Innovative Projects Realized

BusPas Inc – Internationalization opportunities of smart mobility technology.

BusPas is a dynamic development company committed to advancing today’s mobility and shaping tomorrow’s. Through cloud and mobile technologies, we are transforming the mobility experience for users and supporting transit agencies in the digitalization of the city.

We have developed something new for transit operators and agencies. An innovative technology that will be deployed massively at bus stops in the coming years. We are a Montreal-based company and officially launched our product in November at the APTA 2021 Expo in Orlando. Our smart bus stop sign can help Transit Agencies to improve their customer experience, improve safety for their bus operators and users, and increase data availability for operational and planning purposes.

Today, we are seeking to internationalize our activities in order to further develop a new and high potential market in the transport industry for years to come.

We try to understand what is being done abroad, what solutions exist in major cities around the world. Therefore, in our international business intelligence, business development and marketing approach, we want to conduct international research to better understand the level of maturity of transit agencies around the world.

As well as having a rewarding work experience, Ashish will be able to put to good use the knowledge and skills developed in the courses at HEC Montreal. He will be mentored by our new Executive Director of Operations, Brad Cameron, and myself, having also been offered a Mitacs SSE with BusPas and HEC in 2020.

Enhancing a sales forecasting predictive model within the business rule engine using AI techniques

Flex Travel Solutions is preparing to launch its Machine Learning Module to enhance its BRE tool through predictive analysis. The objective of the project is to provide its clients a sales forecasting predictive model within the BRE to facilitate their business strategies and decisions.

Rapid control prototyping of adjustable speed drive systems for OPAL-RT coursewares with the OP8666 module

This project is focusing on the development of RCP coursewares for adjustable speed drive (ASD) systems with the OP8666 module. The OP8666 RCP module will be used to implement, test, and validate the control algorithms in real-time before being connected in the real-world ASD systems.
Two ASD courseware module families have been agreed for the development of RCP teaching laboratory. The first courseware module includes different electric motor drives such as PM, IM and DC machines, while the second one includes power electronic converter topologies such as DC-DC choppers, three-Phase Thyristor Bridge Rectifier, single-phase and three-phase 2-level inverter and finally three-phase 3-level NPC inverter.
With the developed RCP teaching laboratories, students and researchers in R&D centers or universities will learn and master adjustable speed drives with the OP866 module, which is fast enough to visualize electrical motor drives and power electronics phenomena that can be seen on more expensive and time-consuming analogue setups.

Performance and Robustness-Aware Deep Neural Network Compression for Next-Gen AI Accelerators

Deep Learning (DL) is seen by many as “the” framework to solve complex problems involved in numerous applications that influence people’s lives. It is widely used in safety and security-critical environments such as self-driving cars as well as drones and robotics. This makes it highly essential to secure DL algorithms and systems from malicious actors. To address this, recent work has focused on developing efficient and effective techniques to defend against such attacks on deep neural networks. One major challenge that state-of-the-art defense methods commonly face is that they often study large networks, even for simple tasks. This is impractical to deploy, especially in resource-constrained settings. This project evaluates and explores different methods to decrease the storage and computational costs of robust deep neural models by exploring the impact of model compression techniques such as quantization and pruning on network performance and adversarial robustness.

Tundra North Tours Sustainability Planning

The project being undertaken by the intern will be to develop a sustainability strategy/plan in the form of a report for the host organization’s Okpik Arctic Village. This report will be developed by the intern as they gain important and relevant information though consultations with related stakeholders, conducting environmental impact assessments (based on fuel consumption, water consumption, food consumption, and waste production), and observe all the activities that are undertaken at the village. This will benefit the intern as it will allow them to apply their learned knowledge from their recent education to a real world business. It will also provide the intern the ability to develop real world skills desired by employers. This opportunity will also benefit the host organization by subsidizing the fees associated with an intern that has a thorough understanding of sustainability and is able to utilize their education for the benefit of the business.

