<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;">
The Autonomous Robots journal invites papers for a special issue <br>
on “Constrained decision-making in robotics: models, algorithms, <br>
and applications.” <br>
<br>
As the complexity of robotic tasks grows, robotic decision makers <br>
increasingly face the problem of trading off different objectives (for <br>
example, safety versus speed, or, in a reinforcement learning framework, <br>
balancing exploration versus exploitation for fast convergence). A natural <br>
framework for this class of problems is constrained decision-making, <br>
whereby a decision maker seeks to optimize a given cost function (often <br>
stochastic) while keeping other costs (usually involving risk assessments) <br>
below given bounds. Aspects of this framework have been addressed in <br>
isolation by the operations research and finance communities (for example, <br>
algorithms for constrained Markov decision processes and modeling of <br>
risk preferences), but the application of such a framework to the robotics <br>
domain is relatively new, fueled by application as diverse as safe <br>
autonomous driving, collision avoidance for unmanned aerial vehicles, <br>
and risk-aware learning for autonomous robots. <br>
<br>
Accordingly, this special issue aims at presenting the state of the art on <br>
the fast growing field of constrained decision-making in robotics. <br>
Specifically, it focuses on models, algorithms, and applications to solve <br>
constrained decision and planning problems for single and multiple <br>
robot systems. We invite submissions of original research papers addressing <br>
constrained decision making problems with an emphasis on theories and <br>
frameworks validated on robotic systems operating in the physical world. <br>
<br>
Topics of interest include, but are not limited to: <br>
<br>
- Modeling of constraints (in particular, risk) for robotic applications; <br>
- Algorithms for risk-aware decision making and learning for robotic systems, with a <br>
focus on online computation; <br>
- Chance-constrained robotic motion planning; <br>
- Hierarchical constrained decision making; <br>
- Applications of Constrained MDPs and Constrained POMDPs to robot planning; <br>
- Applications: ground, underwater, aerial, and space robots; <br>
- Benchmarks and performance metrics for constrained decision-making problems; <br>
- Verification and validation techniques for constrained decision-making problems. <br>
<br>
Guest editors:<br>
Stefano Carpin, University of California, Merced<br>
Marco Pavone, Stanford University<br>
<br>
CFP: <a href="http://static.springer.com/sgw/documents/1459402/application/pdf/CFP+Constrained+decision-making+-+deadline+10-15-2014.pdf">http://static.springer.com/sgw/documents/1459402/application/pdf/CFP+Constrained+decision-making+-+deadline+10-15-2014.pdf</a><br>
<br>
Important dates:<br>
October 15, 2014: Submission deadline<br>
January 15, 2015: First reviews completed<br>
February 15, 2015: Revised papers due<br>
March 30, 2015: Final decision<br>
<br>
Manuscript must be submitted to <a href="http://AURO.edmgr.com">http://AURO.edmgr.com</a>.
<div><br>
</div>
<div><br>
<div>Stefano Carpin, Ph.D<br>
Associate Professor<br>
School of Engineering<br>
University of California, Merced<br>
<a href="http://faculty.ucmerced.edu/scarpin">http://faculty.ucmerced.edu/scarpin</a><br>
<a href="http://robotics.ucmerced.edu/">http://robotics.ucmerced.edu</a><br>
</div>
<br>
</div>
</body>
</html>