Innovative Projects Realized

Intelligent recycled plastic boat, recycling plastics in the ocean

The ocean is receiving much of the plastic waste generated on land and it is calculated that yearly up to 12.7 million tons of them enters in our oceans bringing a drastic environmental problem. This project seeks to create an autonomous boat made with recycled plastics, to clean up plastic waste in the oceans.
The boat will be assisted by a drone, which will map the area in search of suspended debris, and then send their location to the boat for collection, achieving an integration of air and ground in an autonomous system. Part of the recycled plastics will be used later to 3D print new units allowing to clean more area, reducing costs and recycling waste.

Could Community-Based Sign Language Programs Serve as a Bridge for Deaf Adults to Post-Secondary?

The Sign Language Institute Canada (SLIC) is a community-based organization that aims to increase training and certification opportunities for teachers of American Sign Language (ASL) and la langue des signes Québécoise (LSQ) across Canada. SLIC has plans to implement a community-based sign language program, intended to provide deaf adults with pathways to employment, as sign language instructors, or continuing education, including college and university level education. The objective of this study is to explore the possibility of a community-based sign language program serving as a bridge for deaf adults to post-secondary. Through qualitative research with deaf adults who have participated in community-based adult education programs in the province of Ontario, as well as interviews with deaf adult education staff and teachers, this study aims to consider what a bridging program might look like, according to voices/signs of deaf adult learners, and move us in that direction.

Réseaux de Neurones, L’IA appliquée à l’imagerie médicale

Nous allons développer un outils d’aide à la décision pour les radiologues. À l’aide d’imagerie rayon X de la cage thoracique vue de face, il fournira une évaluation sur la probabilité que le patient soit atteint de 14 maladies communes (atélectasie, Cardiomégalie, épanchement, infiltration, Masse, nodule, pneumonie, pneumothorax, consodilation, oedeme, emphyseme, fibrose, épaississement pleural ,hernie). Cette liste précise vient du fait que nous nous basons sur une recherche nommé CheXNet qui a déjà de bon résultats. Nous voulons aggrandir le nombre de maladies détectées afin de couvrir celle au goût du jours comme la Covid par exemple.Cet outil mettra en évidence les régions à hauts potentiels en plus d’une suggestion de diagnostique (Maladie la plus probable) afin d’aider les radiologues dans leur décision.

Les avantages potentiels sont immense. Cet outils améliorera la qualité ainsi que la rapidité des diagnostiques. De plus, s’il y avait une liste d’attente pour les diagnostiques en radiologie, cet outil pourrait envoyer au radiologue en priorité les radiographies dans lesquelles il a détecté des maladies graves. Ainsi, le temps d’attente pour les cas urgent serait encore plus réduit permettant de sauver du temps très précieux dans ces cas.

Impacts of Diverse Cover Crop Mixtures on Soil Health and Soil GHGs

With losses of soil carbon constraining our ability to ensure regional food security, there is increased focus on using regenerative agricultural management to improve soil health and the sustainability of our agri-food system. Numerous knowledge gaps exist around how we can reliably increase soil carbon, and how regenerative principles may be applied to field conditions relevant to farmers in East Central Ontario, where a short growing season, variable topography, and shallow soils challenge farmers’ ability to adopt regenerative practices such as cover cropping. In this collaboration, Common Earth and Trent University researchers will establish a duplicated, controlled field trial to study soil carbon and nitrogen dynamics and greenhouse gas release as impacted by cover crop management, to better understand how we may best put and keep carbon in soil using regenerative practices.

UWB localization for Autonomous landing / Instrumentation landing of VTOL vehicles.

This project will focus on developing a system to find the precise location of a landing pad as seen from a drone / helicopter. The proposed system uses radio frequency ranging between the devices onboard and devices deployed at the landing pad. Together with accelerometer and compass data, the system can figure out the position and orientation of the drone / helicopter relative to the landing pad. This system will help to automate drone operations as well as will enable air ambulance services and offshore supply helicopters to be operated in poor visibility weather conditions.

Integrated Mobile Waste-to-Energy Process Technology for Marine Infrastructures

The field of plastic waste management is essential for a sustainable society that utilizes plastic waste for energy production and environmental safety. Plastic waste’s land filing and incineration have large environmental impacts due to greenhouse gas emissions. The mentioned challenges could be eliminated using a more reliable and controllable thermal source such as an inductively coupled thermal plasma torch, direct current plasma torch and microwave plasma torch. In this project, advanced torches will be simulated, designed and constructed to generate a thermal plasma jet and integrated into a processing tank on a lab-scale to enhance the chemical reactivity and produce clean gasoline and diesel fuels suitable for usage in ignition engines from plastic waste from marine vehicles. This integration shall improve the quality of end-products, eliminate tar and wax content, provide better temperature control in chemical reactors, as well as enhance the chemical reactivity and thermal cracking reaction parameters.

