, as air conditioning units maintained the testing room at standard conditions.
format Article
id doaj-art-5ec8799d7ec94f668e6fa14a0783d577
institution Kabale University
issn 2076-2615
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Animals
spelling doaj-art-5ec8799d7ec94f668e6fa14a0783d5772024-12-13T16:21:04ZengMDPI AGAnimals2076-26152024-11-011423340310.3390/ani14233403Preliminary Analysis of Collar Sensors for Guide Dog Training Using Convolutional Long Short-Term Memory, Kernel Principal Component Analysis and Multi-Sensor Data FusionDevon Martin0David L. Roberts1Alper Bozkurt2Department of Electrical Engineering, North Carolina State University, 890 Oval Dr., Raleigh, NC 27695, USADepartment of Computer Science, North Carolina State University, Campus Box 8206, 890 Oval Drive, Raleigh, NC 27695, USADepartment of Electrical Engineering, North Carolina State University, 890 Oval Dr., Raleigh, NC 27695, USAGuide dogs play a crucial role in enhancing independence and mobility for people with visual impairment, offering invaluable assistance in navigating daily tasks and environments. However, the extensive training required for these dogs is costly, resulting in a limited availability that does not meet the high demand for such skilled working animals. Towards optimizing the training process and to better understand the challenges these guide dogs may be experiencing in the field, we have created a multi-sensor smart collar system. In this study, we developed and compared two supervised machine learning methods to analyze the data acquired from these sensors. We found that the Convolutional Long Short-Term Memory (Conv-LSTM) network worked much more efficiently on subsampled data and Kernel Principal Component Analysis (KPCA) on interpolated data. Each attained approximately 40% accuracy on a 10-state system. Not needing training, KPCA is a much faster method, but not as efficient with larger datasets. Among various sensors on the collar system, we observed that the inertial measurement units account for the vast majority of predictability, and that the addition of environmental acoustic sensing data slightly improved performance in most datasets. We also created a lexicon of data patterns using an unsupervised autoencoder. We present several regions of relatively higher density in the latent variable space that correspond to more common patterns and our attempt to visualize these patterns. In this preliminary effort, we found that several test states could be combined into larger superstates to simplify the testing procedures. Additionally, environmental sensor data did not carry much weight, as air conditioning units maintained the testing room at standard conditions.https://www.mdpi.com/2076-2615/14/23/3403Conv-LSTMKPCAautoencodermanifold learningpattern recognitionguide dogs
spellingShingle Devon Martin
David L. Roberts
Alper Bozkurt
Preliminary Analysis of Collar Sensors for Guide Dog Training Using Convolutional Long Short-Term Memory, Kernel Principal Component Analysis and Multi-Sensor Data Fusion
Animals
Conv-LSTM
KPCA
autoencoder
manifold learning
pattern recognition
guide dogs
title Preliminary Analysis of Collar Sensors for Guide Dog Training Using Convolutional Long Short-Term Memory, Kernel Principal Component Analysis and Multi-Sensor Data Fusion
title_full Preliminary Analysis of Collar Sensors for Guide Dog Training Using Convolutional Long Short-Term Memory, Kernel Principal Component Analysis and Multi-Sensor Data Fusion
title_fullStr Preliminary Analysis of Collar Sensors for Guide Dog Training Using Convolutional Long Short-Term Memory, Kernel Principal Component Analysis and Multi-Sensor Data Fusion
title_full_unstemmed Preliminary Analysis of Collar Sensors for Guide Dog Training Using Convolutional Long Short-Term Memory, Kernel Principal Component Analysis and Multi-Sensor Data Fusion
title_short Preliminary Analysis of Collar Sensors for Guide Dog Training Using Convolutional Long Short-Term Memory, Kernel Principal Component Analysis and Multi-Sensor Data Fusion
title_sort preliminary analysis of collar sensors for guide dog training using convolutional long short term memory kernel principal component analysis and multi sensor data fusion
topic Conv-LSTM
KPCA
autoencoder
manifold learning
pattern recognition
guide dogs
url https://www.mdpi.com/2076-2615/14/23/3403
work_keys_str_mv AT devonmartin preliminaryanalysisofcollarsensorsforguidedogtrainingusingconvolutionallongshorttermmemorykernelprincipalcomponentanalysisandmultisensordatafusion
AT davidlroberts preliminaryanalysisofcollarsensorsforguidedogtrainingusingconvolutionallongshorttermmemorykernelprincipalcomponentanalysisandmultisensordatafusion
AT alperbozkurt preliminaryanalysisofcollarsensorsforguidedogtrainingusingconvolutionallongshorttermmemorykernelprincipalcomponentanalysisandmultisensordatafusion