Projets novateurs réalisés

Investigating the nutrient profile of customized pet treat products

The founder of Pawsome Concierge has worked in the Dairy industry for over 10 years and has lots of experience and knowledge in the overall food industry in Canada. When she realized there were no standards and regulations in Canada for the pet food industry, she took it upon herself to start making homemade 100% natural treats that are like no other, such as Pumpkin & PB product, Heart Beets product, Oatmeal Pumpkin product, Vegan product, Pupcakes, and Cat treat. Pawsome would like to advocate for higher standards in the pet food industry and ask for stricter laws and regulations. Currently, pet products must only abide by the packaging laws, which is very worrisome, especially with all the recalls that have happened in the last year. The health and wellness of our companion animals is a priority for PAWSOME Concierge. This collaborative project will focus on analyzing the nutrient profile of selected Pawsome pet treat products and the results of the nutrient analysis will allow Pawsome them to make the appropriate claim for their pet food products and increase the consumers’ awareness of food safety and nutrients for their pets. Ultimately, this could help Pawsome grow and expand their market, create more jobs and bring social benefits to our society.

Search Result Aggregation Approaches for Multi-Source Content

Scrawlr is a platform for unconstrained, global interaction with all internet content and users. Scrawlr allows for user evaluation and unconstrained classification of any Scrawlr-hosted or non-Scrawlr content. For non-Scrawlr content, this evaluation and classification allowance will be first at the URL level but will subsequently be provided at the individual content component level.

The research conducted in this project will focus on designing, building, and studying novel search interfaces that effectively aggregate and present results retrieved from multiple data sources to satisfy searcher intent.

Optimizing a Fungal Leather Process

MycoFutures focuses on producing renewable, low-carbon materials from natural sources such as fungal mycelia. Traditional and synthetic leather (e.g. PVC) are in high demand but also have negative environmental impacts, including effects on global warming and eutrophication. MycoFutures has determined the production process for fungal-based leather material and now seek to improve their process by switching from solid state fermentation to liquid fermentation. A successful liquid fermentation process would create cost savings for the company due to increased ease of mycelial harvesting and processing.

Characterization of ionomer properties of the catalyst layers of polymer electrolyte fuel cells (PEFCs) Year Two

The state-of-the-art polymer electrolyte fuel cells have catalyst layers (CLs) made of Platinum catalyst on carbon support (Pt/C) bound together by proton-conducting polymer or ionomer. To overcome the challenges of high cost of Platinum catalyst ad corrosion of carbon support, alternative materials for catalyst and catalyst support are being considered. The interaction of ionomer with catalyst and its support materials controls two factors that profoundly affects the CL performance -(i) the micro-scale structure of the CL and (ii) ionomer properties in the catalyst layer. The proposed research aims to carry out a systematic study of – characterizing the properties of ultra-thin films of different ionomers on various substrates. This work fills in the gap of knowledge of how different ionomer types interact with different catalyst/catalyst support and ultimately affect the catalyst layer structure, property, and performance. The fundamental knowledge generated will be used to fabricate CLs layer with desirable characteristics of high catalyst utilization and enhanced electrochemical performance.

AI-based Model Predictive Control for Energy Management in Smart Buildings using Wireless Sensor Networks

The purpose of this project is to make use of the AI algorithms to enhance the energy consumption related to indoor heating ventilation and air conditioning (HVAC). The project involves collecting data using wireless sensors network, developing thermal dynamic model of a building, and developing Model predictive control (MPC) solver that reduced the energy consumption. Two selected PDFs are involved. The first PDF worked deeply in the developing the thermal dynamic models for buildings as well as developing the corresponding MPC-based solvers. The second PDF worked on AI-based design of wireless networks and will contribute in the data collection and the AI-model development. The partner organization main activity is to offer AI-based services in many fields and is interested to invest in the HVAC field given its AI expertise.

Effect of far infrared reflecting clothing on sleep physiology and sleep-dependent memory consolidation in healthy adults

Sleep is crucial for the formation of novel memories, and it underpins much of our psychological well-being. Unfortunately, millions of Canadians suffer from sleep difficulties, which are especially prevalent among women and populations with low education and income. These difficulties result in poorer cognitive performance and reduced well-being. Finding solutions to these difficulties is often complicated, as pharmacological interventions often have unwanted side effects, and the most efficient intervention (cognitive behavioral therapy) is hard to access for most of the population. The goals of the proposed project are to (1) test a new intervention based on far infrared technology, designed to help participants fall asleep faster and sleep better, and (2) identify which aspects of daytime functioning (memory, alertness, etc.) are impacted by this intervention. Our results will help clarify the viability of far infrared technology as an accessible, long-term sleep intervention.

