Posts

Context Presentation

Anomaly detection in real fluid distribution applications is a difficult task, especially when we seek to accurately detect different types of anomalies and possible sensor failures. Our case study is based on a real context: sensor data from the SGE (Rangueil campus management and operation service in Toulouse).

We propose an automatic pattern-based method for anomaly detection in time-series called Composition-based Decision Tree (CDT). We use a modified decision tree and Bayesian optimization to avoid manual tuning of hyper-parameters. Our method uses sequences of patterns to identify remarkable points corresponding to multiple anomalies. The compositions of patterns existing into time-series are learned through an internally generated decision tree and then simplified using Boolean algebra to produce intelligible rules.

Our approach automatically generates decision rules for anomaly detection. All our experiments were carried out on real and synthetic data. We show that our method is precise for classifying anomalies compared to other methods. It also generates rules that can be interpreted and understood by experts and analysts, who they can adjust and modify.

Keywords

Anomaly detection, Time-series, Machine learning, Classification rules

Scientific goals

•    To detect different types of anomalies observed in real deployment

•    To generate interpretable rules for anomaly detection

•    To use learning methods for anomaly detection on static and continuous data

Contact

ines.ben-kraiem@irit.fr

Context Presentation

The current digital transformation requires the creation of autonomous applications that can be adapted to complex, dynamic, heterogeneous, and unpredictable environments. These systems must be equipped with proactive learning capabilities.

To this end, Self-Adaptive Multi-Agent principle allows the decentralization and self-observation of the learning process. Each knowledge granule is an autonomous agent that cooperates with its neighbors to improve learning from exogenous and endogenous feedbacks. Detecting and solving concurrences, conflicts and incompetence leads to active endogenous learning.This work on an adaptive decentralized learning mechanism is applied to the learning of a robotic arm inverse kinematics.

 Keywords

Self-Adaptive Learning, Endogenous Learning, Adaptive Multi-Agent Systems, Robotics

Scientific goals

•    Design a Self-Learning System

•    Lifelong and Endogenous Learning

•    Genericity and Scalability

Contacts

bruno.dato@irit.fr, frederic.migeon@irit.fr, marie-pierre.gleizes@irit.fr

Context Presentation

This research project deals with energy efficiency in buildings to mitigate the climate change. Buildings are the highest source of energy consumption worldwide. However, a large part of this energy is wasted, mainly due to poor buildings management. Therefore, being accurately informed about consumptions and detecting anomalies are essential steps to overcome this problem. Currently, some software exists to record, store, archive, and visualize big data such as the ones of a building, a campus, or a city. Yet, they do not provide Artificial Intelligence (AI) able to automatically analyze the streaming data to detect anomalies and send alerts, as well as adapted reports to the different stakeholders.

The system designed in the SANDFOX project has for objective to fill this gap. To improve the energy management, an innovative system should aim at visualizing the streaming data, editing reports, and detecting anomalies, for different stakeholders, such as policy makers, energy man-agers, researchers, technical staff or end-users of these buildings.

The paper presents the User-Centred Design approach that was used to collect the required needs from different stakeholders. The developed AI system is called SANDMAN (semi-Supervised ANomaly Detection withMulti-AgeNt systems). It processes data in a time constrained manner to detect anomalies as early as possible. SANDMAN is based on the paradigm of self-adaptive multi-agent systems. The results show the robustness of the AI regarding the detection of noisy data, of different types of anomalies, and the scaling.

Keywords

Anomaly detection, dashboard, multi-agent system, smart buildings, energy management

Scientific goals

•    Anomalies detection in smart buildings streaming data by AI,

•    Restitution of the information to different stakeholders through an adapted dashboard.

Contacts

berangere.lartigue@univ-tlse3.fr, stephanie.combettes@irit.fr, marie-pierre.gleizes@irit.fr,  mathieu.raynal@irit.fr

Context Presentation

The measurement of indoor air quality is important for health protection against chemical and gaseous pollutants ... The indoor air can contain many pollutants such as CO, CO2, NO2 and VOCs. These pollutants exist in different materials and products that can be used in housing (furniture, cleaners ...), but can be also coming from human activities or outside source. In this case, the detection, measurement and monitoring of these gazeuse contaminants is necessary.

In view of its high performance and low cost, the innovative gas multi-sensor based on metal oxides semiconductors for analyzing and controlling indoor air quality is a good alternative to electrochemical and infrared sensors. This project is currently in progress in LAAS in collaboration with the LCC and Laplace and as part of a thesis funded by neOCampus and the Occitanie region.

This thesis focuses on the characterization of multiple MOX-based gas sensors and integrates these multi-sensors in electronic card to achieve a connected object to control the indoor air quality in offices and classrooms in University Paul Sabatier in Toulouse. The gas multi-sensor is a microsystem composed by four sensors on a microchip, realized to detect target gases.  

