Posts from category "réalisation"

SANDMAN: Anomaly Detection in a Data Stream Issued from Smart Buildings

IRIT and  LMDC , Toulouse University


Anomaly detection, multi-agent system, smart buildings, energy management, data stream

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 existing software can 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. To improve the energy management, an innovative anomaly detection system should aim at analyzing raw data, detect any kind of anomalies (point, contextual, collective) in an open environment, at large scale. The developed AI system is called SANDMAN (semi-Supervised ANomaly Detection with Multi-AgeNt systems). The system is semi-supervised by an expert of the field who confirms or overturns the feedback of SANDMAN. 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 system. The results show the robustness of the AI regarding the detection of noisy data, of different types of anomalies, and the scaling.  

Scientific goal

Anomalies detection in smart buildings streaming data by a semi-supervised multi-agent system.


ReCoVAC: conditions for REtaking COntrol by self-obserVAtion of situations within a Connected autonomous vehicle

IRIT - CLLE, Toulouse University


Autonomous vehicles, self-adaptive multi-agent systems, driving control recovery.

Connected autonomous vehicles of level 3, called "conditioned automation", are vehicles in which the human driver delegates driving control in specific situations. During these situations, it may be necessary for the human to regain control of the driving activity. The main objective of this thesis is to develop a supervision system adapted to each driver, by integrating human factors, to allow a safe and efficient transition of two-way control between the human and the autonomous vehicle. For this, the system must identify, by self-observation and in real-time, situations in which the current driver will no longer be able to ensure driving. He must also provide a context for assessing the criticality of the situation as quickly as possible to anticipate and react to it as best as possible. The driving context is composed of indicators that characterize the elements that describe part of the driving process: human, vehicle, and environment. The system is based on self-adaptive multi-agentlearning systems.

Figure 1: Dynamic Learning using self-adaptive multi-agent system

Scientific goals

- Dynamic learning using multi-agent systems

- Generic approach to supervise the activity of a system

- Study the impact of the factors describing the different elements present in the system context on the ability of the system to converge towards a solution.

- Insure the acceptability of the system by human driver


Interaction between patient and exoskeleton for cerebral palsy in children



Cerebral palsy, gait analysis, robotic rehabilitation

Children with cerebral palsy (CP) have an altered locomotion with abnormal gait patterns such as crouch gait, for which the lower limb joints display too much flexion. An exoskeleton adapted to the needs of the patient could improve their walking abilities. An experimental session of gait analysis was conducted to analyze and compare precisely the gait of two twin sisters, one with CP, with the aim of developing an exoskeleton to treat CP. The healthy child showed no differences with the standard results except for the frontal plane kinematics of the knee and hip. The study of the muscular activity revealed an over-activation of all the muscles of the child with CP. The kinematic results showed too much flexion in the sagittal plane for the hip, knee and ankle, as well as asymmetric deviations in the frontal and transverse planes. The kinematics of the pelvis and lumbar region were also altered. Although the data was scarce due to experimental difficulties, the study of joint moments and powers showed altered profiles compared to the results of the healthy child. These findings will allow to run simulations of an exoskeleton and to develop a control strategy adapted to this particular child. 

Figure: General configuration of the experimental configuration 

Figure: Kinematics of the right hip, knee and ankle in the sagittal plane. The red results stand for the healthy child and the blue results for the CP child


ECONECT: Developing connected environmental sentinel systems to better understand the degradation of rivers, the decline of bees and birds

Context Presentation

The ECONECT project began in early 2020, with the objective to develop a communication infrastructure allowing the remote monitoring of autonomous, connected, and versatile systems to measure the responses of bioindicator organisms to chemical contamination, habitat degradation and global warming.

Three sentinel systems are considered:(1) the connected hive, allowing to monitor the dynamics of bee colonies (colony mass, temperature and location of the bee cluster, foraging traffic, etc.) and the cognitive capacities of bees; (2) the connected bird-feeder to submit individually monitored tits to behavioral tests to assess their cognitive abilities; (3) the aquacosm, a floating enclosure allowing the measurement of eco-markers in an aquatic environment (growth dynamics of phototrophic biofilms, relative importance of autotrophic and heterotrophic processes within the ecosystem ...).

