AmIciTy |

| Ambient Intelligence for the Collaborative Integration of Your Tasks


Our Aim:

to create a software infrastructure for Ambient Intelligence (AmI) applications, that handles context at its constructive level, that works naturally with the user's tasks and activities, and that manifests the robustness and reliability that is required for future ambient systems.

Directions

tATAmI-2 | towards Agent Technologies for AmI & IoT
agent-based platform for privacy-aware AmI and IoT applications

  • a powerful and flexible platform for dynamic multi-agent systems
  • control and visualize the system from a single machine
  • use the adequate means of communication between the adequate machines
  • behavior of agents is described by the simple-to-understand agent-oriented-programming language S-CLAIM

Positions
  • MAS-Enabled IoT for Common Devices | positions for 2 master students
  • Privacy-Awareness in IoT through Smart Data Management | positions for 2 master students

read more
AmIciTy:Ctx | AmIciTy Context reasoning
represent user's context with graphs, detect situation with patterns

  • context graph: use a graph of concepts and relations to represent the user's current context
  • context patterns: use graph patterns (graphs with generic nodes and edges) to represent potentially interesting situations
  • match context patterns against the context graph to detect situation and reason on appropriate action
  • uses AmIciTy:Graph for graph tooling

Position
  • Distributed Context Learning on Smart Devices | position for 1 master or diploma student

read more
AmIciTy:Task | AmIciTy user Task integration
task-based environment for assisting users in their daily lives

  • uses software agents to manage the user's tasks and create context-aware, pro-active notifications that come at the right time, in the right manner
  • deployment on PC and Android platforms
  • integrates existing components for context gathering and management

read more
AmIciTy:incubator | technologies for AmILab scenarios
technologies to support deployment of AmI scenarios in AmILab

  • test and deploy various technologies that can be used in detecting the user's activity
  • the focus is on sensing information about the physical, that can be intelligently aggregated to detect activity
  • the sensors used are the sensors of Android devices and Kinect sensors in AmILab

Positions
  • Ubiquitous Applications in AmILab | positions for 3 diploma students

read more
AmIciTy:Graph | AmIciTy graph library
lightweight library for context-oriented graphs and patterns

  • supports the management and manipulation of context graphs in patterns (see AmIciTy:Ctx)
  • uses a purpose-built algorithm for graph matching, more efficient than classic graph-matching algorithms

read more
AmIciTy:Sim | AmIciTy Simulation tools
rapidly simulate a large number of agents on a local machine

  • start a simulation of a large number of agents with one click
  • control simulation speed or execute step-by-step
  • use simulation description files for all simulation data, including agent positions, trajectories, initial knowledge, etc.
  • offers monitoring tools that aggregate data over all agents or groups of agents
  • view individual information about agents selected dynamically at execution time
  • repeat any simulation, including all random events

read more
Project 3A | Agent-based Attendant Assistance
smart assistance for conference attendants

  • using the Envived framework for management of smart spaces and artifacts.
  • using tATAmI agents
  • using context graphs for situation detection

read more
© AI-MAS group 2017

tATAmI-2 | towards Agent Technologies for AmI & IoT

agent-based platform for privacy-aware AmI and IoT applications4 open positions7 past projects20 collaborators

Features



Positions

MAS-Enabled IoT for Common Devices
positions for 2 master students

objectives

While commercial IoT ecosystems exists, they are difficult to acquire and to inter-operate. On the other hand, smartphones are full of sensors and everyone has them.

The goal of this research is to create an open IoT ecosystem in which common smart devices such as phones and tablets are the main actors, being able to get information from the environment, communicate it to each other, and making it available to the users, all securely and privacy-aware. As an appropriate enabler for the subject, the system will be underpinned by software agents.

After the first year of research, the expected result is a state-of-the art multi-agent system that manages the information in a context- and privacy-aware manner, all while being flexible and lightweight. During the second year, an application will be built that uses machine learning to learn from the acquired data and detect unusual situations, such as medical emergencies.


references
[1] Gubbi, J., Buyya, R., Marusic, S., & Palaniswami, M. (2013). 'Internet of Things (IoT): A vision, architectural elements, and future directions', Future Generation Computer Systems, 29(7), 1645-1660.
[2] Swan, M. (2012). 'Sensor mania! the internet of things, wearable computing, objective metrics, and the quantified self 2.0', Journal of Sensor and Actuator Networks, 1(3), 217-253.
[3] Olaru, Andrei. 'tATAmI-2--A Flexible Framework for Modular Agents.' 2015 20th International Conference on Control Systems and Computer Science. IEEE, 2015.
Privacy-Awareness in IoT through Smart Data Management
positions for 2 master students

objectives

In the Internet of Things devices are accessible via the network from anywhere in the world. An important aspect is to make data and control available the the authorized users without breaking privacy.

The goal of this research is to develop mechanisms and algorithms for data to be managed so that, while mobile through the network, is made available only to the intended / authorized users. At the same time, these mechanisms should be lightweight enough to work on resource-constrained devices.

