The Cognitive Robotic Systems Laboratory at AASS

(Previously "Mobile Robotics Lab")

Our Collaborative Projects

Please note: This page has not been updated in a while. If you are interested in the current projects of our lab, you can either wait that we all get a bit of quiet time and update it, or contact directly the Lab Leader.

The research themes at our laboratory are implemented in a number of projects, that tightly combine theoretical investigation, empirical evaluation, and concrete applications. Below are the currently on-going projects. (Please contact the lab director for information about past projects.) Each project is typically associated to one research theme, altough a few of them bridge two or more themes. Most projects are co-financed by Örebro University.

GeRT: Generalizing Robot Manipulation Tasks

	Research theme:	Cognitive robots
	Project staff:	Alessandro Saffiotti (P.I.) Lars Karlsson
	Partners:	DLR, Germany (project coordinator) University of Birmingham, UK Max-Planck Institute, Germany
	Main funder:	European Commission FP7 ICT programme (504,000 EUR for 2010-2013)
	Project website:	GeRT

Synopsis: In order to work naturally in human environments such as offices and homes, robots of the future will need to be much more flexible and robust in the face of novelty than those of today. Humans cope so seamlessly with novel objects that we do not think of grasping a new cup, or screwing the lid off a jar we haven't seen before as challenging. But this kind of everyday novelty in manipulation tasks is hard for a robot. Currently the most advanced robots can perform a task such as making a drink, which involves grasping, pouring, and twisting off a cap from a jar. But the rules for how to pick up every single object must be programmed. So the ability to manipulate fifty different objects means writing fifty different programs. Also when the robot encounters an object it hasn't seen before, it can't grasp the object. Even worse than this is the fact that if one object in a task changes then the program for the whole task may need to be rewritten. This is because how we manipulate an object depends on the task, and the other objects involved. In GeRT we will develop new methods to cope with novelty in manipulation tasks. Our approach is to take a small set of existing robot programs, for a certain robot manipulation task, such as serving a drink and to give the robot the ability to adapt them to a novel version of the task. These programs constitute a database of prototypes representing that class of task. When confronted with a novel instance of the same task the robot needs to establishing appropriate correspondences between objects and actions in the prototypes and their counterparts in the novel scenario. In this way the robot can solve a task that is physically substantially different but similar at an abstract level. The project's results will be demonstrated on the DLR platform Justin. The main role of the AASS Mobile Robotics Lab is to develop hybrid planning techniques able to: (1) generalising the high-level behaviour in the robot programs, and (2) generating new high-level behaviours for novel but analog tasks.

ExCITE - Enabling SoCial Interaction Through Embodiment

	Research theme:	Robots for humans
	Project staff:	Silvia Coradeschi (project coordinator) Amy Loutfi Federico Pecora Annica Kristoffersson
	Partners:	Giraff AB, Sweden ISTC National Research Council, Italy RatioConsulta SpA, Italy University of Malaga, Spain Örebro City Council, Sweden
	Main funder:	European Commission FP7 AAL programme (332,000 EUR for 2010-2012)

Synopsis: The main objective of ExCITE is to evaluate user requirements of social interaction that enables embodiment through robotic telepresence. This evaluation is performed in situ, on a PanEuropean scale and with a longitudinal perspective. An existing prototype is deployed to the targeted end users, and is refined by tightly involving the users in the development cycles of the prototype throughout the project. The prototype used is called the Giraffe system and consists of a screen and web camera mounted on a simple robotic base that can be teleoperated.