Keywords

Multi-sensors, MOS, Indoor Air Quality, Smart Building, neOCampus

Scientific goals

•    To characterize new nanomaterials (SnO2, CuO, ZnO, WO3 ...) designed by the LCC by using an experimental set-up,

•    To define an operating protocol by trying different operating modes.

Contacts

aymen.sendi@live.fr, menini@laas.fr, pierre.fau@lcc-toulouse.fr, katia.fajerwerg@univ-tlse3.fr, myrtil.kahn@lcc-toulouse.fr, vincent.bley@laplace.univ-tlse.fr

Context Presentation

For several decades, the hot water for heating and domestic hot water has been a major issue because it accounts for approximately 55% of heating needs in France for both commercial and residential buildings. District Heating Systems have been considered as a most economical, efficient, and environmentally friendly solution for providing heating services to the building stock of large cities. This is essentially related to the reason that these systems can use renewable energies.

The Rangueil’s scientific campus, benefits from a superheated water district heating system built in the 1960’s. It has recently been connected to a biomass boiler that covers approximately 82% of the demand. Campus buildings undergoes significant evolutions (renovations, new buildings). The network consists of 4 main loops, 63 primary substations, 97 buildings, 11 km of underground network and cover a surface of 576 000 m². Our aim is to study the performance of this heating system, starting from one heating loop.

Keywords

Rangueil’s District Heating Systems (DHS), Dynamic modeling, MATLAB / Simulink.

Scientific goals

•    The objective is to do a pre-study of the entire system by studying just one loop

•    Identify the parameters influencing the performance of a district heating system

•    Develop on MATLAB / Simulink a model of one loop of the district heating system

•    Analysis of data to identify the impacts of recent evolutions of the network and buildings.

Contacts

khouloud.koteich98@gmail.com, oms@insa-toulouse.fr, moisson@insa-toulouse.fr, francoise.thellier@univ-tlse3.fr

Context Presentation

When it comes to evaluate the quantifiable effects of products or services on the environment, Life Cycle Assessment (LCA) is probably the most efficient and recognized tool. Thanks to a “cradle to grave” approach, LCA identifies and quantifies, throughout the life of products, the physical flows of matter and energy associated with human activities (extraction of raw materials, manufacturing of the product, distribution, use, collection and disposal towards end-of-life). For each of its flows correspond impact indicators which allow to establish the overall potential impact of the system on our environment. With regard to lighting, “smart” technologies have made it possible to improve energy efficiency during use phase and thus greatly limit its impact on the environment.

Before the development of these new technologies, lighting represented 14% of European consumption and 19% of global electricity consumption (2009). Today, the UNEP (United Nations Environment Program) estimates it at 15 % worldwide (2,940 TWh) for 5% of global greenhouse gas emissions (1,150 million tonnes of CO2). However, despite major advances in terms of energy efficiency, many direct or indirect impacts on our environment, our health, well-being and productivity are not considered, and we can no longer neglect these impacts.

It is then necessary to define a new methodology, which will allow the extension of the classic LCA by taking into account several health and social criteria, in particular regarding the potential ”impacts on human” ( blue light and impacts on circadian rhythms); the "impacts on ecosystems" (light pollution, potential impacts on insects and plants population); the several “uses of light” (residential, commercial, public lighting, etc.); or even "social acceptability on and by the user of the system" (security, comfort, working conditions, etc.). The aggregation of these criteria, with a classic life cycle assessment and a life cycle cost analysis (cumulative cost of a product throughout its life cycle), will give a global vision (economic, social and environmental) of the potential impacts of lighting and will helps to offer a decision support tool for establishing coherent and appropriate strategies around the transformation of our lighting systems.

Keywords

Life Cycle Assessment, LED, Lighting Systems, Environmental Impacts

Scientific goal

•    Evaluating the global impacts of lighting technologies, and helping the decision making regarding lighting strategies.

Contact

kevin.bertin@laplace.univ-tlse.fr

Context Presentation

Fog computing has emerged as a strong distributed computation paradigm to support applications with stringent latency requirements. It offers almost ubiquitous computation capacities over a large geographical area. However, Fog systems are highly heterogeneous and dynamic which makes IoT services placement decision quite challenging considering nodes mobility that may decrease the placement decision quality over time.

IoT-Fog Services placement problem needs to be thoroughly investigated to ensure the efficiency of such environments. In this thesis, we consider various parameters such as nodes mobility, energy efficiency and applications Quality of Service (QoS) requirements to propose efficient strategies for IoT services placement in the Fog.

Keywords

Internet of Things, Optimization, Mobility, Fog Computing, QoS, Energy.

Scientific goals

•    Propose efficient approaches for IoT applications (services) placement in the Fog,

•    Analyze their impact on the energy consumption of Fog infrastructures and the Quality of Service (QoS) of  applications.

