Artificial Intelligence-related Projects

Jonglera - an agent-based coaching system for stress management

A stress management system based on a multi-agent system platform

STATUS: ongoing
Key people:

• Proj. manager: Helena Lindgren
• AI research manager: Esteban Guerrero
• Field expert manager: Anna Stigsdotter-Neely
• Field expert: Hanna Malmberg

Objective Develop a system to be used by cronical stress patients for managing daily activities. The AI-based system reasons about daily activities suggesting tailored activities for relief stress.
Formal contributions:

• An agent-based coallition method using formal argumentation. (incoming publication)
• A multi-agent based system based on JaCaMo platform
• ...

Android application available now in Google Play Jonglera in Google play
Architecture: Close Project

Intelligent skiing coach

A coaching platform for skiing technique analysis

STATUS: ongoing
Key people:

• Proj. manager and AI research manager: Esteban Guerrero
• Field expert manager: Tor Söderström
• Field expert: Stefan Sandlund

Objective A sensor-based platform to coach skiing atheltes analysing biomechanical features using a hybrid artificial intelligence approach. The AI-based platform gets data from 3D cameras and others sensors, compare in real-time data with expert field rules and other learned techniques, then provides feedback in a game-like user interface .

Contributions:

• E. Guerrero, Nieves, J. C., Sandlund, M., Lindgren, H., & Söderström, T. (2017). Towards automated evaluation of posture quality and its relation to performance in the execution of double poling. In proceedings of the 1st NVC Conference, Performance in snow sports: Translating science into practice. Nationellt vintersportcentrum, NVC Östersund, Sweden, 2017.
• An hybrid AI-based platform to detect activity goal achievement (incoming publication)
• ...

Close Project

Autonomous adaptation of software agents in the support of human activities

Augmented reality + multi-agent systems for assisting human activities

STATUS: ongoing
Key people:

• Proj. manager and AI research manager: Esteban Guerrero
• HCI expert manager: Ming-Hsin Lu
• Field expert: Madeleine Blusi

Objective Explore augmented reality (AR) as a tool for human-computer interaction using a multi-agent platform.
We explore the automatic adaptation of intelligent assistive systems (IAS) in the personalized support of individuals, centered in older adults population. Two main contributions from human-computer interaction (HCI) and artificial intelligence (AI) are developed: 1) a general framework for designing and developing IAS integrating the participation of different users (participatory design); and 2) a compendium of \good practices" from human-computer interaction HCI and AI perspective to improve the adequacy of IAS technology.
Contributions:

• E. Guerrero, Lindgren, H. (2018). Autonomous adaptation of software agents in the support of human activities. To appear in proceedings of the Joint Workshop on Artificial Intelligence in Health. AIH2018 Stockholm, Sweden .
• Intelligent assistive systems using augmented reality for supporting human activities: a home-based medication distribution study in older adults (incoming publication)
• ...

Close Project

Towards a trusted intelligent coach

An intelligent tutoring system to provide biomechanic accurate feedback about performance

STATUS: ongoing
Key people:

• Proj. manager and AI research manager: Esteban Guerrero
• AI researcher: Juan Carlos Nieves
• HCI experts: Karolina Thorsén and Anna Lindström

Objective Investigate trust in intelligent support systems.
We explore the automatic adaptation of intelligent assistive systems (IAS) in the personalized support of individuals, centered in older adults population. Two main contributions from human-computer interaction (HCI) and artificial intelligence (AI) are developed: 1) a general framework for designing and developing IAS integrating the participation of different users (participatory design); and 2) a compendium of \good practices" from human-computer interaction HCI and AI perspective to improve the adequacy of IAS technology.
Contributions:

• Trust in human-computer relationships. Do cross country skiers have trust towards a physical intelligent tutoring system as an accurate feedback on performance? Karolina Thorsén and Anna Lindström. Bachelor’s Thesis Programme in Cognitive Science Umeå University. Advisor: Juan Carlos Nieves, Department of Computing Science Umeå University.

Close Project

Mobile sports science platform

A mobile platform for tracking and evaluate sports activities

STATUS: closed
Key people:

• Proj. manager CS: Helena Lindgren
• Proj. manager SS: Tor Söderström
• AI research manager: Esteban Guerrero
• Field expert: Stefan Sandlund

Objective Develop a mobile platform to track and evaluate physical exercises.
Close Project

Balansera

A mobile platform for evaluating balance and strength in older adults

STATUS: closed
Key people:

• Proj. and AI research manager CS:Esteban Guerrero
• Proj. manager physiotherapy: Marlene Sandlund
• AI advisor: Helena Lindgren
• AI advisor: Juan Carlos Nieves

Objective We explore a novel approach based on argumentation theory for quantifying the evaluation of physical activities from the perspective of rational agents, by resembling the kind of assessment reasoning performed by clinicians: (1) gathering data through observations; (2) handling ambiguous and uncertain observation information; (3) generating current function status hypothesis; (4) deduce an explanatory outcome of explanation; and (5) retracting the explanation under new evidence.
Contributions

• Guerrero, E., Nieves, J. C., Sandlund, M., & Lindgren, H. (2018). Activity qualifiers using an argument-based construction. Knowledge and Information Systems, 54(3), 633-658.

Close Project

Semantic-based Construction of Arguments: an Answer Set Programming Approach

We introduce an approach to building arguments, taking as input an extended logic program and obtaining as result a set of consistent and rational arguments.

STATUS: closed
Key people:

• AI researcher: Esteban Guerrero
• AI researcher: Esteban Guerrero

Objective We introduce an approach to building arguments, taking as input an extended logic program and obtaining as result a set of consistent and rational arguments. Three major contributions are introduced in this paper:

• We establish the concept of argument by considering a semantic interpretation of the argument supports. This notion allows us to unequivocally identify arguments with stratified programs as support, even when the input for the argument engine is a non-stratified program. We have observed that our notation of argument can be based on any logic programming semantics which coincides with stable model semantics in the class of stratified logic progams.
• We introduce a set of rationality postulates for argument-based systems under extended logic programs. These postulates judge the consistency of the underlying formalisms for building arguments. These postulates are formulated in terms of argument conclusions, considering at the same time a skeptical output of the system. We introduce these concepts redefining quality criteria introduced by Gelfond-Lifschitz in terms of normal logic programs and a definition of closure of a set of clauses that consider the well-known Gelfond-Lifschitz reduction
• By taking advantage of the computational properties of WFS, we develop an implementation which, to our knowledge, is the first argumentation engine for extended logic programs. Therefore, our engine can serve as a starting point for other argument-based systems in a wide application spectrum, such as decision-making frameworks and context-based argument reasoning.

Close Project

Human-Robot interaction - Assistive robots

Human reactions are studied from a formal perspective to allow a robot interact and support a human

STATUS: open
Key people:

• AI researcher: Esteban Guerrero
• AI researcher: Esteban Guerrero

Download the code generated for this scenario here: Close Project