Automatic Segmentation of SCG Signal for Ischemic Patients

Research has shown that by monitoring the vibration of the heart using a simple sensor mounting on the human chest, the mechanical characteristics of heart can be measured. The purpose of this research is to design software that can automatically find some important points on the SCG signal. The software will be used to find hemodynamic parameters of the heart that will be used for diagnosis of ischemic heart patients. The outcome of this project is anticipated be extremely beneficial for both academia and the company. By successful completion of this project, multiple publications and patents will help improve the field of cardiac engineering.

Intact : Tarification par apprentissage par renforcement

La tarification en assurance de dommages est une des fonctions fondamentales et elle est traditionnellement entreprise par des actuaires. La tarification demande notamment d’estimer les coûts d’assurance (pertes attendues), les dépenses encourues ainsi que le profit attendu par la compagnie. En plus de ces éléments, la compagnie doit prendre en compte les forces du marché, les compétiteurs, les tendances et bien d’autres éléments qui peuvent affecter le prix du client. Les techniques de tarification sont en constante évolution et utilisent de plus en plus des méthodes d’apprentissage machine ou profond pour améliorer les capacités de segmentation des clients et aller chercher un avantage concurrentiel.
Intact cherche constamment à améliorer ses pratiques de tarification. Dans cette optique, l’utilisation de technique d’apprentissage profond par renforcement semble un filon intéressant à exploiter. En effet, l’intégration d’une politique de tarification qui serait capable de s’adapter aux tendances du marché, aux habitudes d’achat des clients et à l’évolution du risque de chaque assuré est un défi de taille que Intact aimerait relever.

Multi-view Neural Scene Inpainting

Capturing and manipulating 3D scenes is a topic of great interest with applications in photography, augmented reality, and robotics. Neural Radiance Fields, or NeRFs, is an emerging technology that enables high fidelity scene modeling and manipulation. This is of great interest to Samsung, as it can help create new photographic experiences for users. Our project focuses on a particular type of manipulation – object inpainting. This involves two components: segmenting the object from the scene with simple user interaction and inpainting the scene after object removal. Unlike existing NeRF manipulation work, we propose to train both components on large offline datasets of scenes and objects, enabling efficient and robust processing at test time without having to perform such learning on the fly. The intern will work in collaboration with other SAIC-Toronto staff and focus on the segmentation component of the project.

Ecology of Charlottetown Ponds Water Quality Parameters Biotic Indices and GIS Mapping Mitacs BSI Project

This project is based on both a continuation of yearly biomonitoring and ecological assessments of Charlottetowns ponds and extracting additional value from the data over the past five years (2017-2021). Each year and season is unique, so that continuing to capture this information means it will be permanently available allowing continuous trend analysis. The extra value from building a comprehensive report includes identifying changes in species counts and changes in water chemistry that may occur because of environmental factors (eg warmer water temperatures, decreased or increased rainfall or other inputs). An understanding of any negative water ecology trends would allow for development of a remediation plan.

Anonymisation et synthétisation de données transactionnelles

La science des données est une discipline clé pour la Banque National. Au cœur de sa pratique se trouve la gestion des données. La banque souhaite toujours créer plus de valeur pour ses clients grâce aux données, mais elle souhaite éga-lement protéger leurs données et empêcher tout mauvais usage qui amènerait un bris de confiance. Pour cette raison, elle cherche à réduire son risque de données sans perdre en vélocité ou en pouvoir décisionnel.
À l’aide de techniques innovantes telles que l’intelligence artificielle et une méthodologie scientifique robuste, ce projet vise à générer des données transactionnelles non sensibles, mais tout aussi utiles que les vraies données de la banque.

Exploring Digital Asset Regulatory Frameworks

In order to address expanding blockchain technology adoption and rapidly developing digital asset regulatory frameworks, this project, through research and reporting, will assist in developing an updated and clear understanding of regulatory environments to better inform businesses seeking to engage in this market.

Feasibility of high temperature thermal energy storage for the Calgary district heating system

See attached proposal doc for full overview (done on the old form and will not be reformatting for the online submission)