Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

U-TOE: Universal TinyML On-board Evaluation Toolkit for Low-Power IoT

Zhaolan Huang, Koen Zandberg, Kaspar Schleiser, Emmanuel Baccelli

TL;DR

U-TOE, a universal toolkit designed to facilitate the task of IoT designers and researchers, by combining functionalities from a low-power embedded OS, a generic model transpiler and compiler, an integrated performance measurement module, and an open-access remote IoT testbed is presented.

Abstract

Results from the TinyML community demonstrate that, it is possible to execute machine learning models directly on the terminals themselves, even if these are small microcontroller-based devices. However, to date, practitioners in the domain lack convenient all-in-one toolkits to help them evaluate the feasibility of executing arbitrary models on arbitrary low-power IoT hardware. To this effect, we present in this paper U-TOE, a universal toolkit we designed to facilitate the task of IoT designers and researchers, by combining functionalities from a low-power embedded OS, a generic model transpiler and compiler, an integrated performance measurement module, and an open-access remote IoT testbed. We provide an open source implementation of U-TOE and we demonstrate its use to experimentally evaluate the performance of various models, on a wide variety of low-power IoT boards, based on popular microcontroller architectures. U-TOE allows easily reproducible and customizable comparative evaluation experiments on a wide variety of IoT hardware all-at-once. The availability of a toolkit such as U-TOE is desirable to accelerate research combining Artificial Intelligence and IoT towards fully exploiting the potential of edge computing.

U-TOE: Universal TinyML On-board Evaluation Toolkit for Low-Power IoT

TL;DR

U-TOE, a universal toolkit designed to facilitate the task of IoT designers and researchers, by combining functionalities from a low-power embedded OS, a generic model transpiler and compiler, an integrated performance measurement module, and an open-access remote IoT testbed is presented.

Abstract

Results from the TinyML community demonstrate that, it is possible to execute machine learning models directly on the terminals themselves, even if these are small microcontroller-based devices. However, to date, practitioners in the domain lack convenient all-in-one toolkits to help them evaluate the feasibility of executing arbitrary models on arbitrary low-power IoT hardware. To this effect, we present in this paper U-TOE, a universal toolkit we designed to facilitate the task of IoT designers and researchers, by combining functionalities from a low-power embedded OS, a generic model transpiler and compiler, an integrated performance measurement module, and an open-access remote IoT testbed. We provide an open source implementation of U-TOE and we demonstrate its use to experimentally evaluate the performance of various models, on a wide variety of low-power IoT boards, based on popular microcontroller architectures. U-TOE allows easily reproducible and customizable comparative evaluation experiments on a wide variety of IoT hardware all-at-once. The availability of a toolkit such as U-TOE is desirable to accelerate research combining Artificial Intelligence and IoT towards fully exploiting the potential of edge computing.
Paper Structure (12 sections, 3 figures, 4 tables)

This paper contains 12 sections, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. Background on TinyML Performance Analysis
  4. Performance Metrics
  5. Measurement Granularity
  6. U-TOE Toolkit Design & Implementation
  7. Toolkit Architectural Design
  8. Measurement Procedure
  9. Experiments Using U-TOE
  10. Analysing U-TOE Measurements
  11. Reproducible & Custom U-TOE Experiments
  12. Conclusion

Figures (3)

  • Figure 1: Hardware setup of U-TOE. Users can connect local boards with host PC via serial, or use remote board service on IoT-Testbed.
  • Figure 2: Compilation and deployment workflow of U-TOE. uTVM optimizes and translates model from mainstream ML framework into model library, which is co-compiled and flashed with RIOT and U-TOE components onto target boards.
  • Figure 3: Software architecture and components of U-TOE.