-
On-Sensor Convolutional Neural Networks with Early-Exits
Authors:
Hazem Hesham Yousef Shalby,
Arianna De Vecchi,
Alice Scandelli,
Pietro Bartoli,
Diana Trojaniello,
Manuel Roveri,
Federica Villa
Abstract:
Tiny Machine Learning (TinyML) is a novel research field aiming at integrating Machine Learning (ML) within embedded devices with limited memory, computation, and energy. Recently, a new branch of TinyML has emerged, focusing on integrating ML directly into the sensors to further reduce the power consumption of embedded devices. Interestingly, despite their state-of-the-art performance in many tas…
▽ More
Tiny Machine Learning (TinyML) is a novel research field aiming at integrating Machine Learning (ML) within embedded devices with limited memory, computation, and energy. Recently, a new branch of TinyML has emerged, focusing on integrating ML directly into the sensors to further reduce the power consumption of embedded devices. Interestingly, despite their state-of-the-art performance in many tasks, none of the current solutions in the literature aims to optimize the implementation of Convolutional Neural Networks (CNNs) operating directly into sensors. In this paper, we introduce for the first time in the literature the optimized design and implementation of Depth-First CNNs operating on the Intelligent Sensor Processing Unit (ISPU) within an Inertial Measurement Unit (IMU) by STMicroelectronics. Our approach partitions the CNN between the ISPU and the microcontroller (MCU) and employs an Early-Exit mechanism to stop the computations on the IMU when enough confidence about the results is achieved, hence significantly reducing power consumption. When using a NUCLEO-F411RE board, this solution achieved an average current consumption of 4.8 mA, marking an 11% reduction compared to the regular inference pipeline on the MCU, while having equal accuracy.
△ Less
Submitted 21 March, 2025;
originally announced March 2025.
-
Benchmarking formalisms for dynamic structure system Modeling and Simulation
Authors:
Aya Attia,
Clément Foucher,
Luiz Fernando Lavado Villa
Abstract:
Modeling and simulation of complex systems is key to explore systems dynamics. Many scientific approaches were developed to represent dynamic structure systems but most of these approaches are efficient for some kinds of systems and inefficient for others. Which approach can be adopted for different dynamic structure systems categories is a topic of interest for many researchers and until now has…
▽ More
Modeling and simulation of complex systems is key to explore systems dynamics. Many scientific approaches were developed to represent dynamic structure systems but most of these approaches are efficient for some kinds of systems and inefficient for others. Which approach can be adopted for different dynamic structure systems categories is a topic of interest for many researchers and until now has not been fully resolved. Therefore it is essential to explore the existing approaches, understand them, and identify gaps. To fulfil this goal, we identified criteria at stake for a smooth flow from model creation to its simulation for dynamic structure systems. Using these criteria, we benchmark the existing modeling formalisms focusing more on DEVS extensions, and use the results to identify approaches gaps and discuss them.
△ Less
Submitted 25 January, 2024;
originally announced April 2024.
-
Non-Line-of-Sight Tracking and Mapping with an Active Corner Camera
Authors:
Sheila Seidel,
Hoover Rueda-Chacon,
Iris Cusini,
Federica Villa,
Franco Zappa,
Christopher Yu,
Vivek K Goyal
Abstract:
The ability to form non-line-of-sight (NLOS) images of changing scenes could be transformative in a variety of fields, including search and rescue, autonomous vehicle navigation, and reconnaissance. Most existing active NLOS methods illuminate the hidden scene using a pulsed laser directed at a relay surface and collect time-resolved measurements of returning light. The prevailing approaches inclu…
▽ More
The ability to form non-line-of-sight (NLOS) images of changing scenes could be transformative in a variety of fields, including search and rescue, autonomous vehicle navigation, and reconnaissance. Most existing active NLOS methods illuminate the hidden scene using a pulsed laser directed at a relay surface and collect time-resolved measurements of returning light. The prevailing approaches include raster scanning of a rectangular grid on a vertical wall opposite the volume of interest to generate a collection of confocal measurements. These are inherently limited by the need for laser scanning. Methods that avoid laser scanning track the moving parts of the hidden scene as one or two point targets. In this work, based on more complete optical response modeling yet still without multiple illumination positions, we demonstrate accurate reconstructions of objects in motion and a 'map' of the stationary scenery behind them. The ability to count, localize, and characterize the sizes of hidden objects in motion, combined with mapping of the stationary hidden scene, could greatly improve indoor situational awareness in a variety of applications.
△ Less
Submitted 2 August, 2022;
originally announced August 2022.
-
DEIS: Dependability Engineering Innovation for Industrial CPS
Authors:
Erik Armengaud,
Georg Macher,
Alexander Massoner,
Sebastian Frager,
Rasmus Adler,
Daniel Schneider,
Simone Longo,
Massimiliano Melis,
Riccardo Groppo,
Federica Villa,
Padraig OLeary,
Kevin Bambury,
Finnegan Anita,
Marc Zeller,
Kai Hoefig,
Yiannis Papadopoulos,
Richard Hawkins,
Tim Kelly
Abstract:
The open and cooperative nature of Cyber-Physical Systems (CPS) poses new challenges in assuring dependability. The DEIS project (Dependability Engineering Innovation for automotive CPS. This project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 732242, see http://www.deis-project.eu) addresses these challenges by developing…
▽ More
The open and cooperative nature of Cyber-Physical Systems (CPS) poses new challenges in assuring dependability. The DEIS project (Dependability Engineering Innovation for automotive CPS. This project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 732242, see http://www.deis-project.eu) addresses these challenges by developing technologies that form a science of dependable system integration. In the core of these technologies lies the concept of a Digital Dependability Identity (DDI) of a component or system. DDIs are modular, composable, and executable in the field facilitating (a) efficient synthesis of component and system dependability information over the supply chain and (b) effective evaluation of this information in-the-field for safe and secure composition of highly distributed and autonomous CPS. The paper outlines the DDI concept and opportunities for application in four industrial use cases.
