Applications of AI on Resource-ConstrainedHardware with a focus on Anomaly Detection

This thesis addresses the challenges of improving the performance of AI models on resource-constrained microcontrollers (MCUs). As the complexity of modern models continues to grow and the demand for smaller mobile devices increases, optimizing model latency, memory usage, and accuracy on tiny devic...

Full description

Bibliographic Details
Main Author: Ziegler, Travis
Other Authors: Oliva, Aude
Format: Thesis
Published: Massachusetts Institute of Technology 2023
Online Access:https://hdl.handle.net/1721.1/151408
_version_ 1811071600725852160
author Ziegler, Travis
author2 Oliva, Aude
author_facet Oliva, Aude
Ziegler, Travis
author_sort Ziegler, Travis
collection MIT
description This thesis addresses the challenges of improving the performance of AI models on resource-constrained microcontrollers (MCUs). As the complexity of modern models continues to grow and the demand for smaller mobile devices increases, optimizing model latency, memory usage, and accuracy on tiny devices remains a persistent problem. This thesis makes contributions to the field by (1) benchmarking common AI inference engines to identify trade-offs between them, (2) developing a framework that can assist neural-architecture searches to discover more efficient models, (3) proposing model conversion techniques that enable online-learning on MCUs, resulting in improved real-world accuracy, (4) creating a novel visual anomaly detector for MCUs, and (5) collecting a new dataset for anomaly detection benchmarks. The task of visual anomaly detection is to discern between known "Good" objects and objects that are slightly damaged or have imperfects. Being able to spot defective parts has huge applications in industrial and manufacturing settings. The proposed anomaly detector, MCU-PatchCore, is based on PatchCore – a state-of-the-art anomaly detector. MCU-PatchCore achieves a mean accuracy of 86% on the widely used MVTec AD dataset, which contains images of screws, cloth, glass bottles, etc., and their anomalous chipped, torn, cracked, etc., counterparts. While MCU-PatchCore’s accuracy is not as competitive as the GPU-based PatchCore detector, it only requires 200KB of RAM and less than 1MB of storage to run. Additionally, MCU-PatchCore outperforms a few other anomaly detectors in the literature, and shows promising potential for future improvement.
first_indexed 2024-09-23T08:53:45Z
format Thesis
id mit-1721.1/151408
institution Massachusetts Institute of Technology
last_indexed 2024-09-23T08:53:45Z
publishDate 2023
publisher Massachusetts Institute of Technology
record_format dspace
spelling mit-1721.1/1514082023-08-01T04:16:11Z Applications of AI on Resource-ConstrainedHardware with a focus on Anomaly Detection Ziegler, Travis Oliva, Aude Martie, Lee Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science This thesis addresses the challenges of improving the performance of AI models on resource-constrained microcontrollers (MCUs). As the complexity of modern models continues to grow and the demand for smaller mobile devices increases, optimizing model latency, memory usage, and accuracy on tiny devices remains a persistent problem. This thesis makes contributions to the field by (1) benchmarking common AI inference engines to identify trade-offs between them, (2) developing a framework that can assist neural-architecture searches to discover more efficient models, (3) proposing model conversion techniques that enable online-learning on MCUs, resulting in improved real-world accuracy, (4) creating a novel visual anomaly detector for MCUs, and (5) collecting a new dataset for anomaly detection benchmarks. The task of visual anomaly detection is to discern between known "Good" objects and objects that are slightly damaged or have imperfects. Being able to spot defective parts has huge applications in industrial and manufacturing settings. The proposed anomaly detector, MCU-PatchCore, is based on PatchCore – a state-of-the-art anomaly detector. MCU-PatchCore achieves a mean accuracy of 86% on the widely used MVTec AD dataset, which contains images of screws, cloth, glass bottles, etc., and their anomalous chipped, torn, cracked, etc., counterparts. While MCU-PatchCore’s accuracy is not as competitive as the GPU-based PatchCore detector, it only requires 200KB of RAM and less than 1MB of storage to run. Additionally, MCU-PatchCore outperforms a few other anomaly detectors in the literature, and shows promising potential for future improvement. M.Eng. 2023-07-31T19:37:23Z 2023-07-31T19:37:23Z 2023-06 2023-06-06T16:35:04.179Z Thesis https://hdl.handle.net/1721.1/151408 In Copyright - Educational Use Permitted Copyright retained by author(s) https://rightsstatements.org/page/InC-EDU/1.0/ application/pdf Massachusetts Institute of Technology
spellingShingle Ziegler, Travis
Applications of AI on Resource-ConstrainedHardware with a focus on Anomaly Detection
title Applications of AI on Resource-ConstrainedHardware with a focus on Anomaly Detection
title_full Applications of AI on Resource-ConstrainedHardware with a focus on Anomaly Detection
title_fullStr Applications of AI on Resource-ConstrainedHardware with a focus on Anomaly Detection
title_full_unstemmed Applications of AI on Resource-ConstrainedHardware with a focus on Anomaly Detection
title_short Applications of AI on Resource-ConstrainedHardware with a focus on Anomaly Detection
title_sort applications of ai on resource constrainedhardware with a focus on anomaly detection
url https://hdl.handle.net/1721.1/151408
work_keys_str_mv AT zieglertravis applicationsofaionresourceconstrainedhardwarewithafocusonanomalydetection