UPEC-LISSI INRIA PAL-INRIA                                                                                                                 


08:20 - 08:30

Introduction and welcome

Session 1

08:30 - 10:00

Invited speaker 1
Jean Pierre Merlet
, INRIA Sophia Antipolis, France
Title “Mobility, a necessity for preserving the autonomy of frail people” (30')

Abstract: A two years interview period among users, their helpers and the medical community has allowed us to identify the priorities of assistance robotics for frail people (e.g. elderly and handicapped people) together with the guidelines that must be satisfied to fulfill them. Mobility, local or global, is a major priority both for the user and the helpers and should be satisfied with low-cost and non intrusive solutions that must be adaptable to the users and their environment. Furthermore medical monitoring of mobility is a major request from the medical community as mobility analysis provides pertinent information on the functional and cognitive status of the user. We will illustrate what are the major scientific and technical issues that must be solved to satisfy these requirements together with possible solutions that allow for mobility assistance, fall detection/prevention and medical monitoring.
Invited speaker 2
Yiannis Demiris
, Imperial College London, UK
Title “Personalising assistive robotics for people with disabilities” (30')

Abstract: To personalise the interaction with human users and maximise the impact of the assistance they can give, robots need to develop life-long user models that can be used to recognise human actions, predict human intentions and assist intelligently, while constantly adapting to changing human profiles. In this talk, I will describe our latest research in generative models for embodied action perception, social cognition and machine learning mechanisms for humanoid robots and smart robotic wheelchairs. I will describe their application towards adaptive robotic assistants for children and adults with disabilities.
Invited speaker 3
Kyoungchul Kong
, Sogang University, Seoul, Korea
Title “Actuation and Control in Robotic Assistance for Incomplete Paraplegic” (30')

Abstract: Mechatronic technologies are steadily penetrating in our daily lives. We are surrounded by mechatronic products and interact with them in many ways. In particular, mechatronic devices may potentially improve the quality of life of complete/incomplete paraplegics. Robotic assistance technologies for the complete and incomplete paraplegics require different technologies in actuation, sensing, and controls. In this talk, several key technologies for effectively assisting incomplete paraplegics are introduced. These technologies include 1) sensing technologies for observing the dynamic state of the human wearing a robot, as well as for identifying the intent of humans, 2) decision making algorithms to decide about the right amount of assistance, 3) actuation technologies to provide precise assistive forces, and 4) control algorithms for providing assistance as needed. In the design of mechatronic devices interacting with humans, the dynamic characteristics of the human body are an important issue. Furthermore, the compatibility between machines and humans must be optimized when the human and the robot must share the control of the combined overall system as in the case of incomplete paraplegic assistance. If the assistive device is interacting with normal and healthy humans, the design may take advantage of the robust and intelligent controllability of the human brain system. In the case of assistive devices for patients with incomplete paraplegia, such approaches may not be appropriate, and the controller is required to be precise, robust and intelligent. In these aspects, various engineering approaches will be introduced in this talk, and experimental demonstrations will also be provided to verify the proposed methods.

10:00 - 10:30

Coffed break 30'

Session 2

10:30 - 12:30

Invited speaker 1
Mohamed Chetouani
, ISIR, UPMC, Paris, France
Title “Computational Modeling and Personal Robotics for Extracting Social Signatures” (30')

Abstract: Social signal processing is an emerging research domain with rich and open fundamental and applied challenges. In this talk, I'll focus on the development of social signal processing techniques for real applications in the field of psycho-pathology. I’ll overview recent research and investigation methods allowing neuroscience, psychology and developmental science to move from isolated individuals paradigms to interactive contexts by jointly analyzing behaviors and social signals of partners. From the concept of interpersonal synchrony, we’ll show how to address the complex problem of evaluating children with pervasive developmental disorders. These techniques are also demonstrated in the context of human-robot interaction by a new way of using robots in autism (moving from assistive devices to clinical investigations tools). I will finish by closing the loop between behaviors and physiological states by presenting new results on hormones (oxytocin, cortisol) and behaviors (turn-taking, proxemics) during early parent-infant interactions.
Invited speaker 2
Hatice Kose
, Istanbul Technical University, Turkey
Title: “An Assistive Humanoid Robot System for Teaching Sign Language” (30')

