Program of the workshop
Proceedings can be downloaded here
Program of the workshop
Proceedings can be downloaded here
8.50am: Tracking Mobile Objects with Several Kinect using HMMs and Component Labelling (17' presentation and 3' questions)
Amandine Dubois Francois Charpillet
Abstract: This paper proposes a markerless system whose
purpose is to detect falls of elderly people at home. To track
human movements we use Microsoft Kinect camera which
allows to acquire at the same time a RGB image and a depth
image. One contribution of our work is to propose a method
for fusioning the information provided by several Kinects based
on an occupancy grid. The observed space is tesselated into
cells forming a 3D occupancy grid. We calculate a probability
of occupation for each cell of the grid independently of its
nearby cells. From this probability we distinguish whether the
cells are occupied by a static object (wall) or by a mobile
object (chair, human being) or whether the cells are empty.
This categorization is realized in real-time using a simple three
states HMM. The use of HMMs allows to deal with an aliasing
problem since mobile objects result in the same observation as
static objects. This paper also deals with the issue regarding the
presence of several people in the field of view of camera by using
the component labelling method. The approach is evaluated in
simulation and in a real environment showing an efficient realtime
discrimination between cells occupied by different mobile
objects and cells occupied by static objects.
9.10am: Localization and Navigation of an Assistive Humanoid Robot in a Smart Environment (17' presentation and 3' questions)
Enric Cervera, Amine Abou Moughlbay, Philippe Martinet
Abstract: Assistive humanoids that manipulate objects in everyday environments are potentially useful to improve the lives
of the elderly or disabled. To retrieve or deliver objects at home,
precise localization is needed. But localization of humanoid robots
is a challenging issue, due to rough odometry estimation, noisy
onboard sensing, and the swaying motion caused by walking.
To overcome these limitations, we advocate for the use of
external sensors for localization and navigation in a smart home
environment. As opposed to a stand-alone self contained robot,
our humanoid benefits from the information coming from other
sensing devices in the environment. In order to achieve robust localization while walking, and retrieve an object from the floor,
we use RGBD camera information from an external Kinect
sensor. Monte Carlo localization estimates the 6D torso pose
estimation of the humanoid, which is then used for closed-loop
navigation control. Experiments with a NAO humanoid point out
that, by cooperating with the environmental sensors, the overall
precision of robot navigation is dramatically improved.
9.30am: Autonomous Shopping Cart Platform for People with Mobility Impairments (17' presentation and 3' questions)
Luca Marchetti, Daniele Pucci, Pascal Morin
Abstract: Providing a platform able to interact with a specific
user is a challenging problem for assistance technologies.
Among the many platforms accomplishing this task, we address
the problem of designing an autonomous shopping cart. We
assume that the shopping cart is set-up on a unicycle-like
robot endowed with two sensors: an RGB-D camera and a
planar laser range finder. To combine the information from
these two sensors, a data fusion algorithm has been developed
using a particle filter, augmented with a k-clustering step
to extract person estimations. The problem of stabilizing the
robot’s position at a fixed distance from the user has been
solved through classical control design. Results on a real mobile
platform verify the effectiveness of the approach here proposed
10.10am: Keynote speech from Rachid Alami
Senior scientist at the Robotics and Artificial Intelligence Group from LAAS
10.45am: Intention Driven Human Aware Navigation for Assisted Mobility (17' presentation and 3' questions)
Jorge Rios-Martinez, Arturo Escobedo, Anne Spalanzani, Christian Laugier
Abstract: Ensuring proper living conditions for an ever
growing number of elderly people is an important challenge for
many countries. The difficulty and cost of hiring and training
specialized personnel has fostered research in assistive robotics
as a viable alternative. In particular, this paper studies the case
of a robotic wheelchair, specifically its autonomous navigation
and user adapted control. Integration of a technique to interpret
user intention using head movements and a human aware
motion planner is presented. Test results exhibit emerging
behavior showing a robotic wheelchair interpreting gesture
commands and taking the user to his desired goal, respecting
social conventions during its navigation.
11.05am: Robot Navigation Taking Advantage of Moving Agents
(17' presentation and 3' questions)
Procopio Stein, Anne Spalanzani, Vitor Santos and Christian Laugier
Abstract: A crucial requirement for service robots is to be
able to move in dynamic environments shared with humans as
well as interact with them. Navigation in such environments is
a challenging task, as the environment is constantly changing,
future states have to be predicted and planning and execution
must be carried on-line.