Sensor Fusion for Ocean Surface Wind, Wave and Current parameter Estimation and AIS Target Detection and Anti-Spoofing

High-frequency surface wave radar (HFSWR) is recognized as one of the essential tools for remote sensing of the ocean surface because it provides wide-area, all-weather, and near-real-time surveillance. It operates between 3 to 30 MHz and can detect a large area of thousands of square kilometers. Wind speed, wind direction, surface current, and significant wave height can be extracted from HFSWR Doppler spectra.
Although such methods can perform satisfactorily for ocean surface wave parameter estimation, there are other sensors that can provide better spatial resolution than HFSWR on its own. These sensors include Xband radar, which provides very good spatial resolution for parameter estimation, though usually with a reduced maximum range. Satellite imagery can provide parameter estimation over a very large area, however the spatial resolution is limited by the quality of the imagery and the temporal resolution may also be limited. By combining the data from all of these sensors, and possibly others, it is believed that estimates with high spatial and data resolutions for ocean surface wind, wave and current parameters may be extracted from the data from these various sources.

Natural Language Processing (NLP) powered Clinical Documentation

Our company is developing an AI-powered clinical documentation platform. In the proposed project, the intern will assist us in improving our Natural Processing (NLP) models designed to automate clinical documentation for Orthopedic Physical Therapy clinics. These NLP models will be capable of processing text comprising medical vocabulary and terminologies. With this project, we aim to develop more robust, reliable, and high-accuracy language models which meet the expectations and needs of our customers.

Partnership network growth & relationship management

We are designing this project as this year, one of our primary growth strategies is to leverage partnerships to accelerate our entry into global markets. We are looking for an ambitious individual who will lead this crucial operation and help drive this sales channel by creating and managing all international partnerships. This will greatly expand our reach and help introduce our user friendly and affordable software to dealers all over the world. The challenge is to design, test and modify an effective partner success strategy that will strengthen the Glo3D brand while still being scalable as we grow and enter new verticals. We also would like to automate a good chunk of this partner onboarding experience and require someone who can lead this initiative.

Stage en Analyse d’Affaire

Better Being œuvre dans le domaine du bien être corporatif en aidant les entreprises créatives ou Bcorp à améliorer leur culture de mieux être en créant des activités et programmes ludiques et instructifs à l’aide d’une équipe d’expert pluridisciplinaire et une approche holistique. Notre équipe d’experts est constituée de personnes passionnées, créatives et talentueuses (chef, nutritionniste, thérapeute, préparateur mental, kinésiologue et professeur de yoga et méditation) qui ont un intérêt pour le partage de connaissance et la centralisation d’expertise.
Le stagiaire aura pour but de formaliser et standardiser les différents processus d’affaires de l’organisation. BetterBeing est encore une start-up qui a besoin de mieux se structurer afin de faciliter les opérations quotidiennes. Le stagiaire devra alors à l’aide du fondateur cartographier l’architecture d’affaire de l’entreprise et communiquer à toute l’équipe les différents processus mis en place: Processus de vente, de créations d’activité ou de programme, de consultation en entreprise, de recrutement… etc Ce projet est extrêmement important pour l’entreprise car à l’aube de la commercialisation, il est important d’assurer que tout fonctionne au mieux et cela passe par une meilleure compréhension de toutes les étapes à suivre pour assurer la livraison de la valeur du service.

Visualization Platform Development and Optimization

Terris is creating an Earth Intelligence SaaS platform to help solve major challenges that many industries are facing, including the renewable energy sector and emergency response sector. Through this project, the intern will be engaged to take on specific tasks that will help build on the work Terris’ team has been developing. This work will help Terris to advance its’ 3D offerings from the advanced prototype stage to being commercially viable products.

Reinforcement Learning based Graph Convolutional Recommender Systems

This project aims to use and experiment deep learning technique on modern recommender systems such as Graph Convolutional Network. The purpose of this implementation will be to drastically improve recommendation structure’s benchmark. This will allow extract user’s embedding by mapping from pre-existing features that describe the user such as ID and relevant attributes.
In this project students will be integrated as a member of the advanced analytics research team that includes multiple PhD holders in relevant domains.
Students would work on the following main topics:
1. Implementing Graph Convolutional Networks as a recommender system.
2. Implementing an Actor-Critic Deep Deterministic Policy Gradient models for graph-based recommender systems.