Beyond Horizons - the Rise of the Edge AI Processing Paradigm.

Format:

Book

Author/Creator:

Vermesan, Ovidiu.

Series:

River Publishers Series in Communications and Networking Series

Language:

English

Physical Description:

1 online resource (117 pages)

Edition:

1st ed.

Place of Publication:

Milton : River Publishers, 2026.

Summary:

This book, a curated collection of research work presented at the European Conference on EDGE AI Technologies and Applications (EEAI), serves as both a ledger and a beacon for this exciting new era of edge intelligence-driven technologies.

Contents:

Cover

Half Title

Series Page

Title Page

Copyright Page

Dedication

Acknowledgement

Table of Contents

Preface

List of Figures

List of Tables

List of Contributors

Chapter 1: Advancing Edge AI Perception Platforms and Sensor Fusion for Last-Mile Delivery Autonomous Vehicles

1.1: Introduction and Background

1.2: Sensor Fusion in Last-Mile Context

1.3: Autonomous Vehicle Architecture for Last-Mile Delivery

1.3.1: Localisation and High-Definition Map

1.3.2: Perception Implementation

1.3.3: Prediction, Decision-Making, Planning and Route Optimisation

1.3.3.1: Odometry and path planning

1.4: Edge AI Platforms

1.4.1: Robot Operating System

1.5: Future Considerations and Research

1.5.1: Deployment Considerations

1.5.2: Future research

1.6: Conclusion

Chapter 2: AIDGE: A Framework for Deep Neural Network Development, Training and Deployment on the Edge

2.1: Introduction and Background

2.1.1: Related Work

2.2: Our Framework Overview

2.2.1: Internal Graph Representation

2.2.2: Platform interoperability

2.2.3: Graph Regular Expression (GraphRegex)

2.2.4: Network optimization

2.2.5: Export phase

2.3: Conclusion and future work

Chapter 3: A scalable and flexible interconnect-based dataflow architecture for Edge AI Inference

3.1: Introduction

3.2: Related Work

3.3: Background: dataflow execution models

3.4: Interconnect-based dataflow architecture

3.4.1: NGC: Neural Global Controller

3.4.2: NPE: Neural Processing Element

3.4.3: AINoC: Artificial Intelligence Network-on-Chip

3.4.4: Global Buffers

3.5: Execution Model

3.6: Experiments and Results

3.6.1: Evaluation Methodology

3.6.2: FPGA Implementation Results

3.6.2.1: Area

3.6.2.2: Latency

3.6.2.3: Energy consumption

3.6.2.4: Energy efficiency

3.7: Conclusion.

Chapter 4: Federated Learning for Malware Detection in Edge devices

4.1: Introduction and Background

4.2: Federated Learning and Related Work

4.3: Architecture

4.4: Experiments

4.4.1: Dataset

4.4.2: Evaluation results

4.5: Conclusions

Chapter 5: Image Signal Processor (ISP) Tuning using Machine Learning (ML) methods

5.1: Introduction and Background

5.1.1: Tuning problem

5.1.2: Image Signal processor (ISP)

5.1.3: Mathematical Optimization Problem

5.1.4: Static and Dynamic Parameters in ISP

5.1.5: State of Art

5.2: Automatic ISP Tuning

5.2.1: KPIs for Artifact Attenuation

5.2.2: Static Parameters

5.2.3: Dynamic Parameters and Runtime

5.2.4: Test Setup

5.2.5: Results

5.3: Conclusion

Chapter 6: Using Edge AI in IoT devices for Smart Agriculture: Autonomous Weeding

6.1: Introduction

6.2: Material and Methods

6.2.1: BIPBIP: the automatic weeding system

6.2.2: BIPBIP vision system

6.2.3: ANDANTE board integration

6.3: Reference Results

6.4: Work in Progress and Future Work

6.4.1: Work in progress

6.4.2: Future work

6.5: Conclusion

Index

About the Editors.

Notes:

Description based on publisher supplied metadata and other sources.

ISBN:

87-438-0866-2

87-438-0865-4

87-438-0867-0

9788743808657

OCLC:

1569121983

Publisher Number:

CIPO000319283