The expected result is to have the developed mechanisms integrated into a MAS-enabled IoT ecosystem, potentially with help from the features offered by software agents.


references
[1] O'Hara, Kieron. 'Privacy and the internet of things.' (2014).
[2] Samani, Afshan, Hamada H. Ghenniwa, and Abdulmutalib Wahaishi. 'Privacy in Internet of Things: A Model and Protection Framework.' Procedia Computer Science 52 (2015): 606-613.
[3] Perera, Charith, et al. 'Privacy of big data in the internet of things era.' IEEE IT Special Issue Internet of Anything 6 (2015).
[4] Olaru, Andrei, Adina Magda Florea, and Amal El Fallah Seghrouchni. 'Using Context‐aware Multi‐agent Systems for Robust Smart City Infrastructure.' Enablers for Smart Cities: 187-206.


Past projects

Ao Dai: Agent-Oriented Design for Ambient Intelligence    | coordinated together with Amal El Fallah Seghrouchni
research project with Diego Salomone-Bruno and Thi Thuy Nga Nguyen

Done:
  • a demonstrative application dealing with navigation assistance through the space of a building floor;
  • agent hierarchies are used to describe a hierarchy of context elements (such as spatial elements).
Android Platform for Pro-Con Discussion
diploma project with Iulia Moscalenco and Miruna Popescu

Done:
  • the platform was ported to Android, tested and improved using a distributed multi-agent application.
tATAmI / Ao Dai Demo    | coordinated together with Amal El Fallah Seghrouchni and Kenji Tei
research project with Kenji Tei, Susumu Toriumi, Valentina Baljak, Yusaku Kimura, Ryo Shimizu, Amal El Fallah Seghrouchni, Thi Thuy Nga Nguyen, Marius Tudor Benea and Cédric Herpson

Done:
  • successful demonstration of the tATAmI platform using S-CLAIM agents, deployed in the NII Smart Room;
  • the platform successfully interoperated with web services providing location information for the users.
Profiling and Optimization for Android Applications on the tATAmI platform
master research project with Jaber Hassoun

Done:
  • some tATAmI agent applications.
Agent-Based Android Application for Conference Participants
diploma project with Alex Tomescu and Emma Sevastian

Done:
  • a client-server setup based on Envived that allows communication through HTTP long-polling, and also is able to integrate agents (Alex);
  • adaptation of the tATAmI platform to work with context patterns for knowledge representation and integrate the new, component based architecture (Emma).
Deployment of the tATAmI Platform in AmILab -- AmILab interface
internship project with Claudiu Mihai Toma

Done:
  • implementation of an agent component that allows tATAmI agents to interact with AmILab services;
  • implementation of the Follow-me application (v1).
Modern Methods for Communication, Mobility and Portability for the tATAmI Platform
master project with Cosmin Mihai

Done:
  • implementation of WebSocket-based communication for tATAmI-2 agents;
  • implementation of agent mobility for tATAmI-2 agents communicating through WebSockets;
  • help in modularizing tATAmI-2.
© AI-MAS group 2017

AmIciTy:Ctx | AmIciTy Context reasoning

represent user's context with graphs, detect situation with patterns1 open position4 past projects4 collaborators

Features



Position

Distributed Context Learning on Smart Devices
position for 1 master or diploma student

objectives

While machine learning works great when computing resources abound, a privacy-protecting distributed implementation of context-aware, intelligent behavior requires learning on resource-constrained devices.

The goal of this research is to develop means to learn from the user's activity on smart devices only with the help of local resources.

The expected result is to be able to predict the activities of the user, using sensors on the device and also information available from local apps, in order to better assist the user in those activities and anticipate potential unwanted situations.


references
[1] Krishnan, Narayanan C., and Diane J. Cook. 'Activity recognition on streaming sensor data.' Pervasive and mobile computing 10 (2014): 138-154.
[2] Cook, Diane J. 'Learning setting-generalized activity models for smart spaces.' IEEE intelligent systems 2010.99 (2010): 1.
[3] Olaru, Andrei, and Adina Magda Florea. 'A Platform for Matching Context in Real Time.' International Conference on Hybrid Artificial Intelligence Systems. Springer International Publishing, 2015.


Past projects

Graph Matching for Context Recognition
diploma project with Adrian Dobrescu

Done:
  • a thorough study of the state of the art in graph matching algorithms;
  • adaptation of several graph matching algorithms to use with context graphs and patterns;
  • a software platform for the comparison of algorithms, proving that the QuickMatcher algorithm is the most appropriate for the problem at hand.
Using Context Graphs to Help Users in Their Daily Activities
master research project with Adrian Dobrescu

done
  • scenarios for AmI that can be implemented using context graphs and patterns;
  • analysis of the challenges that arise when using context graphs and patterns for describing real-life scenarios.
Using Context Graphs to Help Users in Their Daily Activities
diploma project with Cătălin Badea

Done:
  • an application has been developed that is based on Jung and adequate handlers to allow the display and editing of context graphs and patterns;
  • matches are automatically detected as the graphs are edited.
Generating Context Graphs from Sensor Data using Human Activity Recognition Models
master project with Cătălin Badea