In 2022, a network of 12 sentinel stations will be deployed in the Zone Atelier Pyrénées-Garonne (PYGAR). Each station will be characterized by a spatial analysis of land use and the quality of habitats and by the measurement of concentrations of chemical contaminants (trace metal elements, PAHs, pesticides) in different compartments of the environment. Participatory science protocols will be used to supplement the available data set and to assist in the interpretation of observed trends, while providing environmental education opportunities for the public.

schema (EN) - Arnaud Elger


Environmental sensor; Bioindicator; Animal cognition; Chemical status; Landscape integrity; Artificial intelligence

Scientific goals

•    to design a communicating infrastructure to collect data from different sensors in the field;

•    to develop automated tools for the real-time analysis of collected data, for extracting their ecological significance;

•    to examine the relevance of our sentinel systems to assess the quality of the environment, particularly in terms of chemical status and landscape integrity.



Dynamic Collection of Cooperative Awareness Messages for Collision Avoidance with Vulnerable Road Users

IRIT – Toulouse University                                         


V2X Communication architectures, vulnerable road users safety, cooperative awareness messages.

With the evolution of Intelligent Transportation System (ITS), vehicles are capable of performing intelligent decisions and cooperative communications with other road users to exchange data and expand their environmental awareness. This communication is introduced as vehicle-to-everything (V2X) where vehicles exchange cooperative awareness messages « CAM » that includes important information (eg: position, speed, heading angle, vehicle type…). The generation rules of the CAM messages are defined by the ETSI standard and they are implemented in the facilities layer of each vehicle. The frequency of sending CAMs is between 10 and 1 Hz. However, we found out that following the standard, vehicles must generate a CAM if any change in their behavior is detected. This can lead to overloading the network if the number of road users is high, and might not be necessary if a vehicle has no.

Scientific goal

The aim of this work is to design a novel  neighboring mechanism to enhance the current version of the standard. Through a centralized server with a global vision of the network, the vehicles will be able to efficiently adapt the frequency of sending CAM messages using the information of surrounding neighbors received from the edge-server. Our mechanism is extended to support vulnerable road users (pedestrians, cyclists…) where reducing their transmission frequency undoubtedly helps in saving energy on their connected devices.


Déploiement d’un système de suivi des déplacements et de la pollution sur vélos pour la mise à disposition sécurisée de données atmosphériques.

Cette thèse s’inscrit dans le cadre du projet CLUE : Cycle-based Laboratory for Urban Evolution. Ce projet scientifique vise à équiper une partie des vélos évoluant dans le campus et dans Toulouse d’un ensemble de capteurs afin d’étudier les déplacements des usagers, mais aussi de profiter du réseau de capteurs mobiles ainsi déployé pour collecter des informations sur la pollution atmosphérique sur le campus et plus largement dans la ville.

Objectifs scientifiques

Plus particulièrement, l’objectif de cette thèse s’articule autour des points suivants :

• La collecte d’un jeu de données dans Toulouse (données de mobilité et mesure de polluants atmosphérique) - inexistant à ce jour - et sa mise à disposition.

• Le déploiement d’un noeud de collecte sans fil des informations, grâce à la technologie LoRa (longue portée, basse consommation d’énergie), et la sécurisation des données sensibles (localisation).

• La présentation des données aux différents acteurs/utilisateurs (chercheurs en aérologie, cyclistes, personnes en charge de l’aménagement du campus) :

– Système de contrôle d’accès aux données multi-roles

– Compromis protection de la vie privée/utilisabilité des données

• L’intégration de différents capteurs existants et tests en environnement réel, en particulier pour les capteurs “black carbon” et oxydes d’azote

• Le raffinement et la validation in situ des modèles de diffusion de polluants utilisés en aérologie


- Christophe Bertero (LAAS) : 

- Jean-François Léon (LA) :

- Matthieu Roy (LAAS) :

- Gilles Tredan (LAAS) :


Model Self-Calibration using Self-Adaptive Multi-Agent System

Context Presentation

The purpose of this project is to propose a cooperative agent model, based on the self-adaptive multi-agent system theory (AMAS), allowing an efficient and fast exploration of the parameter space, autonomously and automatically. This exploration should allow a continuous readjustment of the simulation until convergence, improving the control of the macro-level over the micro-level.

