Innovative Projects Realized

Deep learning-based Image Style Transfer with Generative Adversarial Network

The project aims to develop the deep learning-based algorithm that translate the image style of specific object to the reference style. Firstly, the proposed research focuses on identifying the accurate region in image for style transfer, and then translating the image style in that region. Current techniques about image style transfer are struggling to focus on translating the desired objects while keeping the rest of regions in the image unchanged. The competitive advantage gained by the new technologies developed through this project will help Crater Labs to further grow and expand its business in the computer vision, thus creating new employment opportunities in Canada.

Intégration d’enjeux de gestion durable des milieux riverains en aménagement forestier

La forêt canadienne est réputée pour sa diversité biologique, pour la qualité des bois qu’on en extrait, et de plus en plus pour la multitude de services écologiques qu’elle fournit aux humains, comme par exemple la villégiature, la chasse ou la cueillette de petits fruits, Malgré qu’elle soit très vaste, cette forêt n’est pas résistante à toutes les activités humaines et ses ressources ne sont pas illimitées. Pour assurer la meilleure conciliation des usages de la forêt, pour aujourd’hui comme pour le futur, il faut identifier les enjeux environnementaux, sociaux et économiques de chaque territoire forestier et planifier en conséquence. Rayonier Advanced Materials, une entreprise forestière opérant au Québec et en Ontario, s’associe à des chercheurs pour mieux intégrer l’information sur les multiples enjeux de ses territoires d’opérations et utiliser des outils de planification de pointe dans ses processus de gestion des milieux riverains forestiers.

Sustainable Funding Models for Watershed Co-Governance

This research project will analyze different funding options for watershed co-governance in the Nicola River watershed in British Columbia. There are ongoing efforts in the Nicola River watershed to develop and implement a co-governance structure involving the Province of BC and five Nicola First Nations. These efforts are made possible by short-term funding that is intended to last over the course of a two-year pilot project. Upon completion of the pilot in 2020, new sources of permanent, long-term funding will be required. Through background research and interviewees with those involved in the Nicola co-governance initiative, this project will analyze the trade-offs associated with different funding options. Rather than being prescriptive, the project is intended to provide information that will aid decision-makers in choosing a funding model that works in the context of the Nicola River watershed. This project will employ Structured Decision Making (SDM), a framework that allows for systematic analysis of complex, multi-stakeholder decisions.

Diversity Measures and Equal Opportunity Practices in the North American Corporate Labour Force and Global Diversity Management Practices

Visier Solutions Inc hopes to create a new standard for the measurement of employee diversity in organizations based on scientific research and measured baselines. With the help of the interns, this study will set a standard for the practical measurement of diversity in organizations as well as best practices in the analysis of metric results. The findings of this study will enrich our understanding of diversity and equal opportunity treatment of diverse groups in the corporate environment. Further, the study will expand Canadian knowledge on diversity management practices through engaging with the literature on global legal diversity definitions and management practices. The study will utilize quantitative and qualitative techniques through a statistical analysis of corporate employee data and a literature review on international practices leading to a sustained increase in the diversity of global enterprises.

Centrifuge modeling of liquefaction in layered silty sand deposits

Soil liquefaction or excess pore water pressure buildup reduces a soil’s stiffness and strength and can result in significant damages during earthquake shaking. The current state of practice on liquefaction evaluation techniques relies mainly on simplified empirical procedures based on case observations and free-field laboratory testing which have been widely limited to clean sands (zero fines). In this study, an investigation of the seismic response of layered saturated silty sand deposits is conducted by performing a series of dynamic geotechnical centrifuge experiments. The influence of fine-grained particles and layering on the material’s behavior are evaluated in both free-field and soil-structure interaction (SSI) model conditions. To the best of the author’s knowledge, this is the first application of centrifuge modeling on these liquefiable deposits. This project provides an important opportunity to study this material’s complex dynamic behavior by using an advanced testing program that will address the existing gaps in the literature.

Experimental investigation of controlled-rocking systems

The proposed project will investigate the behavior of high-performance reinforced concrete (RC) core walls, which are designed to rock about their base in two horizontal directions. For this purpose, a 1/6.5 scaled specimen of the 24-story reinforced concrete wall will be built and tested at the Multi-Array Shaking Table (MAST) facility at Tongji University, Shanghai, China. The experimental program includes a large array of unidirectional and bidirectional shaking tests of the specimen, which will investigate the performance of the system at different shaking intensities and different directions. The applicant research will focus on unidirectional and bi-directional rocking behavior of the walls. This experimental program will be performed at four phases of shaking table tests. In phase A, a cantilever-rocking-wall, equipped with supplemental dampers, will be tested unidirectionally. In phase B, the specimen will be tested in coupled-rocking-wall direction. In phase C, the cantilever-rocking-wall and the coupled-rocking-wall systems will be tested under bi-directional shaking. Finally, in phase D, all dampers will be removed and the cantilever-rocking-wall will be tested unidirectionally. The results of the experimental program will be used to calibrate advanced 2-D and 3-D numerical models to perfectly simulate the controlled rocking systems.