DHRS-CIM: Distributed Human-Robot System for Chemical Incident Management

	Research themes:	Cognitive robots, Robot ecologies
	Project staff:	Alessandro Saffiotti (P.I.) Marcello Cirillo Federico Pecora Amy Loutfi Lars Karlsson Kevin LeBlanc
	Main funders:	European Commission FP7 IAPP program (553,000 EUR for 2008-2012)

Synopsis: DHRS-CIM is a EC FP7 project under the scheme "Industry-Academia Pathways and Partnerships" (IAPP). The overall objective of the project is to combine forces of Europe's leading industrial and academic organizations in the field of Disaster Management, in order to deliver an intelligent system which can be deployed to detect (potential) chemical incidents and help humans to take informed decisions to manage chemical incidents. The research within this project deals with human- and artificial sensing and sense-making. An essential contribution is the alignment and integration of both artificial- and human-system research: the artificial systems and humans together become an intelligent collaborative system. The consortium consists of industrial and academic partners from the Netherlands, United Kingdom, Belgium, Portugal, and Sweden. AASS participates in this project through the cooperation of two labs: the Mobile Robotics Lab, and the Learning Systems Lab. The main contributions of AASS are in the areas of artificial olfaction, of perception planning, and of integration of sensor data and symbolic information.

Anchoring in a Symbiotic Robotic System

	Research theme:	Cognitive robots
	Project staff:	Silvia Coradeschi (P.I.) Marios Daoutis Jonas Ullberg Amy Loutfi Federico Pecora Alessandro Saffiotti
	Main funders:	Vetenskapsrådet, Swedish Research Council (2 Million SEK for 2009-2011) Örebro University (internal faculty funds, ongoing)

Synopsis: In this project we investigate the problem of anchoring, that is connecting symbols to sensor data corresponding to objects, in a system where humans, robots, and pervasive devices are cooperating. The main goal of this project is to enable human participation in the anchoring process through human-like forms of communication. This will be done by integrating knowledge representation and reasoning techniques to an anchoring framework. In our previous work, we have examined the problem of perceptual anchoring in single and distributed robotic systems considering multiple and heterogeneous devices. The key elements explored in this project will be the augmentation of the symbolic levels of the current anchoring framework by considering a shared ontology and common grounding between the actors in the system. Strong emphasis is placed on the human robot interaction,in particular: generating object descriptions that are meaningful to the human and interpreting requests about objects. Our driving motivation is the symbiotic system where human, robot, and environment coexist and in particular we examine the system in the context of elderly living. The overall significance of the project lies in the integration of state-of-the-art techniques in knowledge representation and reasoning to robotic platforms in order to shift the application of robotics from the industrial venue to home-like environments.

Remote - Robotic tElepresence fOr disTributed CarE

	Research theme:	Robots for humans
	Project staff:	Silvia Coradeschi (P.I.) Annica Kristoffersson Amy Loutfi Federico Pecora
	Main funders:	RobotDalen, on funds from Vinnova and the EU (300,000 SEK for 2009-2010) NovaMedTech, on funds from the EU (150,000 SEK for 2009-2010)

Synopsis: The main objective of this project is to evaluate user requirements of distributed care that enables embodiment through robotic telepresence. This evaluation is performed in situ, and focuses on the healthcare professionals perspective in terms of the effectiveness and benefit to alternative forms of providing patient care. A prototype robot which enables telepresence, called the Giraffe, is deployed to the targeted end-users, and is refined by tightly involving the feedback in the development cycles of the prototype throughout the project. Specific focus is placed on to best augment the prototype in order to satisfy the requirement of providing distributed care to patients located in the geographical regions of $Ouml;rebro and Västerås. The project is closely coupled to the AAL initiative ExCITE.

Odour Recognition in Intelligent Systems

	Research theme:	Artificial olfaction
	Project staff:	Amy Loutfi (P.I.) Marco Trincavelli Silvia Coradeschi
	Main funder:	Vetenskapsrådet, Swedish Research Council (3.6 Million SEK for 2007-2010)

Synopsis: Olfaction is a useful sense for humans and other animals, and this has motivated us to study how olfaction can be added to mobile robots and other kinds of intelligent systems. This project takes as starting point existing olfactory sensor technology, in particular electronic noses, and has as aim to integrate such devices into autonomous and semi-autonomous mobile robots, distributed sensing networks and other intelligent systems. The potential applications include rescue robots, home assistance robots, intelligent homes, and environment surveillance. This project investigates one of the main challenge related to artificial olfaction which is to robustly classify odours, and in particular this project deals with the challenge of classification when e-noses are integrated in intelligent systems and the presence of air streams, multiple odour sources or other external factor inherently caused by the distributed network or the mobile robot (movement) influence the readings.

