Innovative Projects Realized

Guidance and Control of a Tandem-rotor Helicopter Landing on a Moving Platform

The development of autonomous unmanned aerial vehicles (UAVs) is a growing area of interest. An important step in the creation of a fully autonomous flying vehicle is the ability to precisely and smoothly land on a target. The goal of this research is to develop a UAV landing system that is able to track and land on a moving platform. The project will involve developing a guidance and control system that can plan a descent trajectory and track it down to the platform. The proposed system will be robust to strong wind gusts and still provide a smooth touchdown to avoid damage. The advantage to NGC is that this autonomous system would expand the capability of their UAV platform. This system can be applied not only to the current project. but to a variety of different aerospace vehicles and applications.

A novel water-based air filtration technology using cryogel biomaterials

We have a new technology at the Heart Valve Performance Laboratory, University of British Columbia, Okanagan Campus, which can be used for the new generation of the facial mask. This biomaterial which is highly hydrophilic will be utilized for the construction of a water-based air-filter mask because it is practically 99% water and 1% polymer, i.e., solid state water. Through the process of filtration, all small and large particles along with viruses and bacteria are trapped by the proposed mask and then can be properly disposed.

Self-tuning servers within IBM cloud

Modern cloud-based applications are deployed as isolated processes in containers or virtual machines. These applications frequently require tuning by the application developer (or a DevOps engineer) in order to extract the requisite performance. For example, a Java application executing within a Docker container may have radically better performance if the host JVM runs with a specific garbage collector policy, or if the Docker container has the appropriate amount of memory and virtual CPU resources. This application-specific tuning is today a manual process. And, it may be unwieldy for engineers to perform this tuning for complex applications that execute across multiple machines and communicate through message passing.

In this project we will work with the IBM hybrid cloud product team to develop new algorithms and systems to improve the automatic tuning and management of cloud-based applications. Our focus will be initially on improving the performance of Java applications. We may later branch out to consider clusters of applications (e.g., Microservices) that must be tuned in aggregate to optimize overall system performance.

Sources and sinks for potentially hazardous elements in mine waste

Mining is important to the Canadian economy but also generates large quantities of waste rock. Water percolating through some waste rock piles dissolves metals that can become hazardous to ecosystems. Laboratory tests are used to simulate this process. The intern will examine a suite of samples from a multi-year laboratory test to identify the source minerals and new minerals (sinks) that may have formed during the experiment. She will also work with the partner to determine the mineral reactions that occurred during the tests. The focus will be on nickel and cobalt, two metals mined in Canada and used for electric car batteries. The partner organization will benefit from access to cutting-edge analytical equipment at Queen’s

Efficient Deep Learning Methods that Only Require Few Labeled Data

In this postdoc, we plan to focus on computer vision tasks where existing deep learning methods require lots of labeled samples to work well. Acquiring labeled samples is time-consuming and often impractical. Thus, we investigate three different classes of methods to alleviate the label scarcity problem: active learning, weakly-supervised learning, and few-shot learning. In active learning, the goal is to label the most important samples to maximize the performance of the model while reducing labeling costs. In weakly supervised learning, the goal is to train models using weak labels. As a result, we can annotate samples with incomplete labels that are cheaper to collect. For instance, instead of annotating full masks for the task of segmentation, a cheaper alternative is to only annotate a single point per object. In the few-shot setup, the goal is to have models that can perform a task with only a few labeled samples. For these problem setups, we plan to propose new methods by building on our previous work and on promising methods that exist in the literature. Having deep learning models that can learn well with few labeled data can lead to significant contributions to the scientific community and many real-life applications.

Eliminating weight stigma in prenatal health care: Development of sensitivity training modules for prenatal care providers

As obesity (body mass index [BMI] >=30.0kg/m2) increases, there is also an increased risk for experiencing weight stigma (WS). WS is defined as having negative attitudes towards individuals because of weight. WS experienced in health care settings may cause poor communication with physicians, and negative health outcomes. A population that may experience WS are pregnant women who have a BMI >=40.0kg/m2 (high level of obesity). To prevent WS during pregnancy, health care providers may need sensitivity training. The proposed project will develop and test a sensitivity module for prenatal health care providers. A obstetrician will take the sensitivity training and deliver a appointment for pregnant woman, while we measure their response by questionnaires, interviews and heart rate monitoring to determine the importance of sensitive communication. The women will inform us how we can improve the sensitivity module. The updated module will be delivered to health care providers who are members of The Society of Obstetricians and Gynaecologists of Canada (SOGC). Health care providers will complete a questionnaire right after finishing the module and one month later, to determine if the module improved sensitivity in their practice. Finally, the module will be updated and made available to all SOGC clinicians.