However, even in very complex situations, humans can easily
find a path that avoid both dynamic agents and static obstacles.
This paper proposes a technique to take advantage of the
human movement in such populated environments, selecting
a leader to be followed in a probabilistic fashion, according to
the robot’s desired destination.
In this way, the robot can take advantage of the paths
traveled by humans, effortlessly avoiding dynamic and static
features as the human leader does, relieving the robot from the
burden of having to generate its own path.
11.25am: Keynote speech from Matei Ciocarlie (30' presentation and 5' questions)
Research Scientist and Area Lead at Willow Garage
"Mobile Manipulation Through An Assistive Home Robot"
Abstract: We present a mobile manipulation platform operated
by a motor-impaired person using input from a headtracker,
single-button mouse. The platform is used to perform
varied and unscripted manipulation tasks in a real home,
combining navigation, perception and manipulation. The operator
can make use of a wide range of interaction methods
and tools, from direct tele-operation of the gripper or mobile
base to autonomous sub-modules performing collision-free base
navigation or arm motion planning. We describe the complete
set of tools that enable the execution of complex tasks, and share
the lessons learned from testing them in a real user’s home. In
the context of grasping, we show how the use of autonomous
sub-modules improves performance in complex, cluttered environments,
and compare the results to those obtained by novice,
able-bodied users operating the same system.
1.30pm: Keynote speech from Hiroshi Ishiguro (30' presentation and 5' questions)
Professor of Osaka University, Japan and Fellow of ATR,
Ishiguro@sys.es.osaka-u.ac.jp
"From teleoperated androids to cellphones as surrogates"
In order to understand the meaning of human presence, we have developed Geminoid which is an teleoperated android of myself.
With the android, we could learn how people can adapt the new media. Based on the knowledge, we have recently developed a simpler
teleoperated android with the minimal humanlike appearance. The new android is called Telenoid. People can easily adapt to Telenoid
and enjoy conversations by using it. Further, we are remaking it with a cell-phone size. It is called Elfoid. We believe the new type
of cell-phone can transmit our presence to distant places and changes our life style again.
2.05pm: Social Inclusion of Senior Citizens by a Teleoperated Android: Toward Inter-generational TeleCommunity Creation* (17' presentation and 3' questions)
Ryuji Yamazaki, Shuichi Nishio, Hiroshi Ishiguro, Takashi Minato, Marco Norskov, Nobu Ishiguro, Masaru Nishikawa and Tsutomu Fujinami
Abstract: As populations continue to age, there is a
growing need for assistive technologies that help senior
citizens maintain their autonomy and enjoy their lives.
We explore the potential of teleoperated androids, which
are embodied telecommunication media with humanlike
appearances. Our exploratory study focused on the social
aspects of Telenoid, a teleoperated android designed as a
minimalistic human, which might facilitate communication
between senior citizens and its operators. We conducted
cross-cultural field trials in Japan and Denmark
by introducing Telenoid into care facilities and the private
homes of seniors to observe how they responded to it.
In Japan, we set up a teleoperation system in an elementary
school and investigated how it shaped communication
through the internet between the elderly in a care
facility and the children who acted as teleoperators. In
both countries, the elderly commonly assumed positive
attitudes toward Telenoid and imaginatively developed
various dialogue strategies. Telenoid lowered the barriers
for the children as operators for communicating with
demented seniors so that they became more relaxed to
participate in and positively continue conversations using
Telenoid. Our results suggest that its minimalistic human
design is inclusive for seniors with or without dementia
and facilitates inter-generational communication, which
may be expanded to a social network of trans-national
supportive relationships among all generations.
2.25pm: Multi-sensors engagement detection with a robot companion in a home environment(17' presentation and 3' questions)
Wafa Benkaouar and Vaufreydaz Dominique
Abstract: Recognition of intentions is an subconscious cognitive
process vital to human communication. This skill enables
anticipation and increases the quality of interactive exchanges
between humans. Within the context of engagement, i.e. intention
for interaction, non-verbal signals are used to communicate
this intention to the partner. In this paper, we investigated
methods to detect these signals in order to allow a robot to
know when it is about to be addressed. Classically, the human
position and speed, the human-robot distance are used to detect
the engagement. Our hypothesis is that this method is not
enough in the context of a home environment. The chosen
approach integrates multimodal features gathered using a robot
equipped with a Kinect. The evaluation of this new method of
detection on our corpus collected in spontaneous conditions
highlights its robustness and validates use of such a technique
in real environment. Experimental validation shows that the
use of multimodal sensors gives better precision and recall
than the detector using only spatial and speed features. We
also demonstrate that 7 multimodal features are sufficient to
provide a good engagement detection score.
