Projets novateurs réalisés

Adaptive Biofilm-Based apparel (ABBA)

Consumers are becoming increasingly interested in “smart” clothing with advanced properties that can enhance their comfort and performance. The next-generation of “smart” textiles can be fabricated by integrating living cells into the textiles that make up our clothing. For users of fitness apparel, living cells can endow clothing t-shirts or yoga pants with the ability to respond to changes in temperature, to the presence of sweat or to excessive stretching of the fabric. Working together, Lululemon and McGill engineers, aim at pioneering this field and demonstrating that “living textiles” can grow and adapt to the user and the environment. By incorporating harmless bacteria in textiles, the researchers will create some prototypes of “smart” textiles that can release fresh scents when soaked in sweat, “self-heal” after a tear or become more breathable in hot and humid environments. This project will position Canada as a leader in the emerging field of “smart living wearables”.

Digital Humanities, research-creation, and emerging hybrid media publication platforms: reimagining Public Journal

Academic journals rarely feature research-creation projects: as such, they are often limited by conventions inherited from print publishing. Recent efforts to create digital and/or online journals can potentially result in new publishing opportunities for scholars and artists. This project will examine how international online f journals in art history, visual studies, and film and media studies deal with hybrid media and research-creation. Public, a Canadian journal known for featuring both scholars and artists, is currently looking to increase its digital presence and to launch a new website that will enable it to better integrate research-creation projects and hybrid media scholarship. This project will enable Public to look at best practices and to design an effective digital publication strategy.

Development of an Information Sharing Platform for a Virtual Manitoba Oil Museum

This research project will organize and develop museum exhibits for a virtual Manitoba Oil Museum, using materials collected over the years by the museum as well as the dissemination of a PhD thesis on the Place of Oil in Manitoba. There is very little information in Manitoba broadly about the oil extraction that has been occurring since the 1950’s so this virtual museum would provide a site for sharing stories to a broader audience. This project plans to share a diversity of perspectives and experiences of living with oil extraction, to provide a more nuanced view into the struggles, benefits and costs of those in south-western Manitoba, particularly as dialogue around climate change and energy transition (including divestment from oil and gas extraction) becomes more important. The benefit to the partner organization, the Manitoba Oil Museum, will be to have a full-time coordinator to arrange and bring into being a site for sharing these stories and artifacts.

Simulation-based decision support system for data analytics deployment

Data has been recognized as one of the most valuable assets of modern business. The capacity to gather, store, analyze and interpret data in great quantities can determine to a large degree the ability of a company to achieve goals and adapt to largely volatile environments. This is especially true for financial institutions where data is directly connected to profitability. In the presence of a large number of relevant solutions to support automatic data analytics, implementation of analytics tasks and their deployment on computation infrastructure, as in cloud, involve complex decisions that can often lead to suboptimal results. Besides accuracy of analytics, this lack of optimality may lead to misused or wasted computation resources and loss of productivity for data scientists. In this project, our objective is to model this variety of alternative options by capturing their performance, such as latency and throughput, as well as their business aspects, including ROI and productivity. Based on this model, we will build a simulation engine capable of replaying what-if scenarios and guide data scientists towards more optimal solutions.

One-Shot Learning Based 3D Object Picking

Automation of processes is becoming more popular within logistics and material handling industries. As the amount of transferred objects grows there is a continuously increasing need for high efficiency, reduce cost, and of course reliability. With this project, we are utilizing the latest machine learning technology to improve the automation rate to a new level. This project is focusing on how to learn new object picking with limited training samples. In this way, our picking system is able to adapt to new scenarios easily. Meanwhile, since deep learning is more robust to shape or color changes of the object, we are able to apply this system to the warehouse, logistics, and agricultural industries.

Spraying of CO2-emulsions for the production of small droplets

Trapping of single mutated Amyloid-? peptides by heat

Acting big on the small scale: Designing a miniaturized soft robot

Implantation d’un systeme de gestion des stocks avec systeme de reapprovisionnement continu

L’entreprise Marc Villeneuve inc. est une PME qui oeuvre dans I’entreposage et la distribution de produits surgeles sur le marche quebecois. Afin de gerer I’inventaire de quelques 350 produits, elle doit effectuer des commandes a ses fournisseurs de maniere quotidienne. Presentement, le gestionnaire s’appuie sur l’inventaire du systeme informatique et sa connaissance du marche pour commander la quantite necessaire de chaque produit afin de minimiser les penuries. Le projet consiste en l’implantation d’un systeme de gestion des stocks qui appuiera le gestionnaire en fournissant un point de reapprovisionnement minimum et une quantite maximale a atteindre specifique a chaque produit. La portee du projet sera de 25-50 produits qui seront choisis par l’entreprise. L’etudiant devra d’abord etablir des previsions de ventes pour chaque produit concerne par le projet. Il devra ensuite determiner le type de systeme de gestion des stocks approprie pour l’entreprise et etablir les parametres requis pour son utilisation.

Speciation in Microorganisms: Pheromones and the Dating Game

Model Evolution and Data Transformation for Space Systems

Olympus : Aide à la décision pour imagerie PAUT

La méthode d’inspection ultrasonore PAUT (acronyme pour ‘Phased Array Ultrasonic Testing’) est une technique d’imagerie permettant la détection et la caractérisation de défauts de fabrication dans les assemblages et composants les plus critiques de l’industrie. Cette méthode, offerte avec certains produits Olympus tels que l’Omniscan ou le Focus PX (des produits conçus et fabriqués au site de Québec), est aujourd’hui principalement dépendante de l’expertise et la compétence d’analystes spécialisés dans le domaine du CND (Contrôle Non-Destructif). Le projet présenté ici permettra au stagiaire de participer à la réalisation d’une preuve de concept pour explorer le potentiel de l’IA (Intelligence Artificielle) avec pour objectif de réduire au maximum le temps d’analyse (sans compromettre la fiabilité) dans le cadre d’une application ciblée par Olympus.
Une problématique particulière au domaine CND pour le développement d’outils IA est la faible occurrence de défauts réels disponibles pour l’entrainement. Dans cette optique, Olympus souhaite se doter des meilleures pratiques pour exploiter l’augmentation de données réelles par la simulation; un générateur d’images synthétiques de défauts a été conçu à cette fin. Pour l’application ciblée, des données d’inspection sans défauts sont par ailleurs disponibles en quantité importante.