Ecology of Physically Embedded Intelligent Systems

	Research theme:	Robot ecologies
	Project staff:	Mathias Broxvall (P.I.) Jayedur Rashid Federico Pecora Alessandro Saffiotti
	Main funders:	ETRI, Korea (430,000 USD for 2004-2007) Vetenskapsrådet, Swedish Research Council (2.1 Million SEK for 2006-2009) Örebro University (internal faculty funds, ongoing)
	Project home:	http://aass.oru.se/~peis

Synopsis: In a decade or two, physically embedded intelligent systems (PEIS) will become part of our daily experience and improve the quality of life for every citizen, and especially for those in need of special physical and/or cognitive care like senior citizens. Examples of such systems include autonomous cleaners, intelligent appliances, smart user interfaces, and dedicated devices able to carry objects or perform assistive tasks. The goal of this project is to allow these systems to break their isolation, and to form an ecology of communicating and cooperating compontens. The functionality of each individual unit will be improved by this interaction: in practice, each system may use functionalities from another system in order to compensate or complement its own. By giving our systems a design that allows distributed knowledge, deliberation and cooperation, the whole ecology of systems becomes more robust and flexible than their individual parts. In this project we investigate the kind of systems described above and develop missing key technologies required to enable this kind of emergent systems. This entails research within distributed intelligence (knowledge and deliberation) as well as considering theories from fields such as cooperative anchoring and planning in light of the new problems posed by these complex emergent systems.

Human-Aware Planning and Execution

	Research themes:	Cognitive robots, Robot ecologies
	Project staff:	Lars Karlsson (P.I.) Marcello Cirillo Alessandro Saffiotti
	Main funders:	CUGS, National Graduate School in Computer Science (1.8 Million SEK for 2006-2008) Örebro University (internal faculty funds, ongoing)

Synopsis: This project considers a joint "symbiotic" system that includes both robots and humans. It aims at developing sound techniques to dynamically generate courses of actions for this system that ensure that execution will progress toward the system's goal, while avoiding negative interference and safety hazards when combining the robot's and the human's actions. To do so, the state of the distributed robotic system and the one of the humans must both be taken into account. The humans however must be considered with a mixed role: partly as independent agents on whose actions the robot may have expectations, and partly as a resource from which the robot can request actions (thus altering the above expectations).

Monitoring and Execution in Smart Environments

	Research themes:	Robot ecologies, Cognitive robots
	Project staff:	Federico Pecora (P.I.) Marcello Cirillo Alessandro Saffiotti
	Main funders:	Örebro University (internal faculty funds, ongoing)

Synopsis: In this project we address the problem of realizing a service-providing reasoning infrastructure for proactive human assistance in intelligent environments. Specifically, we are building an architecture which combines two key capabilities for contextualized service provision, namely (1) human activity recognition and (2) planning for controlling pervasive actuation devices. The system leverages temporal knowledge represented as relations in Allen's interval algebra and constraint-based temporal planning techniques (the current implementation of the system is based on OMPS, a temporal planner developed at ISTC-CNR). The system seamlessly interleaves context deduction capabilities and plan generation/execution capabilities, both at the modeling level and at the reasoning level. Some of the research directions entailed by this project are: efficient temporal reasoning algorithms for on-line inference; how to deal with multiple hypotheses in context recognition; combining data-driven approaches (e.g., Hidden Markov Models) with knowledge-driven approaches (e.g., temporal reasoning) to obtain a more flexible and adaptable system; dealing with sensory uncertainty.

Navigation System for Automatic Loaders

	Research themes:	Cognitive robots
	Project staff:	Alessandro Saffiotti (P.I.) Johan Larsson Mathias Broxvall
	Partners:	Atlas Copco AB
	Main funder:	RobotDalen, Swedish Robot Valley (on funds from Vinnova and the EU)

Synopsis: This project is an industrial co-operation with Atlas Copco Rock Drills AB. The project aims at the development of a robust and flexible navigation system for underground mine loaders, which has only modest requirements in terms of infrastructure. The goal is to enable autonomous and semi-autonomous operation of these vehicles. The approach is to build upon the techniques for office navigation based on behavioral plans, previously developed in our lab, to navigate in the mine tunnels.