2.45pm: Can smart rollators be used for gait monitoring and fall prevention ? (17' presentation and 3' questions)
C. Dune, P. Gorce, J.-P. Merlet
Abstract: Clinical evaluation of frailty in the elderly is the
first step to decide the degree of assistance they require. This
evaluation is usually performed once and for all by filling
standard forms with macro-information about standing and
walking abilities. Advances in robotics make it possible to
turn a standard assistance device into an augmented device.
The existing tests could then be enriched by a new set of
daily measured criteria derived from the daily use of standard
assistance devices. This paper surveys existing Smart Walker
to figure out whether they can be used for gait monitoring
and frailty evaluation, focusing on the user-system interaction.
Biomechanical gait analysis methods are presented and compared
to robotics system designs, to highlight their convergences
and differences. On the one hand, monitoring devices try to
estimate accurately biomechanical features, whereas, on the
other hand, walking assistance and fall prevention do not
systematically rely on an accurate human model and prefer
heuristics on the user-robot state.
3.25pm: Keynote speech from Charles Fattal
MD, PhD, President of the Scientific Committee for the National Expert Center on Assistive Robotics
Laboratoire m2h, Centre Mutualiste Neurologique Propara, Montpellier (France)
For the past 20 years, assistive robotics for manipulation offers practical,
useful solutions that are available on the market for persons with quadriplegia.
However, there is still the need to validate the effectiveness and reliability of the impact of
these technologies in the quest for a more independent life and better quality of life in this population.
In order to achieve these objectives, it is essential to proceed with fine and thorough analysis of the users’ needs
and expectations and ensure that the evolution potential, adaptation and intuitivism of the command interfaces
are optimal in order for the user to appropriate these new technologies. Finally it is also important that
maintenance and after-sale services be reactive and efficient to avoid adding a detrimental technological
dependence on top of physical impairments.
4.00pm: User studies of a mobile assistance robot for supporting elderly: methodology and results
(17' presentation and 3' questions)
A. Garzo, L. Martinez, M. Isken, D. Lowet and A. Remazeilles
Abstract: The FP7 European project Florence aims at investigating how current state of the art robotic technology
can be used to support elderly to live longer independently at home. During the whole development for our robotic platform
users and caregivers were strongly involved through focus groups, Wizard of Oz tests and functional validation within controlled environments.
In this project we focus on type of services: on the one hand we consider lifestyle activities like tele-presence and coaching services
and on the other hand we also propose safety services like fall handling. This article describes the user-centered mechanisms we put
in place during this project and compiles the information we could gather during interaction with potential end-users. The collected
data naturally strongly influences our development. They are provided here as well as general guidelines for any further assistive
mobile robot development.
4.20pm: ANG, a family of multi-mode, low-cost walking aid(17' presentation and 3' questions)
J-P. Merlet
Abstract: ANG is a family of low cost modular walking
aids based on commercially available Rollator. We present
two models of this family: the simplest ANG-light which is
intended to be used as a diagnosis tool for walking and as a fall
detection tool and ANG-II, a motorized walker with over 20 onboard
sensors. Trajectography obtained through ANG-light
with 24 healthy subjects has been used to obtain gold standards
of walking pattern and the walker is currently being used at
Nice hospital with 30 elderly end-users in order to determine
if trajectory records obtained during the daily use of a walker
may allow doctors to objectively characterize abnormal walking
patterns, to follow the progress of a rehabilitation process
and to detect emerging pathologies. The more sophisticated
ANG-II is intended to be used as a test platform for original
functionalities, some of which will be presented.
4.40pm: Velocity Control for Walk Assistance by Endeffector Force in the Leg Coordinate based on the Biarticularly-actuated System
(17' presentation and 3' questions)
Shinta Sonokawa, Yunha Kim, Sehoon Oh and Yoichi Hori
Abstract: This paper proposes a novel velocity control of
the center of mass (COM) of a human body with attached
ankle foot orthosis (AFO) during the stance phase. We propose
a novel coordinate system for COM that achieves model simplificatio
with the biarticularly-actuated system. This allows
for simple control design of the velocity and position of COM.
In addition to simplifie control, the proposed mathematical
model for AFO has a simple structure that reproduces the
biarticularly-actuated system using passive elements such as
springs. Simulation results and comparison with conventional
methods verify the effectiveness of the proposed control design.