Abstract: In this talk, I’ll present a socially assistive humanoid robot system for teaching sign language to children with communication impairments, by means of imitation based interaction games. In this study, we use a specially designed 5 fingered robot platform with expressive face (Robovie R3) which is able to interact and communicate with hearing impaired children using signs and visual cues. The robot is able to recognize a selected set of words from Turkish Sign Language (TSL) and able to generate these signs accurately using hand, arm and head gestures. The test studies with hearing impaired children from collaborating schools, and their results will be presented in details.
Invited speaker 3
Pericle Salvini
, Scuola Superiore Sant'Anna, Pisa, Italy
Title: “On Ethical, Legal and Social Issues of Care Robots” (30')

Abstract:According to recent figures, in the next years, Western developed societies will be supposed to face the “aging problem”. The population aged 60 will surpass that of younger people and, to make things even worst, current trends in society indicate that family carers are no more willing to look after their older relatives. Such a situation has given more emphasis to the rise of robotics as a possible solution to deal with the demographic change and the new social norms in the care of elderly and disabled people. The ways in which robotics has been proposed to address the “aging problem” are manifold: ranging from humanoids, general purpose robots, to less invasive, distributed and task-specific systems. This talk intends to provide an overview of the main ethical, legal and societal challenges concerning the use of care robots, which, together with appropriate design procedures, may contribute to design better robots.
Interactive session (contributed papers) (2×15')
Authors: Luigi Palmieri, Andrey Rudenko and Kai Arras
Abstract 1: We introduce and show preliminary results of a fast randomized method that finds a set of K paths lying in distinct homotopy classes. We frame the path planning task as a graph search problem, where the navigation graph is based on a Voronoi diagram. The search is biased by a cost function derived from the social force model that is used to generate and select the paths. We compare our method to Yen's algorithm, and empirically show that our approach is faster to find a subset of homotopy classes. Furthermore our approach computes a set of more diverse paths with respect to the baseline while obtaining a negligible loss in path quality.
Authors: David Fischinger, Astrid Weiss and Markus Vincze
Abstract 2: Household service robots are intended to support their users in a variety of tasks, where fetch-and-carry scenarios are a key element. One of the reasons why these universally usable robot-butlers have not reached marketability yet is the limited capability of service robots to physically interact in “real” environments, in specific object manipulation and grasping. In this paper we present an approach to enable robust grasping for service robots. We extend previous work to enable grasping of known and unknown objects. The method is based on the topography of a given scene and abstracts grasp-relevant structures to enable machine learning techniques for grasping tasks. We describe how to use the approach to adapt the robotic hand opening width before grasping to enable grasping in situations were a fully opened gripper could not approach the object. A proof-of-concept study with a service robot prototype demonstrates the feasibility of our approach.

12:30 - 14:00

Lunch 1h30

Session 3

14:00 - 15:30

Invited speaker 1
Auke Jan Ijspeert
, EPFL, Lausanne, Switzerland
Title: “Using modular robotics for making adaptive and assistive furniture” (30')

Abstract: This talk will present the design and control of modular robots, called Roombots, to be used as building blocks for furniture that moves, self-assembles, and self-reconfigures. Modular robots are robots made of multiple simple robotic modules that can attach and detach. Connectors between units allow the creation of arbitrary and changing structures depending on the task to be solved. Compared to “monolithic” robots, modular robots can potentially offer higher versatility and robustness against failure, as well as the possibility of self-reconfiguration. The type of scenario that we envision is a group of Roombots that is merged with the house environment and the furniture to assist persons with limited mobility. In the long run we aim at a smart assistive environment where the robotic furniture is at the service of the user and helps with important aspects of his/her daily life.
Invited speaker 2
Laura Marchal Crespo
, ETH, Zurich, Switzerland
Title “Rehabilitation robots that enhance patient motivation and participation” (30')