A precalibrated rate-dependent constitutive model for cyclic softening

When an earthquake happens, the soil underneath structures loses all or parts of its resistance. This condition is called liquefaction (for loose sands) or cyclic softening (for sensitive clays) as the soil acts as water during the earthquake. Liquefaction or cyclic softening can damage considerably man-made and natural earth structures and cause serious loss of human properties and lives. Examples can be found in several earthquakes such as the 1964 Niigata earthquake, the 1964 Alaskan earthquake, the 1999 Chi-Chi earthquake in Taiwan, and the 2011 Christchurch earthquake.

My project consists of developing a predictive model which can integrate the behavior of sensitive clays at the time of earthquakes. This model can be used in practice to simulate the effect of seismic loading on sensitive clays to predict if cyclic softening happens or not, or more simply, if clays lose strength and if they can hold the structures in an earthquake or not. One of the focus of this research effort is ensuring that the model is simple to use, yet with robust predictive capabilities so that practitioners can use of the model even if they do not have a solid background in mathematical equations.

Ecosystem Services of Native and Non-Native Urban Trees

In 2013, North America’s 1st ‘sustainable’ landscaping rating system will be launched: the Sustainable

Sites Initiative (SITES: www.sustainablesites.org). A major component of this guideline is the

preference for native plants because they are assumed to provide greater ‘ecosystem services’ than

non-natives. However, to date no research has been conducted in North America that compares the

relative value of native and non-native urban trees, and as such data on this crucial aspect of SITES is

lacking. The research proposed here will obtain the 1st North American estimates on the relative value

of native and non-native urban tree in terms of three major ecosystem services: (1) pollination, (2) biocontrol,

and (3) habitat for biodiversity. The data resulting from this research will be of broad

applicability to North American sustainable landscaping initiatives, such as the SITES.

How predictable is evolution: repeatability and variability in salmonids

Within species, there is variability among individuals in their characteristics and resource use. This variability is frequently represented as discrete forms, or morphs, that have differences in genetics, physical characteristics, and resource use. The evolution of morphs within fish species has occurred independently dozens of times within lakes at northern latitudes. This phenomenon is perhaps most incredible within species of the family Salmonidae, as there are several salmonid species in which three or four co-occurring morphs have been observed. Although multiple morphs have evolved in several independent instances, it is unknown to what degree the process of evolution has been similar in each case. For example, it is unknown whether the evolution of multiple morphs is associated with particular features of the environment, or whether the morphs exhibit repeated patterns in terms of their physical characteristics and resource use. In this project, a meta-analysis collating data from studies addressing evolution of salmonid morphs in individual lakes will be used to address these questions.

Dynamic Control Algorithms for Autonomous Path Planning and Tracking of Agricultural Wide-Span Implement Carriers (WSIC)

The proposed project aims to develop dynamic control systems for the autonomous operation of agricultural Wide-Span Implement Carriers (WSICs). The dynamic effects will be integrated into the control algorithms for the stationary and mobile mode of WSIC to further improve the precision and efficiency for path tracking of the machine. The proper control laws will be designed for path tracking and point-to-point displacement. The simulation or experiment for the WSIC platform based on the dynamic model will be compared with the results based on the existing kinematic models. The proposed project takes advantage of the expertise in dynamic control systems by the host professor Katupitiya (UNSW) and the WSIC designed by the home professor Laguë (uOttawa). Peer-reviewed publication is expected in this project. This project can also be the basis and a specific subject for my doctoral research project “Development of dynamic control algorithms for obstacle detection and avoidance by autonomous agricultural WSICs”.

A Prototype for a Privacy-Aware Health Sharing App

Patient-Reported Outcome Measures (PROMs) are validated questionnaires with which respondents self-report their health status. This may include symptom severity, functional status, and quality-of-life data. PROMS are an increasingly important part of the communication between a patient and their healthcare team. Beyond the individual encounter, PROMs data is also critical to healthcare systems providing more personalized care through “precision medicine.” PROM data helps give context to all the other sources of big data (e.g., laboratory/radiologic, genomic, wearable sensor) collected in healthcare.
The challenge, however, is collecting PROM data in a way that is secure and private, and in conformance with the specific desires of individual patients. By providing this assurance individuals can share such data with their healthcare providers, researchers, care givers, or their family with the assurance that it will only be used as intended. This is the challenge Promera is trying to address. TO BE CONT’D

Hey Neighbour Collective

Widespread decline in social capital is well-documented and has been attributed to a range of underlying root causes, from policies regarding the physical design of our neighbourhoods, increasing geographic mobility, shifting away from more localized economies, the complexity of increasing cultural diversity in many areas, to societal beliefs and norms related to individualism. We know these issues touch many of us, however, where you live makes a difference. Levels of social connection are lower in multi-unit buildings than in single detached houses. In Metro Vancouver, based upon 2017 research published by the Vancouver Foundation, renters, young people and low-income people are less likely than homeowners and older people to say they “feel welcome” in their neighbourhoods and to know neighbours well enough to ask for help. TO BE CONT’D