Search for: All records

In quantitative ethnography (QE) studies which often involve large da-tasets that cannot be entirely hand-coded by human raters, researchers have used supervised machine learning approaches to develop automated classi-fiers. However, QE researchers are rightly concerned with the amount of human coding that may be required to develop classifiers that achieve the high levels of accuracy that QE studies typically require. In this study, we compare a neural network, a powerful traditional supervised learning ap-proach, with nCoder, an active learning technique commonly used in QE studies, to determine which technique requires the least human coding to produce a sufficiently accurate classifier. To do this, we constructed multi-ple training sets from a large dataset used in prior QE studies and designed a Monte Carlo simulation to test the performance of the two techniques sys-tematically. Our results show that nCoder can achieve high predictive accu-racy with significantly less human-coded data than a neural network.

The ability to build structures with autonomous robots using only found, minimally processed stones would be immensely useful, especially in remote areas. Assembly planning for dry-stacked structures, however, is difficult since both the state and action spaces are continuous, and stability is strongly affected by complex friction and contact constraints. We propose a planning algorithm for such assemblies that uses a physics simulator to find a small set of feasible poses and then evaluates them using a hierarchical filter. We carefully designed the heuristics for the filters to match our goal of building stable, free-standing walls. These plans are then executed open-loop with a robotic arm equipped with a wrist RGB-D camera. Experimental results show that the proposed planning algorithm can significantly improve the state of the art in robotic dry stacking.

Fluorescent portable monitoring systems provide real-time and on-site analysis of a sample solution, avoiding transportation delays and solution degradation. However, some applications, such as environmental monitoring of bodies of water with algae pollution, rely on the temperature control that off-site systems provide for adequate solution results. The goal of this research is the development of a temperature stabilization module for a portable fluorescent sensing platform, which is necessary to prevent inaccurate results. Using a Peltier device-based system, the module heats/cools a solution through digital-to-analog control of the current, using three surface-mounted temperature modules attached to a copper cuvette holder, which is directly attached to the Peltier device. This system utilizes an in-house algorithm for control, which effectively minimizes temperature overshooting when a change is enacted. Finally, with the use of a sample fluorescent dye, Rhodamine B, the system's controllability is highlighted through the monitoring of Rhodamine B's fluorescence emission decrease as the solution temperature increases.