Autonomous robot navigation in unstructured outdoor environments is a challenging area of active research and is currently unsolved. The navigation task requires identifying safe, traversable paths that allow the robot to progress toward a goal while avoiding obstacles. Stereo is an effective tool in the near field, but used alone leads to a common failure mode in autonomous navigation in which suboptimal trajectories are followed due to nearsightedness, or the robot's inability to distinguish obstacles and safe terrain in the far field. This can be addressed through the use of machine learning methods to accomplish near-to-far learning, in which near-field terrain appearance and stereo readings are used to train models able to predict far-field terrain. This paper proposes to enhance existing, memoryless near-to-far learning approaches through the use of classifier ensembles that allow terrain models trained on data seen at different points in time to be preserved and referenced later. These stored models serve as memory, and we show that they can be leveraged for more effective far-field terrain classification on future images seen by the robot. A five-factor, full-factorial, repeated-measures experimental evaluation is performed on hand-labeled data sets taken directly from the problem domain. The experiments result in many statistically significant findings, the most important being that the proposed near-to-far Best-K Ensemble Algorithm, with appropriate parameter selection, outperforms the single-model, nonensemble baseline approach in far-field terrain classification. Several other findings that inform the use of near-to-far ensemble methods are also presented.

No abstract

S U M M A R YWe propose a new method to analyse seismic time-series and estimate the arrival times of seismic waves. Our approach combines two ingredients: the time-series are first lifted into a highdimensional space using time-delay embedding; the resulting phase space is then parametrized using a non-linear method based on the eigenvectors of the graph Laplacian. We validate our approach using a data set of seismic events that occurred in Idaho, Montana, Wyoming and Utah between 2005 and 2006. Our approach outperforms methods based on singular-spectrum analysis, wavelet analysis and short-term average/long-term average (STA/LTA).

A month-long quasi-experiment was conducted using a distributed team responsible for modeling, simulation, and analysis. Six experiments of three different time durations (short, medium, and long) were performed. The primary goal was to discover if synchronous collaboration capability through a particular application improved the ability of the team to form a common mental model of the analysis problem(s) and solution(s). The results indicated that such collaboration capability did improve the formation of common mental models, both in terms of time and quality (i.e., depth of understanding), and that the improvement did not vary by time duration. In addition, common mental models were generally formed by interaction around a shared graphical image, the progress of collaboration was not linear but episodic, and tasks that required drawing and conversing at the same time were difficult to do.