Projets novateurs réalisés

Suivi au champ de la croissance des racines d’espèces fourragères en fonction de la fertilisation et des coupes de foin

La croissance racinaire des plantes fourragères destinées à l’alimentation des bovins a été peu étudiée. Ce projet vise à évaluer l’effet de certaines pratiques agronomiques sur la croissance des racines de fléole des prés, de fétuque élevée et de luzerne. Pendant deux saisons de croissance, les racines de ces plantes ont été photographiées chaque semaine, directement dans le sol, à l’aide d’une caméra de type minirhizotron. Le projet de recherche international proposé consistera en l’analyse et l’interprétation des données issues de ces deux années d’expérience dans un laboratoire français spécialisé dans l’écologie fonctionnelle des systèmes racinaires des plantes pérennes. Les objectifs spécifiques seront d’établir l’effet de la source d’azote (lisier de bovin ou engrais de synthèse) et des coupes de foin sur la croissance racinaire de ces mélanges fourragers, ainsi que d’établir le taux de renouvellement annuel des racines, une information essentielle pour déterminer quelle quantité de carbone ces plantes peuvent entreposer dans les sols. Une meilleure connaissance du système racinaire des plantes fourragères pérennes permettra d’améliorer la performance agronomique et environnementale des systèmes agricoles.

Signal processing of EEG for detection and classification ofclinical brain states

measures the brain activity via electrodes placed on the scalp, or in the case of evaluation for epilepsy brain

surgery, intracranially. The EEG measures very complex time varying signals from multiple and spatially

complicated sources. Clinically, EEG interpretation relies on visual interpretation of signals less than 30 Hz,

ignoring the rich motherlode of information from higher frequencies. Advanced signal processing is required to

make the EEG more useful for assessing clinical states. This proposal applies advanced signal processing to: 1)

assess the levels of anesthesia in patients undergoing surgery, 2) study the neurodynamics of focal seizure

activity measured both with intracranial and extracranial EEG electrodes with a view to understanding the

features and location of most active seizure activity, and 3) extract gamma frequency from the scalp EEG from

controls and patients so as to develop a biomarker for disease states such as dementia and schizophrenia. All of

these projects clearly have significant clinical applicability and commercial potential.

Characterizing lytic bacteriophages against pathogenic E. coli: receptor specificity, efficacy in vivo, and genomic analysis

Many facets of human life are affected by bacterial pathogens. Bacteria can cause both animal and plant diseases. Extensive use of antibiotics to contend with pathogenic bacteria has resulted in an increase of multi-resistance to antimicrobials. Even when antibiotics are used vigilantly, their application in poultry or other fields of animal production can lead to increased spreading of drug resistance. Natural adversaries of bacteria – bacteriophages – are now considered as an alternative to antibiotics. Unlike antibiotics which show broad range killing, bacteriophages only kill specific bacterial species or strains, and are harmless for others. In poultry, avian pathogenic E. coli (APEC) is the predominant cause of respiratory and systemic infection in chickens and turkeys. During a Mitacs Accelerate grant, we isolated lytic phages with a broad killing activity against a variety of APEC strains. TO BE CONT’D

Développement d’une formation en ligne au travail avec interprète pour le réseau de la santé de la ville de Québec

La communication entre un patient et un intervenant est la base pour le travail en santé et service sociaux. L’intervention d’un interprète peut être essentielle en situation interculturelle, pour aller au-delà des différences linguistiques et culturelles qui conditionnent la qualité des soins. Le but de ce projet est de concevoir et valider un court programme de formation continue pour les intervenants en santé (notamment médecins, infirmières et professionnels de la santé mentale) portant sur le travail avec interprète. Cette formation permettra un meilleur accès aux services de santé et services sociaux pour les patients allophones et améliorera la qualité des services (évitant des erreurs, garantissant le suivi des traitements, etc.). Elle devrait aussi diminuer la discrimination systémique entraînée par le manque de connaissance sur ce type d’intervention.

Microfluidics Lab on a Chip device

There is no definite knowledge about how cells transforms and responds to the physical forces in human body. Recent studies shows that chemical signals, environment as well as mechanical properties can steer the cell behavior. Recent advancement in MEMS and micro fabrication techniques leads to the ability to study biological responses using a lab on a chip and organ on a chip devices. The potential to mimic exact same cellular level conditions made LOC a vibrant and rapidly-progressing field. Mechanotransduction can be studied in details by placing the cells in a micro porous membrane and applying a uniaxial stretch. Recent research solved the challenges to create a biaxial stretch which is more close to biological force condition compared to the uniaxial. Our objective of this research is to develop a micro fabricated multi axis stretching LOC device for studying mechanotransduction and biomedical applications.

An investigation of the effectiveness of competency-based management implementation

Competency-based management (CBM) is an approach to human resource management (HRM) that has grown in popularity over the years. CBM involves objectively defining the necessary competencies (defined as job-related knowledge, skills, abilities, and other attributes) to effectively perform a job, then integrating this framework into the organization’s HRM processes. While this approach makes sense in theory, there is little hard evidence available about its effectiveness for organizations. In an effort to bridge this knowledge gap, this project will involve conducting a comprehensive interdisciplinary review of the literature to determine what evidence currently exists about the effectiveness of CBM for learning and development initiatives in organizations. Importantly, the review will also assist organizations with preparing themselves to evaluate future CBM initiatives.