On an application standpoint, the purpose of this project is to produce a realistic traffic that satisfies the best a set of objectives and constraints at both micro and macro levels. This traffic should also allow interaction with humans and adapt to events that could occur in the virtual environment. 

CALICOBA_simple - Darmo


Self-adaptive Multi-agent Systems, Self-Calibration, Multi-Agent Simulation

Scientific goals

1. Enrich the AMAS theory with general learning mechanisms andstrengthen the coupling between micro and macro levels.

2. Propose a new generic calibration method of models.

3. Enrich GAMA tools


A Smart Clean Garden for Toulouse 3 University

Context Presentation

This project is built in cooperation with Epurtek factory and INRAe who displays planted filters in Occitany, France and other countries. The actual network extends with the help of the regional GIS ‘EAU Toulouse and the national group of water Re-Use from Aqua-Valley and CapEnergy platforms. Major research and innovation issues are: 

1. To obtain and feed a pluri-topic data base from environmental sensors settled down into the filters for in-field, providing a spatially and temporally improved datasets that allows the implementation of deterministic models and open the black boxes of the filters sediment and soil. These models of transport and reaction (ex MIN3P) will describe the water and pollutants flows through the filters. This task started from cooperation between the research labs (UMR ECOLAB) at Toulouse, the university of La Rochelle, and UMR ECO&SOL from Montpellier.

 2. To demonstrate how the invertebrate addition in the filters will improve the purification efficiency in agreement with the biodiversity influence that governs natural ecosystem functioning. Research hypotheses will be tested in filter replicates at the level of lab microcosms, in campus pilots and across-campus comparisons.

 3. The assessment of biodiversity effects on filter benefits and services in terms of water quality providing, water quantity resource management, energy cost, recreational area for campus users, teaching and innovation supports.

 4. To provide a new generation filter that matches with the requirement of the recent European policies about water reuse quality and may be recognized as a new technology for local water cycle in our territories

Diapositive1 - magali Gerino


Planted Filter, Water, Biodiversity

Scientific goals

This garden provides the potential to become an outdoor living lab as a demonstrator of sustainable and low carbon solution for wastewater treatment and recycling though nature-based solution. The main challenge is to demonstrate the advantages of having a biodiverse and smart clean garden on a campus or a smart city, in terms of environment, economy, society (quality of life) and energy, by comparison with the classic filters. This implies an adaptative management of the water resource in cooperation with other research laboratories and stakeholders, as public water managers and private factories that wish to favor the ecological and energetic transition in this field.


Bio-inspired connected filter for campus water

LEFE, SGE, IRIT, Toulouse University / IMFT, La Rochelle University / PME Epurteck


Living Lab, water, filter, biodiversity, tomography

A Water oriented Living lab on the campus gets applied and fundamental research components with the main goal to reduce the surface area of the “regular” planted filters by making them more performing toward filtration with the involvement of an increased biodiversity. This demonstrator makes part of the LL implementation in the Interreg SUDOE Tr@nsnet Project. A cooperation between UT3 Direction du Patrimoine, SGE, UMR IRIT, UMR IMFT, U La Rochelle, PME Epurteck will lead to a bioinpired filter located next to IRIT to treat waste water of the A1 building. This biotech with enhanced biodiversity and soil metabolism for organic mater biodegradation, will increase the green area of the campus, will help at the air temperature regulation , and prevent of any smelt and musquitos for the neighbourhood. The earth worms (are ecological engineers that dig biostructure networks in the soil, that may largely influence the water parameters when flowing through this soil. The current tested research hypotheses is: « how does a burrow network buried in the macroporous substrates of soils influence the water infiltration capacities ? » This is run through the cooperation between research group in Physic “MacroPorous media and Biology” of UMR IMFT and the FERMAT X-ray tomography for images of 3D gallery networks of worms burrow, and modeling of water infiltration water flux in porous media; and the ecological team Bioref (Biodiversity, biological networks and Fluxes in aquatic and terrestrial ecosystems) of UMR Laboratory of Functional Ecology and Environment

Constant-head permeameter