Accelerating the Healthcare Leader’s Career Pathways: Determining pathways of leadership, and developing and testing a mobile app prototype

These are challenging times for healthcare leaders. Since the pandemic, it has become abundantly clear that healthcare is at a tipping point. With burnout, retirements, and resignations, healthcare leadership positions are now more available than ever. CHLNet and LEADs Change have studied leadership during COVID-19 and have highlighted the demands on leadership and the need for heightened strategic capabilities. Considering the unprecedented challenges to healthcare due to COVID-19 since March 2020 as well as advancement in technology in healthcare, it is imperative that we take a hard look at the realities of leading in a healthcare setting within the post-pandemic era. In this project, we will examine the emerging capabilities of successful healthcare leaders and the current state of the literature in a COVID context. Then, we will map out the capabilities and create a potential instructional medium using a mobile app design with digital resources for future healthcare leaders’ training.

Robot control using artificial intelligence for high diversity and low volume production sites

NeuroBotIA aims at tackling problematics of manufacturing SME such as the shortage of qualified workers. Currently, Canada is facing a labor shortage of 39% which will increase overtime due to younger generation interest in professional job and population growth decrease. Therefore, NeuroBotIA will develop technology to alleviate labor shortage using intelligent robots.
This project aims at training a Neural Network (NN) to powder coat mechanical parts on smart factory’s production line. The NN will be placed in a virtual environment to learn more rapidly and efficiently. The project is separated into 2 subprojects:
1. Design and training of the NN
2. Simulation of the powder coating
This project is a milestone in intelligent robots performing complex tasks. Indeed, powder coating is among the most complex tasks that has been attempted to be taught to an industrial robot. In addition, this project will contribute to the formation of highly qualified personnel (two graduate students).

Analysis of Canadian Western Red Spring Wheat Using Near Infrared Spectroscopy

The proposed research will develop a reliable analytical method for the quantitative analysis of grain grading factors for Canadian Western Red Spring wheat using near infrared spectroscopy (NIRS). Most commercially available near infrared spectrometers used in agricultural applications require significant expertise, a large footprint, a controlled environment, and are prohibitively expensive. The NIRS technology is effective for understanding grading factors, like protein content, and has demonstrated potential for other factors. This project will support the advancement of NIRS in agriculture by creating a user-friendly, portable, and cost-effective analytical method for wheat that can be performed by farmers in the field.

Blockchain-based federated learning for Agri-food supply chain

This project utilizes blockchain technology, smart contracts, and federated learning techniques to enhance collaboration between multiple parties for the agri-food supply chain application. It aims to promote data trust between multiple stakeholders in the supply chain by adopting blockchain technology and federated learning techniques. Multiple parties can keep their data at their local data storage, protect individuals’ privacy, and still contribute to model training and extracting knowledge. Blockchain technology is adopted as a distributed storage to securely verify data and distribute model training data on a ledger replicated across a peer-to-peer network. Smart contracts are responsible for the automatic execution of aggregating model updates and tasks of multiple parties.

Inspection of Bridges in Ontario for Damage Detection and Rehabilitation

This Mitacs project will undertake bridge inspections of 33 road bridges in Ontario to determine locations and level of damages in these bridges and suggest appropriate rehabilitation techniques to keep these bridges in service. The data collected will be added to the current pool of similar data already collected by the industry partner, MEDA to develop an inhouse bridge inspection manual. This manual will allow MEDA to be more productive and have a competitive advantage for the future bridge inspection projects. This Mitacs project will also undertake lab tests and computer modeling on concrete girders with various levels of defects. The data collected from this part of the project will be used to develop design guidelines for repair of damaged concrete girders of bridges. These design guidelines will help Canadian practicing engineers in designing the rehabilitation technique knowing the location and level of damages in the concrete girders.

Automatic Optical Character Recognition Preprocessing for Custom Gameplay Text

Computer Games are one of the key use cases of graphics cards of AMD. To ensure highest quality and performance, extensive testing of graphics hardware and software is required. However, much of this gameplay testing is manual and requires significant efforts due to varying styles in games and their versions. In this context, an open challenge lies in the difficult to automatically pre-process multiple heavily styled and color instances of text that appear in various games which current requires manual tuning. The goal of this project is to investigate and implement machine learning-based solutions to automatically pre-process, detect and recognize text in gameplay settings. The solution developed should be able to handle varying art, text, lighting, and user-interface styles and configurations as seen in games. The development of this framework would result in saving a significant amount of manual effort needed for automated testing of games.