Human-Robot Collaboration for Remote Operation

	Research themes:	Cognitive robots
	Project staff:	Mathias Broxvall (P.I.) Mattias Seeman Alessandro Saffiotti
	Partners:	Rotundus AB
	Main funder:	RobotDalen, Swedish Robot Valley (on funds from Vinnova and the EU)

Synopsis: This project is an industrial co-operation with Rotundus AB. The project aims at the development of techniques for remote operation, in which the responsibility is shared between the robot and the human operator. The goal is to lower the cognitive burden posed to the human operators, allowing more effective and secure tele-operation. We focus on one specific facet of shared autonomy: presentation of sensor data in a human-friendly way, using a 3D, third person, panoramic view. We plan to experimentally verify the benefit of our approach to increase the effectiveness of tele-operation tasks. The research is being performed on a spherical robot produced by Rotundus AB.

MULTISENS - Continuous Monitoring and Automatized Localization of Soil Pollutants

	Research theme:	Artificial olfaction
	Project staff:	Silvia Coradeschi (P.I.) Marco Trincavelli Amy Loutfi Bert Allard
	Main funders:	RobotDalen, Swedish Robot Valley (on funds from Vinnova and the EU) (150,000 SEK for 2008-2009) Örebro University (internal faculty funds, ongoing)

Synopsis: MULTISENS represents a collaborative effort that combines the leading competences in the two disciplines of environmental science and robotics and automation. Through this collaboration, state of the art in autonomous sensor systems will be applied to soil remediation in order to provide online and continuous monitoring of specific pollutants. This will enable the possibility to cost-effectively and efficiently isolate potential hotspots where high concentrations of pollutants are present and also enable the possibility to follow the remediation treatment and determine its effectiveness in real-time.

MEDNOSE - Electronic Noses for Medical Applications and Diagnosis of ENT Bacteria

	Research theme:	Artificial olfaction
	Project staff:	Amy Loutfi (P.I.) Marco Trincavelli Silvia Coradeschi Bo Söderquist Per Thunberg
	Partners:	University Hospital in Örebro
	Main funders:	NovaMedTech, on funds from the EU (680,000 SEK for 2008-2010) Örebro University (internal faculty funds, ongoing)

Synopsis: Sepsis also known as blood poisoning is caused by the presence of micro-organisms in the blood such as bacteria. Quick administration of antibiotic treatment is crucial as sepsis can lead to septic shock, multiple organ dysfunction and even death. The standard procedure for diagnosis involves routine microbiological blood cultures. Such procedures can take at least 36 hours to several days before diagnosis can be made. The project MEDNOSE's aim is to exploit state of the art technologies in compact gas sensor in order to, at an early stage, detect and identify different bacteria culture that are present in blood samples through detection of the released gases in the first incubation stage.

The Comfortable Sleep Laboratory

	Research theme:	Robots for humans
	Project staff:	Lars Karlsson (P.I.) Mathias Broxvall Martin Längkvist Federico Pecora
	Partners:	University Hospital in Örebro Svensk Förening för Sömnforskning och Sömnmedicin Mälardalens Högskola JonDeTech AB, Uppsala
	Main funder:	NovaMedTech on funds from the EU (500,000 SEK for 2008-2010)

Synopsis: Sleep disorders are a problem that is estimated to affect a considerable part of the Swedish population (10-15%), and a problem which can have a significant negative impact on the health and wellbeing of the affected individual and indirectly also on his/her family. Elderly persons are especially affected. This project aims at developing equipment that can be used in the patient's home to collect information about sleep-related behaviors, including when the patient goes to bed and leaves the bed, how often the patient goes up during night, and how much the patient moves in bed. For that purpose, we are developing a suite of sensors (e.g. light, movement, pressure, temperature) that can either be carried by the patient, or positioned at strategic places in the patient's home. the project also aims at improving the recording of physiological data such as EEG when the patient is at home. In particular, we want to be able to automatically detect when there are problems with the recording, e.g. when an electrode is coming off.