△ Less
Submitted 3 June, 2021;
originally announced June 2021.
-
39 Hints to Facilitate the Use of Semantics for Data on Agriculture and Nutrition
Authors:
Caterina Caracciolo,
Sophie Aubin,
Clement Jonquet,
Emna Amdouni,
Romain David,
Leyla Garcia,
Brandon Whitehead,
Catherine Roussey,
Armando Stellato,
Ferdinando Villa
Abstract:
In this paper, we report on the outputs and adoption of the Agrisemantics Working Group of the Research Data Alliance (RDA), consisting of a set of recommendations to facilitate the adoption of semantic technologies and methods for the purpose of data interoperability in the field of agriculture and nutrition. From 2016 to 2019, the group gathered researchers and practitioners at the crossing poin…
▽ More
In this paper, we report on the outputs and adoption of the Agrisemantics Working Group of the Research Data Alliance (RDA), consisting of a set of recommendations to facilitate the adoption of semantic technologies and methods for the purpose of data interoperability in the field of agriculture and nutrition. From 2016 to 2019, the group gathered researchers and practitioners at the crossing point between information technology and agricultural science, to study all aspects in the life cycle of semantic resources: conceptualization, edition, sharing, standardization, services, alignment, long term support. First, the working group realized a landscape study, a study of the uses of semantics in agrifood, then collected use cases for the exploitation of semantics resources-a generic term to encompass vocabularies, terminologies, thesauri, ontologies. The resulting requirements were synthesized into 39 "hints" for users and developers of semantic resources, and providers of semantic resource services. We believe adopting these recommendations will engage agrifood sciences in a necessary transition to leverage data production, sharing and reuse and the adoption of the FAIR data principles. The paper includes examples of adoption of those requirements, and a discussion of their contribution to the field of data science.
△ Less
Submitted 15 December, 2020;
originally announced December 2020.
-
Spatial images from temporal data
Authors:
Alex Turpin,
Gabriella Musarra,
Valentin Kapitany,
Francesco Tonolini,
Ashley Lyons,
Ilya Starshynov,
Federica Villa,
Enrico Conca,
Francesco Fioranelli,
Roderick Murray-Smith,
Daniele Faccio
Abstract:
Traditional paradigms for imaging rely on the use of a spatial structure, either in the detector (pixels arrays) or in the illumination (patterned light). Removal of the spatial structure in the detector or illumination, i.e., imaging with just a single-point sensor, would require solving a very strongly ill-posed inverse retrieval problem that to date has not been solved. Here, we demonstrate a d…
▽ More
Traditional paradigms for imaging rely on the use of a spatial structure, either in the detector (pixels arrays) or in the illumination (patterned light). Removal of the spatial structure in the detector or illumination, i.e., imaging with just a single-point sensor, would require solving a very strongly ill-posed inverse retrieval problem that to date has not been solved. Here, we demonstrate a data-driven approach in which full 3D information is obtained with just a single-point, single-photon avalanche diode that records the arrival time of photons reflected from a scene that is illuminated with short pulses of light. Imaging with single-point time-of-flight (temporal) data opens new routes in terms of speed, size, and functionality. As an example, we show how the training based on an optical time-of-flight camera enables a compact radio-frequency impulse radio detection and ranging transceiver to provide 3D images.
△ Less
Submitted 4 August, 2020; v1 submitted 2 December, 2019;
originally announced December 2019.
-
A prototype for a serious digital game to teach linguistic ontologies
Authors:
Diana Medina,
Grissa Maturana,
Fernán Villa,
Carlos Mario Zapata
Abstract:
The objective of ontologies is to increase the compression of a given domain by eliminating interpretation problems. Among kinds of ontologies are linguistics ontologies which are ontologies used to simplify the interface between domain knowledge and linguistic components. Digital games have received increasing interest from educators in recent years for their potential to enhance the language lea…
▽ More
The objective of ontologies is to increase the compression of a given domain by eliminating interpretation problems. Among kinds of ontologies are linguistics ontologies which are ontologies used to simplify the interface between domain knowledge and linguistic components. Digital games have received increasing interest from educators in recent years for their potential to enhance the language learning and linguistic learning experience. Within the literature are games to teach ontologies of a specific domain, and games that use ontologies to facilitate the understanding of a given domain. Other educational games teach linguistics or vocabulary in contexts in which language is useful and meaningful. Although games help to understand difficult topics, the use of games that seek to meet the learning objectives of linguistics is not very popular and those focused on teaching linguistic ontologies are scarce. To solve the lack of the recreational resource for teaching linguistics in this document a prototype of a digital game called onto-ling is proposed. The goal is for the player to learn the relationship between concepts according to semantics, types of concepts and relationships through a game of levels.
△ Less
Submitted 18 September, 2019; v1 submitted 15 September, 2019;
originally announced September 2019.
