Projets novateurs réalisés

KASTx Ventures – Taylor Van Hell

The Marketing Analyst will be leading a stakeholder relations project at KASTx Ventures. This project comes at an important time for the company as we expand our service offerings for our portfolio companies.

Our portfolio companies are working to deepen community connections and expand the list of investors and venture capital firms they approach. This project will focus on building out key frameworks that can be adapted based on the needs of each portfolio company. It will require the analyst to build effective processes from lead generation to initial contact and continued follow up.

Strategy: The Analyst will be required to use existing information to prioritize leads and build outreach strategies to approach contacts.

Marketing: The Analyst will work with the marketing team to utilize existing marketing materials and build new materials as required to successfully move leads through the pipeline.

Operations: The Analyst will document existing processes while looking for opportunities and delivering key improvements.

Analytics: The Analyst will need to build effective reporting mechanisms to proactively determine process bottlenecks and communicate stakeholder relations activities.

Projects: The Analyst will work on ad hoc marketing projects at our portfolio companies.

Re-imagining Haryanvi Indo-Canadian Women: Investigating Intersemiotic, Cultural, and Postcolonial Translation in Ahoi Asthami and Dev Uthan Ekadashi

The project attempts to document two visual traditions- Ahoi Asthami and Dev Uthan Ekadashi conducted by and for Haryanvi Indo-Canadian women which happens only once a year, is fast changing due to rapid modernization by offering an ethnographic study and subsequent translation in Canada. An extensive literature review revealed an almost complete lack of theoretical, cultural, and ethnographic research on any Haryanvi Indo-Canadian tradition and therefore my research fills this gap. The project will assess the relationship between Haryanvi Indo-Canadian communities and their folklore by translating their culture through the framework of interrelated theories such as Intersemiotic, Cultural, and Postcolonial Translation under the supervision of host supervisor Prof Paul Bandia who holds expertise and is internationally renowned for translating oral, non-western, and minority cultures. My research tries to transcend the linguistic, geographical, and socio-cultural restrictive translation barriers (since a translator often encounter challenges owing to the lack of knowledge of target language’s linguistic and cultural knowledge systems) and will deliver a research model for similar translation practices within other geographical, socio-cultural, and linguistic spaces and thus can provide an insight into translating new media and digital cultures.

Empowering Companies & Communities to Build the Future Workforce

Purpose
Building a Better World by Bridging Skills Gaps – One Gap at a Time to Future-Ready the Workforce

Problem
More than 85 million jobs could go unfilled by 2030 (KornFerry 2022) because there aren’t enough skilled people. In Canada that’s $50B of unrealized GPD (Deloitte 2022) .

Optimization of DC-DC converters for DC microgrid interfaces

Distributed Energy Resources are key elements in de-centralized power systems. By deploying them close to the consumer
nodes, one can address the increases in power demand while bypassing investment in transmission and distribution systems.
The proposed research work concerns the optimization of DC-DC interfaces for DC microgrids. It will consider different types of
semiconductors (Si, SiC, GaN) of different ratings. This allows the reduction of the switching frequency and consequently power
losses and cooling requirements. The basic bi-directional DC-DC converter (cell) consists of two switching poles, one
intermediate inductor and two filter capacitors. By combining cells in parallel, one can reduce the current ratings of the switching
poles and inductors, while in series, one reduces the voltage ratings of switches and capacitors. In practice, the way the
interfaces are realized have a significant impact on characteristics such as high efficiency and power density with low cost,
volume and weight.

A robot control strategy for automated compound sanding of gypsum plasterboard

Nowadays, many building contractors use modular construction, where various parts of a building are partially or completely built on an assembly line and then put together on the delivery site. Building interior modules often involve building and finishing plasterboard walls, including closing joints between panels and hiding fasteners or defects. To do so, a gypsum-based compound is applied, let to dry, and sanded iteratively to achieve a regular surface. This produces a considerable amount of fine dust, which can be a nuisance, especially in the closed space of a manufacturing plant. In this project, we are investigating how a robot manipulator arm, based on an automatically acquired model of the surfaces, can plan and execute the sanding task. Various force-based and position-based control schemes will be evaluated and prototype will be demonstrated on a real production line.

Enquête sur la commercialisation des nanotechnologies au Canada

Les objectifs de ce projet visent à déterminer l’impact de la collaboration université-entreprise sur la commercialisation des nanotechnologies, à déterminer l’impact du financement public de la recherche universitaire et à travers des contributions directes aux entreprises sur la commercialisation, et à étudier les effets des retombées de connaissances ‘capturées’ par les entreprises avec ou sans l’aide de la collaboration avec les universités sur la commercialisation de nanotechnologies au Canada. Afin d’atteindre ces objectifs, nous conduirons une enquête par questionnaire web auprès des organisations et entreprises canadiennes de nanotechnologie, incluant les ‘spin-offs’ universitaires. Par ailleurs, nous conduirons des entrevues semi-dirigées auprès des directeurs et gestionnaires de la R-D des entreprises canadiennes de nanotechnologie les plus innovantes de façon à obtenir de l’information pertinente sur un grand nombre d’innovations dont la profitabilité des produits qui en découlent. Cette étude inédite permettra de mesurer l’importance des facteurs influençant la commercialisation des nanotechnologies.