Abstract: Recent work in robot-aided training has focused on developing sophisticated robotic mechanisms in order to support movement training of complex movements. Although there is recent evidence that robotic guidance training can improve motor function more effectively than conventional therapist-assisted practice, the benefits seem to be limited to patients with severe impairment. Patient’s effort during physical training is an important factor in order to provoke motor plasticity; hence robotic devices could potentially decrease recovery if they encourage a decrease in effort, energy consumption, or attention during training. Augmenting the intensity of training is therefore, crucial for neurological patients. Intensity can be augmented not only by increasing the duration of training, but also by increasing the physical and mental engagement of the patient. This talk will present possibilities to enhance patient motivation and participation during robot-assisted movement training. Innovative control algorithms that amplify or create movement errors in order to maximize the training benefits of the already existing rehabilitation robot are proposed.
Interactive session (contributed papers) (2×15')
Authors: Kenji Koide and Jun Miura
Abstract 1: This paper describes a person identification method using inertial measurement unit (IMU) and laser range finder (LRF) for mobile service robots. The method is based on the matching of foot strike timings from LRF and IMU. The robot is equipped with two LRFs, and the target person puts an IMU on his/her foot. The foot strike timing of the target person is measured by LRF and IMU independently. By matching these data, the robot reliably identify the target person in LRF data. For a robust target tracking, we apply Bayesian inference to estimation of the target person. Person identification experiments and a person following experiment are conducted in order to show the validity of our method.
Authors: Je Hyung Jung and Jan Veneman
Abstract 2: This paper addresses a perturbation detection method on the basis of calculation of Centroidal Momentum (CM) in human waking. CM is referred to as the combination of linear and angular momenta at Center of the Mass (CoM). In the method, variation of CM patterns between unperturbed and perturbed walking plays a crucial role in detecting perturbations. The method has been evaluated with experimental data of human walking and results show that the method is capable of detecting moderate and strong perturbations determined by combination of diverse durations and magnitudes of disturbance force. Average detection time was about 334 msec. The proposed method will be further applied to human-exoskeleton interactive walking in order to investigate whether CM can be employed as a kind of stability index to assess the actual state of balance of the human supported by a lower limb exoskeleton.

15:30 - 16:00

Coffee break 30'

Session 4

16:00 - 18:00

Invited speaker 1
Norihiro Hagita
, ATR, Japan
Title “What are Service Robots Allowing “Harmonious” Collaboration Using Co-experience Knowledge and Wisdom?” (30')

Abstract: As information communication technology (ICT) reaches deeper and deeper into society, the volume of information emanating from the Internet, mobile devices, sensors, and other sources continues to rise, leading to problems in terms of the inability to effectively apply knowledge gained from these sources and creating new problems regarding matters such as the receptivity of newly produced knowledge in terms of ethical, legal and social issues. Recent sensing technologies for human behavior, gesture recognition, and human-robot interaction may give us more human-centric tools for searching for co-experience knowledge and wisdom on the Internet beyond conventional keyword search. This talk will introduce a new research area targeting intelligent information processing systems that create co-experience knowledge and wisdom through “balanced or harmonious collaboration” of humans and machines partially supported by the Japan Science and Technology Agency (JST). The balanced or harmonious collaboration means that not only users but also human society may feel the services of robots (or systems) acceptable in the human-robot collaboration. Application of intelligent systems is envisioned for the following services: living assistance for elderly or handicapped, medical diagnosis support, guidance for daily life practices, robotic services in human populated environments, etc. These ongoing research projects will be introduced.
Invited speaker 2
Koji Kamei
, ATR, Japan
Title “Semantic de-composition, re-composition and intelligence of robotic services: upon Cloud Networked Robotics Framework” (30')

Abstract: While deployment of IoT technologies and robotic services for life-support tasks is widely spreading, development of such services should be supported by common platform technologies, semantic modeling and reasoning tools based on standards. We start in this talk, by providing an overview of the standardization activities with a special focus on human robotic interaction, IoT and ubiquitous networked robots topics. An implementation based on the current standards named ‘UNR (ubiquitous network robot) Platform, that supports modularized development of both robotic services and components and coordination among them, will be presented also. Next, we will elaborate on the next targeted standardization and implementation activities within the OMG and IEEE RAS organizations in order to enable more collaboration and/or composition of robotic services using standard robotic ontologies and AI based reasoning tools. To this end we will present the Cerebro knowledge representation and reasoning (KR&R) Framework. The latter supports from one hand a semantic meta modeling of the robotic applications in the cloud according to both closed and open world assumption and from the other hand it offers a set of evolving reasoning tools that implements description logic as well as commonsense reasoning with answer set programming. These modeling and reasoning tools will help in developing the next generation of platform for cloud networked robotics with more integration with cloud based knowledge bases and new features such as flexible and semantic interoperability between heterogeneous parts, context and intention awareness, flexible composition and decomposition to the robotic application components through the Cerebro KR&R Framework.
Invited speaker 3
Mehul Bhatt
, University of Bremen, Germany
Title “Spatial Reasoning for Robotics and Smart Environments” (30')

