Innovative Projects Realized

Efficient and low-complexity video coding for virtual reality and 360-degree video streaming

Virtual reality (VR) and augmented reality (AR) offer a unique immersive video experience by providing 360-degree video in a panoramic view. Limited bandwidth, demanding high quality, encoder delay, network latency and lack of standards are the main problems to deliver true VR immersive experience. To address these challenges, in this research, we intend to design a VR system based on learning concepts to provide efficient bandwidth usage where the encoder makes smart decisions to assign different qualities to different parts of the spherical frame based on the user’s view using features such as video content and user’s movement patterns. The 360 frame is split into segments such as tiles where the size, number and the quality of tiles are determined adaptively and on-line. TO BE CONT’D

Application of bacteriophage encapsulation in biodegradable polymers for the prevention of prosthetic joint infections

In 2014-2015, there were over 100,000 hip and knee replacement operations in Canada. Among these, more than 8,500 of these procedures had to be repeated. The main cause of failure was due to bacterial infections at the surface of the implants. Furthermore, this problem is likely to get worse over time due to antibiotic resistance phenomena in bacteria. We propose the development of a surface modification of the implants using bacteriophages, a class of viruses that only targets bacteria. These phages are omnipresent in the environment and our bodies, and therefore safe, and are not subject to antibiotic resistance. This polymeric surface treatment of orthopedic implants thus has the potential to greatly reduce the number of surgical revision procedures and save significant costs in both surgery time and materials.
Phagelux Canada had developed expertise in the use of bacteriophages in many fields including agriculture and animal health. TO BE CONT’D

Photometry- and Radiometry-based Precision Agriculture

Labor cost and availability are identified as the biggest challenges facing Canadian agriculture. The DOT Technology Corporation, as part of the SeedMaster, is to create an autonomous power system that acts as a primary locomotion unit for carrying a multitude of farm implements used for seeding, fertilizing, and tilling. This simple autonomous power will enable farmers to employ a number of farm functions through a shared, smaller, and economical locomotion base, called DOT. This project is on adding a myriad of sensors to the DOT power unit such as: visible-light cameras, Lidar, GPS, and IMUs. They will provide feedback information on the vehicle’s location, its direction of motion, and the environment it moves through, to the control system for real-time navigation and control. Furthermore, the information obtained via the imaging systems will be used for: (1) obstacle detection and avoidance, (2) crop-row identification, (3) motion planning for the DOT for optimal performance of the farm function it carries out.

Energy storage integration into electrified vehicle systems for shared and transit mobility applications

The deployment of electric and alternatively-fuelled vehicles in urban transportation constitutes a core component of current federal and provincial policies vis-a-vis Climate Action Strategies across Canada.
In the heavy-duty vehicle context specifically, the lack of standardized charging infrastructure combined with a lack of understanding as to the value of integrated energy storage devices to reduce or eliminate demand/delivery charges for high-powered charging constitutes an ongoing technology barrier to electric transit integration.
These important challenges will be addressed in this proposal through the partnership between the fellow and the Canadian Urban Transit Research & Innovation Consortium (CUTRIC). The active intervention of utilities and industry members will enable the researcher to capture proprietary data and technical information not normally available to academic researchers who are studying electrified transit and energy storage systems outside of industry partnerships.
TO BE CONT’D

Cancer tissue classification using machine learning/deep learning algorithms

Raman spectroscopy is a non-destructive laser-based optical technique that provides information on the molecular composition of biological tissue. Using this technique, it is possible to distinguish different types of tissue and use this information to develop prognostic tests to evaluate, for example, how a patient will respond to a specific therapy. Combined with other spectroscopy techniques, it is possible to acquires a wide range of information about the cancer/normal tissues in different organs and use this database to perform a very efficient tissue classification using the machine/deep learning algorithms. This research will lead to the development of new prognostic system, using spectroscopy techniques to predict and categorize tissues to for different types of cancers. These developments will setup the stage for future clinical trials using spectroscopy to improve treatment for patients affected with cancer, for example, the potential efficacy of immunotherapies.

Terahertz Distributed Sensing Platform Based on Waveguide Bragg Gratings – Year two

Distributed sensing is an advanced technology that enables real-time monitoring of variations along the entire length of a waveguide, and offers the possibility of sensing from a long distance. In the optics domain, distributed sensing based on optical fibers has been successfully demonstrated. However, the realization of distributed sensing in the terahertz domain is still at an embryonal stage. This project aims at introducing the concept of distributed sensing into the terahertz domain, and developing, for the first time to our knowledge, a prototype of novel terahertz distributed sensing platform by integrating multiple THz Bragg grating sensing units into a low-loss and high coupling-efficiency THz two-wire waveguide. TO BE CONT’D

Terahertz Distributed Sensing Platform Based on Waveguide Bragg Gratings

Distributed sensing is an advanced technology that enables real-time monitoring of variations along the entire length of a waveguide, and offers the possibility of sensing from a long distance. In the optics domain, distributed sensing based on optical fibers has been successfully demonstrated. However, the realization of distributed sensing in the terahertz domain is still at an embryonal stage. This project aims at introducing the concept of distributed sensing into the terahertz domain, and developing, for the first time to our knowledge, a prototype of novel terahertz distributed sensing platform by integrating multiple THz Bragg grating sensing units into a low-loss and high coupling-efficiency THz two-wire waveguide. TO BE CONT’D