Visualization of the Aorta Cusp Using Time-Sampled 3D Ultrasound Data

	Research theme:	Robots for humans
	Project staff:	Mathias Broxvall (P.I.) Per Thunberg
	Partners:	University Hospital in Örebro
	Main funder:	NovaMedTech on funds from the EU

Synopsis: Many commonly occuring forms of heart diseases, such as insufficience, involve the aorta cusp and constitute acute medial emergencies. In many cases where the cusp is repaired or replaced, the surgeon needs to make accurate medical decisions based on the shape and form of the cusps. Due to limitations in visualisation of 3D medical imaging, these decisions currently have be performed during surgery which gives only a very limited timeframe for deliberation. In this project, we investigate a method to pre-compute a quantitatively correct visualisation of the aorta cusps as they will appear during surgey, based on non-invasive 3D ultrasound imaging. The core of the method is to create a 3D model of the cusps as they are working, and to unfold this model to reflect the way the cusps will be opened during the surgery. This will allow the diagnosing and preparation of patients well before surgery. Additionally, it will offer the physician the possibility to investigate cusps as they are operating under normal cirumstances. This gives additional information that can be used for medical assessments.

Ängen: a Physical Testbed Apartment for Prototyping Distributed Care Technologies

	Research theme:	Robots for humans
	Project staff:	Federico Pecora (P.I.) Silvia Coradeschi Amy Loutfi Alessandro Saffiotti
	Partners:	Länsgården AB, Örebro Jatab Care AB, Örebro Giraff AB, Västerås
	Main funder:	NovaMedTech on funds from the EU (150,000 SEK for 2009-2010)

Synopsis: Elderly citizens who choose to live in an assisted living facility require an array of assistive services. The specific nature of these services depends strongly on the health, social and economic situation of the elderly user. AASS has developed a framework for easily integrating and composing off-the-shelf technological devices for domestic assistance. This project will leverage the experience and tools available at AASS (e.g., the PEIS Home, a testbed environment for research prototype development) to develop easily-deployable, scalable products for highly customized distributed care solutions. The results of the project are intended for deployment a real environment, the Ängen senior residence facility in Örebro. To this end, we will develop a new testbed environment at AASS in which a selection of the most stable research products will be re-factored and integrated to fit the needs of specific elderly users and real-world deployment. In this project, AASS will collaborate with Läänsgården AB, a local, publicly-controlled housing company specialized in the development of elderly residences. Also, involved in the project are Jatab Care AB, a company based in Örebro which produces a unique assistive device for patient mobilization, and Giraff AB, a newly created Swedish SME (March 2009) focused on telepresence robotics (see also project ExCITE above).

Evaluation of Domestic Robotic Technologies for the Elderly

	Research theme:	Robots for humans, Robot ecologies
	Project staff:	Amy Loutfi (P.I.) Federico Pecora Gion Koch Svedberg
	Partners:	ISTC, National Research Council, Rome PRO, National Pensioners' Organisation
	Main funder:	KP Pension & Försökring (60,000 SEK for 2008-2010) Örebro University (internal faculty funds, ongoing)

Synopsis: For the elderly, the prolongation of residence in their own homes is an important factor as it increases the feeling of safety and security, may results in economic savings and promotes general well-being. For this to be possible, however, the need for the home to be an active and a dynamic place accomodating to its inhabitants becomes increasingly important. As breakthroughs in technological research are facilitating the integration of embedded technologies in the home, it is becoming essential to examine how these trends are perceived and experienced by the aging population. This is important, as it will determine how well these new technologies will be accepted and used. In this project, we study the interaction between the elderly and a so called "intelligent home" environment that is equipped with a number of features intended to assist its inhabitants with in day-to-day activities. We do this using a variety of tools, including video-based evaluation, cross-cultural studies, and Wizard-of-Oz experiments in a real home setting.