Abstract: People-centred perceptual sensemaking & interaction with cognitively founded conceptualisations of space, actions, and change are crucial components of computational cognitive systems, and multimodal interaction & assistive technologies. Our research in this context focusses on the formal and computational characterisation of space and dynamic visuo-spatial thinking —encompassing spatial abstraction, reasoning, and learning— from the viewpoint of assistive technologies in domains such as robotics, ambient & smart environments, building architecture design, and dynamic geography. In essence, we have prioritised those areas where processing, (semantic) interpretation, and (high-level) control in a dynamic spatial environment are crucial, with the central underlying theme being that of integrated reasoning about space, events, actions, change, and interaction. This talk will focus on two complementary strands of research: - cognitive vision. presenting a declarative model of perceptual narratives of space and motion for the embodied visuo-spatial grounding of observations (e.g., RGB-D data, point-clouds) in the real world - commonsense cognitive robotics. presenting high-level agent control with integrated planning and explanatory inference, in particular focussing on narrative-based (logical) abductive reasoning as a basis to make sense of observations, identify abnormalities etc.
From a methodological viewpoint, I will emphasise the role of specialised knowledge representation and reasoning mechanisms for commonsense reasoning about space, actions, and change, in particular positioning recent advances in logic-based declarative spatial reasoning encompassing - geometric and qualitative spatial reasoning within constraint logic programming using the CLP(QS) spatial reasoning system -non-monotonic (spatial) reasoning with answer set programming as a basis for high-level spatio-temporal abduction
I will demonstrate the methods using case-studies from the viewpoint of applications such as human activity interpretation, semantic model generation from visual data, high-level reasoning in ambient intelligence scenarios etc.
Interactive session (contributed papers) (2×15')
Authors: Claudio Coppola, Oscar Martinez Mozos and Nicola Bellotto
Abstract 1: The life span of ordinary people is increasing steadily and many developed countries are facing the big challenge of dealing with an ageing population at greater risk of impairments and cognitive disorders, which hinder their quality of life. Monitoring human activities of daily living (ADLs) is important in order to identify potential health problems and apply corrective strategies as soon as possible. Towards this long term goal, the research here presented is a first step to monitor ADLs using 3D sensors in an Ambient Assisted Living (AAL) environment. In particular, the work here presented adopts a new 3D Qualitative Trajectory Calculus (QTC$_\text{3D}$) to represent human actions that belong to such activities, designing and implementing a set of computational tools (i.e. Hidden Markov Models) to learn and classify them from standard datasets. Preliminary results show the good performance of our system and its potential application to a large number of scenarios, including mobile robots for AAL.
Authors: Timm Linder, Fabian Girrbach and Kai O. Arras
Abstract 2: People tracking is an important prerequisite for socially compliant service robots that operate in human environments. In this paper, we take this challenge to new extremes by attempting to robustly track people in a 360-degree field around the robot in very crowded environments like a busy airport terminal. As a first contribution, we present a novel multi-modal people tracking framework that is modular, flexible and fully integrated with ROS. We believe that our framework can be of great benefit to researchers as it covers the entire people tracking pipeline, including powerful visualization and evaluation tools. Secondly, we present a number of simple extensions that can make tracking in crowded scenarios more robust. Finally, we compare our tracking method against a complex multi-hypothesis tracking system. Our results on real and synthetic data suggest that it is not the choice of data association which has the largest impact, but the underlying models that, for instance, control track initiation, deletion, and handling of occlusions. By showing that the simpler data association may be sufficient, we provide reasoning why the available resources on a resource-constrained mobile robot might be better spent on other tasks such as higher-level perception and reasoning.

18:00 - 18:10

Closing remarks