Study of thermal energy storage solutions for a long-term, eco-friendly and cost-effective adiabatic compressed air energy storage system

It is vital to reduce today’s consumption of fossil fuels by using renewable energies like wind energy, solar or photovoltaic. However, majority of renewable energies are intermittent and not in phase with human electricity consumption cycle. It is therefore essential to develop energy storage systems to capture renewable energies when they are available and distribute them when needed.
Adiabatic compressed air energy storage (A-CAES) is a very promising technology. These systems are one of the few storage technologies capable of delivering several hours of high power. However, they need a device where a large amount of thermal energy can be stored. Also, this energy must be stored at a very high temperature in order to reach a good system efficiency. TO BE CONT’D

Application of Cold Spray Coatings for Repair of Aerospace Parts

There are currently no practical-economical repair methods for many aerospace parts made

of aluminum/titanium. The Cold Spray technology produces metal coatings by accelerating

metal powders onto the substrate to be coated using high velocity jet at supersonic speeds.

Upon impact the particles plastically deform and produce a dense coating. Metals such as

aluminum/titanium can be applied by this process without damaging the parts to be coated.

As such, it is believed that Cold Spray can be used to restore/repair damaged

aluminum/titanium aerospace coatings/parts as well as other parts. Its successful

implementation as a restoration/repair technology will result in major cost savings and reduce

schedule time and thus will allow the Canadian technologies developed and commercialized

by Centerline to be adopted by aerospace companies such as Boeing and generate

considerable revenues from selling the equipment, parts and know-how. The interns will

perform intensive research and development activities using the Cold Spray technology in

order to develop and complete the feasibility and qualification-certification testing….

Les enjeux éthiques et juridiques en intelligence artificielle appliquée à la santé

De nombreuses questions éthiques, juridiques et de politique publique en matière de régulation se posent dans le domaine de l’intelligence artificielle et restent encore sans réponse, celles qui nous intéressent s’énoncent dans le domaine de la santé mentale. En effet, la création des machines d’apprentissage se fait par l’entrée de données massives traduites en algorithme afin de prédire un diagnostique ou un traitement. La prise de décision de la machine est autonome et autant les développeurs que les utilisateurs ne peuvent l’expliquer. De ce fait, des questions se posent en matière de responsabilité si la machine fait une erreur de prédiction On se demande qui est responsable dans tel cas, le développeur, le médecin utilisateur ou la machine intelligente? Déléguer la décision à une machine, est-ce finalement un choix éthiquement responsable? Qu’en est-il de la protection des données personnelles des patients ? TO BE CONT’D

Facteurs influençant l’utilisation et l’efficacité d’une technologie d’assistance nouvelle génération pour le membre supérieur: le bras robotisé JACO

Les individus ayant des incapacités limitant l’usage de leurs membres supérieurs sont quotidiennement confrontés à des difficultés restreignant la réalisation de leurs activités de tous les jours. Leur autonomie, leur participation sociale et éventuellement leur qualité de vie en sont par conséquent affectées. Les technologies d’assistance nouvelle génération telles que les bras robotisés ont le potentiel d’aider grandement ces individus dans leurs activités et habitudes de vie. L’entreprise québécoise Kinova Robotics est un chef de file mondial des technologies d’assistance nouvelle génération ciblant le membre supérieur. Depuis 2009, l’entreprise commercialise le bras robotisé JACO, maintenant utilisé par des individus présentant des incapacités partout dans le monde, et est fortement engagée en recherche et développement (R&D) pour l’amélioration de ses produits existants et le développement de nouvelles technologies d’assistance et de services associés. Afin d’aider cette entreprise à centrer ses activités de R&D sur les besoins de ses clients potentiels, le projet proposé permettra d’évaluer l’utilisation actuelle du bras robotisé JACO, ses impacts sur les utilisateurs primaires, leurs proches aidants et les utilisateurs ressources ainsi que les attentes et besoins des différentes catégories d’utilisateurs en matière de technologies et de services associés à l’aide d’un devis mixte.

Évaluation de l’éco-efficacité par simulation de procédés dans le domaine de la transformation laitière. – Year two

Un prototype de logiciel d’évaluation de l’éco-efficacité des procédés de transformations des produits laitiers par simulation numérique de procédés industriels a été développé dans le cadre des travaux de doctorat de S. Benoit. Le présent projet vise à rendre cet outil utilisable par l’industrie laitière et à l’enrichir de fonctionnalités supplémentaires. En plus de quelques tâches d’améliorations des algorithmes, plusieurs fonctionnalités élargissant les capacités du logiciel seront ajoutées pour le rendre plus performant, la bibliothèque de procédés unitaires sera enrichie afin d’autoriser la modélisation de procédés plus variés, et la simulation des unités de filtrations par membranes fera l’objet d’une modélisation plus avancée dans le but d’améliorer le réalisme des résultats de simulation.
Sur la base du prototype développé, l’outil permettra d’évaluer parallèlement les dimensions environnementales et économiques des procédés laitiers modélisés afin d’en déterminer l’éco-efficacité et d’identifier les opportunités d’améliorations. TO BE CONT’D