|
Invited Talks
Room Mombert
|
08:10
|
|
Welcome |
|
08:30
|
|
Modeling and Control of Distributed Stochastic Robotic Systems
Alcherio Martinoli
Technological advances in communication, embedded computing,
energy storage, sensors and actuators enable an increasingly ubiquitous
deployment of distributed, mobile, robotic systems in the real world.
However, often such systems are severely constrained in their resources
by cost, volume, or mass considerations imposed by the targeted
application. Such constraints typically result in an increased
stochasticity of the node behavior that has to be captured and
controlled with appropriate methods in order to obtain a more
predictable behavior at the collective system level. In this seminar, I
will describe a few recipes that allowed us to achieve such result
under peculiar scenarios. In particular, I will leverage case studies
concerned with self-aggregation and self-assembly of nodes to
illustrate such methods. Finally, I will conclude my seminar with some
of the lessons we learned over the last eighteen years of research in
this area and extrapolate some hints for future research directions to
overcome limitations of the current methods.
|
|
09:00
|
|
Testbeds, Models and Abstractions: Towards a Science of Distributed Robotics
Gaurav Sukhatme
Coming soon
|
|
Interactive session
Teasers: room Mombert
Interactive presentations: Interactive Sessions Area
|
|
09:30
|
|
Implementation of Cooperative Dynamic Behaviors in Networked Systems with Controllability Preservation (pdf)
Lorenzo Sabattini, Cristian Secchi, Matteo Cocetti and Cesare Fantuzzi
|
|
09:35
|
|
Discussion of Multi-Robot Exploration in Communication-Limited Environments (pdf)
Arnoud Visser, Julian de Hoog, Adrian Jiménez-González and José Ramiro Martínez de-Dios
|
|
09:40
|
|
A Software Framework for Multi-Robot Human Interaction (pdf)
Junaed Sattar, Marc Grimson and James Little
|
|
09:45
|
|
SwarmSimX and TeleKyb: Two ROS-integrated Software Frameworks for Single- and Multi-Robot Applications (pdf)
Johannes Lächele, Martin Riedel, Paolo Robuffo Giordano and Antonio Franchi
|
|
09:50
|
|
Multi-UAS Testbed for Wireless Communication Experiments (pdf)
Maciej Stachura, Jack Elston, Eric W. Frew, Brian Argrow, Timothy X. Brown and Cory Dixon
|
|
10:00
|
|
Coffee Break
|
|
10:30
|
|
Interactive presentations (Interactive Sessions Area)
|
|
11:30
|
|
Transfer back to room Mombert
|
|
Invited Talks
Room Mombert
|
|
11:40
|
|
Robotic Sensor Networks for Environmental Monitoring
Volkan Isler and Pratap Tokekar (presenter)
Robotic Sensor Networks (RSNs) hold the potential to revolutionize
environmental monitoring. Networks of robots and sensors will enable
scientists to collect data at unprecedented spatio-temporal scales,
from hard-to-reach places over long periods of time. Environmental
monitoring applications, in turn, motivate challenging research
problems in robotics. In this talk, I will present our lab's research
on two such problems: sensor planning algorithms and energy optimal
navigation for long term autonomy of RSNs. For sensor planning, I will
focus on the problems of accurately locating targets in minimum time
by designing paths for mobile sensors, or by deploying a minimum
number of stationary sensors. For energy optimal navigation, I will
present our results on finding the optimal paths and velocity profiles
to minimize energy consumption for car-like robots, and modeling solar
energy for efficient harvesting with mobile robots in complex urban
environments. Along with algorithmic results, I will present results
from field experiments on the system we are building for monitoring
radio-tagged invasive fish with autonomous boats in summers, and
mobile robots (on frozen lakes) in winters.
|
|
12:10
|
|
Platform Scale and Group Size Considerations in Architecture Design
Vijay Kumar
Coming soon.
|
|
12:40
|
|
Lunch
|
|
02:00
|
|
Supervision of Heterogeneous Robot Teams with Application to Urban Search and Rescue
Oskar von Stryk and Karen Petersen (presenter)
As part of a larger research effort at TU Darmstadt on cooperation of autonomous
robots and sensor networks for monitoring and response applications, human
supervision of a heterogeneous team of autonomous robots is investigated.