Done:
  • a study into how machine learning could be applied to context graphs in order to obtain context patterns.
  • use of various HAR models to create context graphs starting from raw sensor data.
© AI-MAS group 2017

AmIciTy:Task | AmIciTy user Task integration

task-based environment for assisting users in their daily lives1 past project2 collaborators

Features



Past project

Ambient Intelligence for Task Integration
diploma project with Cătălin Drăgoiu-Prună and Vlad Herescu

Done:
  • an Android application that displays and manages the user's tasks, together with details that link the task to files, contacts and online services, displayed in a compact, comprehensive manner (Cătălin);
  • an Android application that relies on various details about the user's tasks to schedule user's activities, with real-time updates and context detection (Vlad).
© AI-MAS group 2017

AmIciTy:incubator | technologies for AmILab scenarios

technologies to support deployment of AmI scenarios in AmILab3 open positions1 past project3 collaborators

Features



Positions

Ubiquitous Applications in AmILab
positions for 3 diploma students

objectives

The goal of the project is to build a cross-platform application that is able to move from one device to another as instructed by the user. On his/her personal device, the user will slide with the finger in the direction of a PC or gesture in the direction of a PC and the application will move to the designated machine. On all machines the application will feature similar interfaces.


todo

The project will be split into three sub-projects:

  • Android and PC application back-end
  • using a Kinect device to determine the target machine
  • cross platform visual interfaces based on a uniform specification

why

The completion of the project will get us closer to realizing the ideal deployment of AmI in which applications are independent of, and can move between devices. It will also be a good demonstrative application for the capabilities of the AmILab.



Past project

Implementing AmI Scenarios with Android Devices
internship project with Cristian Neagoe, Cristian Grigoras and Vlad Herescu

Done:
  • a method to set the phone's ringer volume so that the ringing is of an adequate volume with respect to the ambient sound;
  • a method to detect the user's location using the wireless networks available in the current location;
  • a method to keep the user's working files on the user's currently used device;
  • a method to detect the user's need for assistance in working on a project.
© AI-MAS group 2017

AmIciTy:Graph | AmIciTy graph library

lightweight library for context-oriented graphs and patterns2 past projects2 collaborators

Features



Past projects

Visualization of Context Graphs - JUNG and Zest
master research project with Nihal Ablachim

Done:
  • a thorough study in the two libraries for graph rendering;
  • analysis of the advantages and disadvantages of each library in the context of the given problem.
Using Context Graphs to Help Users in Their Daily Activities
diploma project with Cătălin Badea

Done:
  • an application has been developed that is based on Jung and adequate handlers to allow the display and editing of context graphs and patterns;
  • matches are automatically detected as the graphs are edited.
© AI-MAS group 2017

AmIciTy:Sim | AmIciTy Simulation tools

rapidly simulate a large number of agents on a local machine3 past projects4 collaborators

Features



Past projects

The AmIciTy:Mi MAS Simulation Platform
research project with Cristian Gratie

Done:
  • a plaform for simulating up to several thousand agents interacting in a complex manner;
  • agents use individual message queues for communication;
  • the simulation follows a scenario that is completely specified in an XML file;
  • powerful visualization tools have been developed, such as individual agent visualization for all agents, grid views of individual agent features, and 2D and 3D graphs for various aggregated measures.
Improvement of AmIciTy:Mi agent balance
internship project with Sofia Neață

Done:
  • new measurements for agent's balance have been integrated and analysed.
Improvement of AmIciTy:Mi as a Powerful MAS Simulation Platform
internship project with Guillaume Masson and Alexandre Hoquard

Done:
  • improvement of AmiciTy:Mi to support advanced description of agent sets, moving agents, and various agent behaviors.
© AI-MAS group 2017

Project 3A | Agent-based Attendant Assistance

Coordinated by Alex Soricismart assistance for conference attendants3 past projects9 collaborators

Features



Past projects

Android Smart Presentation Feedback    | coordinated together with Tudor Berariu
diploma project with George-Cristian Stoica, Dragoș Dincă and Anca Pîrvan

Done:
  • a very useful and interesting application has been created;
  • it uses a client-server architecture to manage the slides and user feedback;
  • it uses PDF rendering to allow the user to navigate through the slides independently of the currently presented slide;
  • it parses the PDF file with the slides, allowing the user to select text or images to comment on.
Semantic Question Clustering    | coordinated together with Tudor Berariu
master research project with Cristina Groapă and Maria-Cătălina Mocanu

Done:
  • an algorithm for semantic clustering of short text was obtained.
Agent-Based Android Application for Conference Participants
diploma project with Alex Tomescu and Emma Sevastian

Done:
  • a client-server setup based on Envived that allows communication through HTTP long-polling, and also is able to integrate agents (Alex);
  • adaptation of the tATAmI platform to work with context patterns for knowledge representation and integrate the new, component based architecture (Emma).
© AI-MAS group 2017