L’intégration de la culture, des attitudes et des pratiques en matière de gestion énergétique comme dimension du profil de durabilité chez les PME

Les PME sont des acteurs essentiels pour répondre aux défis que posent les enjeux climatiques qui nous confrontent. Elles représentent 99% des entreprises au Québec et génèrent des effets négatifs importants sur l’environnement. C’est pourtant le segment de marché le moins connu en matière de culture et de pratiques énergétiques. C’est donc dans l’intérêt commun (politiques publique, accompagnement) et dans celui du partenaire immédiat, soit l’Institut de Recherche d’Hydro-Québec, que le présent projet consiste à développer un questionnaire d’enquête fondé sur les connaissances les plus à jour sur le sujet qui servira à mieux comprendre ces acteurs, les pratiques qu’ils mettent de l’avant et celles qu’ils pourraient envisager, puis ce qui les stimule à s’engager sur la voie de l’efficience énergétique.

Optimal path planning for service robotics in a convention setting

CrossWing Inc. has been developing a service robotics platform called VirtualMe, which is a

multifunctional mobile robot base that can be configured for many applications. CrossWing is

attempting to have the VirtualMe used as an interactive guide/mascot for convention settings. The

goal of this project is to develop autonomous navigation capabilities for VirtualMe that would allow it

to function as a kind of mobile information kiosk that can serve as many people as possible. We

believe that this added functionality would be of use and add an interesting dimension to the gatherings

which invariably occur at public events.

Benefits to Crosswing include an added capability for VirtualMe that will allow them to expand into

new markets. They plan to introduce the interactive kiosk at a major international sporting event,

where they will have the opportunity to showcase their technology to the world.

A set membership filtering approach to low-complexity state estimation from PMU measurements

The widespread use of phasor measurement units (PMUs) in power-grids can greatly enhance state-estimation (SE) by making use of accurate, GPS time-stamped synchronous phasor measurements. Unlike conventional SCADA measurements which are reported every 4 seconds, synchro-phasor measurements are typically available as frequently as 30-60 measurements per second. While the availability of more measurements can provide accurate state estimates in real-time, the sheer amount of data can overwhelm the computational capabilities of most data processing systems. One potential approach to reducing the computational complexity of SE under high measurement rates is set-membership filtering (SMF). SMF algorithms are related to the normalized least mean squares (NLMS) algorithm. However the SMF algorithms not only exhibit better convergence and tracking properties, but also require parameter-updates only for a fraction of observations. This make SMF approach ideal for SE in PMU-equipped power grids. Even though SMF algorithms have been widely considered for many statistical signal processing applications including SE in communication networks, this rich class of algorithms has received very little attention in the context of SE in power grids.

Modélisation et caractérisation des capteurs Lidar embarqués dans les véhicules intelligents et leurs applications en détection/reconnaissance d’obstacle par fusion multi-capteurs Lidar-Vision

Le développement des systèmes de sécurité embarqués dans les véhicules a connu un progrès significatif ces dernières années avec l’apparition, notamment, de systèmes intelligents d’aide à la conduite (ADAS) et de la conduite automatisée. Parmi les capteurs utilisés, la combinaison des capteurs lidars et des caméras permet de mesurer et de reconstruire virtuellement l’environnement du véhicule et ses obstacles. Le projet proposé ici porte sur 1) la modélisation, la caractérisation et la validation des lidars dans différentes conditions d’utilisation et 2) la fusion lidar-vidéo pour la détection, la classification et le pistage d’obstacles en temps réel. Le projet d’une durée de trois ans implique la firme Phantom Intelligence de Québec ainsi que deux doctorants et un maitrisard membres du laboratoire LIV (Laboratoire sur l’intelligence véhiculaire) de la Faculté de génie de l’Université de Sherbrooke.

Development of a reject refining method in the magnesium bisulfite pulping process

Twin Rivers Paper previously installed and ran a reject refining process in order to improve their pulp yield. However, once the process was implemented it decreased the cleanness in the produced pulp which is undesirable for customers therefore the process was shut down. Currently the rejected chips are either being burned or sent to landfill but the issue with this is that there are still good fibers within those rejects that are able to be pulped. The objective of this research is to implement a reject refining process at the mill that will recover the good fibers from the true rejects in order to increase the overall pulp yield, which is economically beneficial.

Anomaly Detection in Financial Data

In this joint collaboration with Scotiabank we hope to solve a commonly faced problem by large financial institutions. It is to detect errors in financial datasets. This could be due to typing errors made by a human or a computer glitch that causes an incorrect value to be stored. To identify these errors, we plan to build an error detection system. It will model how financial variables change in relation to other variables. This will help us identify groups of variables that move, through time, in a similar manner. With this knowledge we will then be able to spot errors in the data. This system will enable the bank to build trading models that are not affected by these errors and reduce the time taken to identify them.