Contacts

tanissia.djemai@irit.fr, patricia.stolf@irit.fr, monteil@laas.fr, jean-marc.pierson@irit.fr

Context Presentation

We spend around 85% of our time indoors. However, indoor air is 5 to 10 times more polluted than outside air. Based on this observation, various laboratories at the Paul Sabatier - Toulouse III University are focusing part of their research on the development of tools for measuring and improving the Indoor Air Quality.

This is the case of LAAS (Systems Architecture Laboratory), LCC (Coordination Chemistry Laboratory) and LMDC (Materials and Construction Durability Laboratory), which work in collaboration on the development and optimization of MOx gaz sensor (semiconductor metal oxides) to detect different gases in the air (LAAS and LCC), and photocatalytic coatings (LMDC and LCC) to degrade or decrease the concentration of gaseous pollutants.

During this internship, the gases mainly treated are nitrogen oxides (NOx). The aim is to assess the effectiveness of the sensors developed by LAAS and the LCC in detecting variations of NOx concentration and the effectiveness of the coatings in degrading them.

Keywords

neOCampus, Indoor Air Quality, Sensors, Photocatalysis, Nitrogen Oxides, Metal Oxides Semiconductor

Scientific goals

•    To integrate the LAAS / LCC gas sensors into the LMDC test device (Figure 1).

•    To compare the detection capacity of the NOx sensors and analyzer currently present at the LMDC.

•    To carry out abatement tests to test the depollution efficiency of various photocatalytic coatings, in particular based on ZnO and TiO2, and compare the results obtained with the two measurement systems (analyzer and sensors).

Contacts

delaveau@etud.insa-toulouse.fr, hot@insa-toulouse.fr, menini@laas.fr, pierre.fau@lcc-toulouse.fr, katia.fajerwerg@lcc-toulouse.fr

Context Presentation

The 5G evolution is the key driving factor that provides promising support for efficient V2X (Vehicle to everything) communications. Various applications with different requirements have been designed on vehicular networks to improve the driving experience by offering multiple services. In this thesis, we are interested by safety-critical applications and focus on collision avoidance systems between vehicles and pedestrians. This type of applications imposes strict reliability, wide connectivity, and a minimum end to end delay requirements. The high dynamic topology and the different types of communication technologies raise up challenges such as scalability, heterogeneity, and high traffic load to handle.

We have chosen MEC (Multiple Access/Mobile Edge Computing) that offers direct communication exchange between the mobile nodes (vehicles and pedestrians) and the network Infrastructure and used the Network Slicing mechanism to separate the critical vehicular traffic with high priority and strict requirements from other traffic. In our scenario a network congestion risk while vehicles and pedestrians send their BSM and CAM messages to the network infrastructure. Indeed, periodical transmissions could be unnecessary or harmful especially when the network density is high. Therefore, an intelligent scheme should be developed to adapt the transmission frequency of these messages to a network server without overloading the network, while considering the dynamic network state.

The main contribution of our work is to exploit the rich environment Information and analyze those data to take future decisions and to predict the network state to decrease the traffic load and finally to meet the objective requirements. Thus, we argue for the use of advanced Machine Learning techniques to learn from the available network data and take the appropriate action by choosing the best parameters' configuration.

Keywords

Autonomous vehicles, V2X Communications, Resource Allocation, Network Slicing, MEC, Machine Learning

Scientific goal

•    Improve the performance of V2X Communications in the context of critical road safety applications.

Contacts

Chaima.Zoghlami@irit.fr, Rahim.Kacimi@irit.fr, Riadh.Dhaou@irit.fr

Context Presentation

The theme addressed in this PhD concerns autonomous and connected vehicles. It is essential, after making a vehicle more reliable, to study how several connected autonomous vehicles will be able to interact to maximize the behavior of the collective (fluidity, fuel consumption and pollution). The Society of American Automotive Engineers (SAE International) has defined 5 levels of autonomy: level 1, in which the driver performs all maneuvers, at level 5, in which the vehicle is completely autonomous and can do without driver and / or passenger.This PhD concerns both level 4 autonomous vehicles in which the vehicle drives and the supervised driver can regain control of driving, as well as level 5 vehicles (total autonomy of the vehicle).

In this PhD, the aim is to study how each vehicle communicates with its neighboring vehicles and how it behaves / reacts in response to the information provided by its neighborhood. A lock consists in determining the information that is relevant to communicate among all the data recovered from the numerous sensors scattered in the vehicle and effectors. 

Keywords

Intelligent Transport System, Connected and Autonomous Vehicles, Multi-Agent System

Scientific goal

•    Learn how to use any data and which ones are the best to communicate

Contact

guilhem.marcillaud@irit.fr