Using Food Bank Client Data to Better Understand the Impacts of COVID-19 in Saskatoon

The SFBLC has been collecting detailed information on their clients and their socio-demographic and economics characteristics for more than five years. However, apart from collating overall summary numbers on things like user counts, which they report annually to Food Banks Canada, they largely lack the internal capacity to do a detailed analysis of this data and explore ways that it can inform their operations and planning. They have even fewer resources that they can allocate for exploring ways that the organization’s data can be used to inform questions of wider concern to the community or to social research in Canada generally. This lack of capacity has become particularly acute in the context of the COVID-19 crisis. The primary value of the internship will be to begin to fill these gaps. It will also lay a foundation for future collaborative work between the SFBLC and the academic research community. Additionally, the knowledge produced from this work, can be used to inform wider-policy in Canada. This project will also provide an adaptable framework and methodology for which organizations and researchers across the country may use for in their own social support efforts.

Influence of rearing and adult housing of laying hens on the behaviour and stress reactivity of offspring

Mothers can influence the behaviour and physiology of their young even before hatch. In birds that occurs through the deposition of hormones and nutrients into the egg. Mother-hens can be housed in various systems that offer different challenges to the bird. Our project will investigate how housing a mother-hen can affect the resilience of their offspring. This data can be used to improve farm practices and to decrease the occurrence of behaviour and welfare issues in egg production.

Canada and Snow Goose Herbivory Effects on Tidal Marsh Recession in the Fraser River Estuary

Tidal marshes are incredibly important bot ecologically and economically. While they provide many natural resources, they also provide flood protection and water filtration. However, many marshes are recessing into unvegetated mudflats. My project aims to find out if and how much goose herbivory contributes to this problem and compare the effects of Canada Geese and Snow Geese. To do this, exclosures will be set up in the mudflats and the marsh edge to keep either one or both geese species out. This will allow a direct comparison of how much damage one species does to the plants, relative to the other. After analyzing this data, I will try different new deterrents to scare and reduce herbivory of geese in tidal marshes. This project is being done in the hope of informing future management and restoration techniques throughout Canadian tidal marshes.

BCCSA Silica Control Tool Gap Analysis

Inhalation of Respirable Crystalline Silica (RCS) generated during construction activities such as cutting or grinding can cause lung cancer and the irreversible lung disease. Exposure to RCS is common on construction sites because silica is a naturally occurring mineral that is present in many construction materials. The Silica Control Tool is a risk assessment tool that was developed by the BC Construction Safety Alliance and the University of British Columbia to assist construction employers to create exposure control plans for managing the health risks of RCS on construction worksites. It uses an RCS exposure database to generate exposure estimates for construction activities both with and without exposure controls The Silica Control Tool was launched in 2017 and is freely available to all employers in British Columbia. It currently has over 2300 registered users. However, there remain gaps within the database which limit the applicability of the Tool. We aim to conduct on-site field measurements to determine the effect of these missing control methods to expand the applicability and effectiveness of the Silica Control Tool.

Polymer Nanoparticle-Based Vaccines for COVID-19

The current COVID-19 pandemic urgently calls for vaccines able to be rolled out on a global scale. This proposal aims to develop nanoparticle-based vaccine delivery vehicles that can enhance vaccination programs by improving vaccine storage and transport without need for refrigeration, as well as enhance the actual performance of the antigen, a protein found on the Sars-CoV-2 virus surface, in eliciting formation of neutralizing antibodies that protect against serious infection. These delivery vehicles are based on novel, responsive polymers able to enhance storage, delivery and action of the antigen.

Plant oil based resins for hemp biocomposite materials

Plant oils are attractive renewable feedstocks for the synthesis of biobased resins. However, the current bioresin technologies are unable to address certain challenges including (1) achieving a sufficiently high biocontent in the final product, (2) the synthesis of bioresins via green chemistry, and (3) a commercially feasible bioresin cost. In this research, the MITACS intern will, under laboratory conditions, make two novel, high biocontent resins that have been proposed for this work. The properties of these bioresins will be evaluated through the preparation of test biocomposite materials, in close collaboration with the company. This project will help the company to understand the potential of two bioresin types in their future planned manufacturing of composite materials.