Reinforcement Learning Based Constrained Control Applications

The main issue of the proposal hinges on the application of Data driven approaches for Model Predictive Control (MPC) applications. MPC is an optimization based control methodology well known in literature which is extremely popular when considering constrained control problems. The main drawback of MPC is the need of an accurate model plant to solve an optimal control problem on-line, under real-time constraints.
Data-driven control approaches mitigate the issue of model construction and tuning in two ways. In a more traditional way, using black (or gray) box system identification to come up with linear or nonlinear (such as neural) prediction models from data (model-based MPC). Alternatively, learning directly the control law without going first through an open-loop prediction model (model-free MPC). The main idea is to combine learning algorithms with control techniques like linear quadratic regulation and MPC, so to synthesize an optimal policy for the real process directly from data, without going first through time-consuming modeling and MPC tuning efforts, therefore reducing the overall numerical burdens to ensure safety, robust performance and overall stability.

Development, techno-economic analysis, and life cycle assessment of a novel CO2 utilization method for polyethylene production

The purpose of the project is to manufacture low- and high- density polyethylene from CO2. These products are the result of an energy-intensive process, steam cracking. This process is known for its high CO2 emissions, and this is something that the whole world is trying to reduce, because of its global warming impacts. Therefore, CO2 will be captured from flue gas, and added to hydrogen, which will be produced via the electrolysis of water. These compounds undergo a reaction known as reverse water-gas shift, and produce CO and water. Adding H2 to CO forms syngas, which will then be converted to ethane. Ethane will then in turn be converted to ethylene via a process known as chemical looping oxidative dehydrogenation. From there LDPE and HDPE can be produced. It is expected that such a process will emit less CO2 than steam cracking, and even have negative CO2 emissions
in areas where renewable electricity is used.

The effect of genre on syntactic complexity in English L2 writing

By using second language (L2) written corpus data, this study explores the effect of different genres (i.e., argumentative,
expositive, narrative, and descriptive) on syntactic complexity measures. Previous studies have noted that specific genres require unique sets of linguistic features, including measures associated with syntactic complexity. Therefore, the analysis will identify which complexity features are uniquely associated with specific genres. The expected outcomes include contributions to the measurement of complexity in L2 writing and pedagogical approaches for helping L2 speakers develop the complexity features associated with expert writing.

Building Footprint Extraction from Remotely Sensed Data

The primary objective of this project is to automatically extract accurate footprint maps of buildings from remotely sensed data
such a satellite/aerial images. In recent times, deep learning approaches have shown significant progress and success in
processing remote-sensing image data. Most prevailing techniques used for this task often involve using pixel-level classification
deep learning techniques that involve extensive post-processing to generate the required building footprints. These methods often
fail to accurately capture the boundaries and corners of building footprints which are crucial in several downstream tasks.
Therefore, this project will focus on developing an end-to-end deep learning neural network architecture that can directly predict
accurate building footprint polygons from remote sensing data. These resulting building footprint polygons can be used directly in
downstream GIS, mapping, and reconstruction tasks, without the need for any post-processing operations.

Enhanced Perception for Autonomous Truck Mounted Attenuator (ATMA) to Increase Work Zone Safety

This project makes an existing Autonomous Truck Mounted Attenuator (ATMA) system fully operational for Canadian harsh weather conditions and develops an augmented perception framework to enhance motion planning of the control system. The primarily focus of ATMA is ensuring the safety of highway workers and transportation infrastructure in work zones. On successfully understanding the existing control system, perception module will be augmented visual and radar information for reliable decision making and control in various environmental and highway driving conditions. This is to address low visibility and perceptually degraded conditions (by Camera or LiDAR) in harsh weather scenarios. To enhance situation awareness, decentralizez sensing methods will be used through connectivity of the system to the leader truck. This communication will improve perception of the leader/follower trucks used in the ATMA system, thus reliability of the control system; its effectiveness will be evaluated through the Safety of the Intended Functionality (SOTIF) metrics.

Improving efficiency of federated learning using access network infrastructure

The first aspect of this project aims to address the so-called digital divide, the challenges for broadband access in remote areas. The second is data privacy in artificial intelligence solutions. With the advent of 5G, we can now use fixed wireless access (FWA) to enable broadband access in remote areas. It consists of installing a few radio equipment while deploying a core network in central locations around a community with connections back to the Internet. In order to support the FWA operation, this research project proposes improvements in the scalability and convergence of the so-called federated learning algorithms (FL). FL enables multiple local actors to build a common, robust ML model without sharing the data, consequently addressing critical issues related to data privacy. We expect the proposed solution to reduce the convergence time and improve the accuracy of FL, making it suitable for use in practice.