The supervisor’s understanding of the current status of robots and mission is a
crucial factor for performance and success of supervisory control. Situation
awareness is usually applied as knowledge base for robot teleoperation. However,
it cannot be maintained for a whole team of robots in a critical application
like urban search and rescue (USAR). To overcome this deficiency the concept of
situation overview has been developed as suitable knowledge base for supervision
which is obtained by support from the robots through an event-based
communication concept. Tags and policies are used to adapt the amount of
communicated messages, allowing to reduce communicated data efficiently to the
required information.
Situation overview enables the supervisor to coordinate a heterogeneous robot
team at a high level of interaction. For this purpose, a method was developed
for interacting with the team's task allocation algorithm. This has the
advantage that inputs from the supervisor can be considered during the mission
without impairing coordination among robot team members. Intuitive commands from
the human are internally translated into changes to the tasks that compose the
mission, and into constraints of a mixed-integer linear program (MILP). Results
with simulated and real robots in USAR missions demonstrate that based on the
proposed concepts already few inputs from a human supervisor can lead to
substantial improvements of the solution quality. Furthermore, the additional
constraints to the MILP lead to significantly reduced computational efforts for
numerical solution compared to a fully autonomous problem approach.
|
|
02:30
|
|
Autonomy and Human Interaction in Multi-Robot Systems
Antonio Franchi and Paolo Robuffo Giordano
Coming soon.
|
|
Interactive session
Teasers: room Mombert
Interactive presentations: Interactive Sessions Area
|
|
03:00
|
|
Communication Integrated Control Architecture in Multirobot Syste (pdf)
Hakan Karaogouz, Haluk Bayram and H. Isil Bozma
|
|
03:05
|
|
A Communication-Bandwidth-Aware Hybrid Estimation Framework for Multi-robot Cooperative Localization (pdf)
Esha D. Nerurkar and Stergios I. Roumeliotis
|
|
03:10
|
|
A Software Suite for the Control and the Monitoring of Adaptive Robotic Ecologies (pdf)
Mauro Dragone, Maurizio Di Rocco, Federico Pecora and Alessandro Saffiotti
|
|
03:15
|
|
Interactive Demo: Haptic Remote Control of Multiple UAVs with Autonomous Cohesive Behavior (pdf)
Thomas Nestmeyer, Martin Riedel, Johannes Lächele, Simon Hartmann, Fiete Botschen, Paolo Robuffo Giordano and Antonio Franchi
|
|
03:20
|
|
Concurrent Bayesian Learners for Multi-Robot Patrolling Missions (pdf)
David Portugal, Micael S. Couceiro and Rui P. Rocha
|
|
03:30
|
|
Coffee Break
|
|
04:00
|
|
Interactive presentations and Teleoperation Demo (Interactive Sessions Area)
|
|
05:00
|
|
Transfer back to room Mombert
|
|
Invited Talk and Final Discussion
Room Mombert
|
|
05:10
|
|
Multi-Agent Coordination for Optimizing Communication in Heterogeneous Robot Networks
Daniela Rus and Stephanie Gil (presenter)
This talk concerns the development of controllers to position a team of aerial vehicles
in a configuration that optimizes proximity-based communication, providing support to a
team of client vehicles performing a collaborative task. We focus on heterogeneous
systems where only part of the system (the aerial vehicles) is assumed to be controlled
and the client vehicles are non-cooperative; moving over a priori unknown trajectories.
This allows client vehicles maximum flexibility to complete their independent task. We
propose a control approach that combines tools from the classic k-center and control
under uncertainty problems where agents’ positions optimally minimize a physically
motivated cost function. For real-time control we present a bounded-error sampling
approach based on "coresets" from computational geometry. The talk will include a
technical overview of the tools developed as well as performance results from several
implementations on heterogeneous hardware platforms including AscTec Hummingbird
quadrotors, KUKA Youbot robots, iRobot Creates, and an autonomous golf car.
|
|
05:40
|
|
Discussion
|
|
06:15
|
|
End of the Workshop
|
