Robotic Exploration, Monitoring, and Information Collection:
Nonparametric Modeling, Information-based Control, and Planning under Uncertainty

A fundamental problem in robotics is efficient exploration and monitoring in uncertain environments. High-impact applications include aerial surveillance, ocean monitoring, urban search and rescue, space exploration, robotic surgery, and manipulation planning. Progress in this area requires solving three major sub-problems: (1) modeling the environment to maximize the accuracy of predictions based on limited information, (2) controlling a robot so that its motion maximally reduces its uncertainty about the environment, and (3) planning a motion for a robot to accomplish a specified task in an uncertain environment. These three problems are intimately linked, but research in these topic areas has largely proceeded independently. Following the success of our workshop on a subset of these topics at RSS12, this expanded workshop will bring together leading researchers in motion planning, information-based control, and nonparametric modeling to fuel an exchange of ideas between these diverse communities.

We solicit contributed presentations in all areas of nonparametric modeling, information-based control, and motion planning under uncertainty, including, but not limited to: planning in unknown environments, planning with uncertain motion and sensing models, planning and control with Bayesian nonparametric models, control with mutual information gradients, information surfing, informative path planning, active sensing, modeling using Gaussian processes and Dirichlet processes in the context of adaptive sampling and robust planning, modeling of nonstationary spatio-temporal dependencies in streaming sensor data, scalability in the face of large amounts of data, and implementation challenges for fielded robotic systems.