Browsing by Author "Bergbreiter, Sarah"
Now showing 1 - 4 of 4
Results Per Page
Sort Options
Item Can You Do That Again? Time Series Consolidation as a Robust Method of Tailoring Gesture Recognition to Individual Users(MDPI, 2022-10-03) Dankovich, Louis J. IV; Vaughn-Cooke, Monifa; Bergbreiter, SarahRobust inter-session modeling of gestures is still an open learning challenge. A sleeve equipped with capacitive strap sensors was used to capture two gesture data sets from a convenience sample of eight subjects. Two pipelines were explored. In FILT a novel two-stage algorithm was introduced which uses an unsupervised learning algorithm to find samples representing gesture transitions and discards them prior to training and validating conventional models. In TSC a confusion matrix was used to automatically consolidate commonly confused class labels, resulting in a set of gestures tailored to an individual subject’s abilities. The inter-session testing accuracy using the Time Series Consolidation (TSC) method increased from a baseline inter-session average of 42.47 ± 3.83% to 93.02% ± 4.97% while retaining an average of 5.29 ± 0.46 out of the 11 possible gesture categories. These pipelines used classic machine learning algorithms which require relatively small amounts of data and computational power compared to deep learning solutions. These methods may also offer more flexibility in interface design for users suffering from handicaps limiting their manual dexterity or ability to reliably make gestures, and be possible to implement on edge devices with low computational power.Item REU: Examining vision sensor for cm-scale robots(2013-08) Frantz, Max; Penskiy, Ivan; Bergbreiter, SarahThe capabilities of the Centeye Stonyman Vision Chip were examined on the Parallax BOE-bot platform. This paper describes the tests that were performed to determine the sensor’s suitability for controlling navigation in an unknown environment. M. Srinivasan’s Image Interpolation Algorithm was used to compute 1D optical flow. Parameters that affect the optical flow’s signal-to-noise ratio were manipulated to acquire a sufficiently strong signal and navigation of a testing arena and collision avoidance were demonstrated.Item REU: Improving straight line travel in a miniature wheeled robot(2012-08-10) Gessler, Katie; Sabelhaus, Andrew; Bergbreiter, SarahThe TinyTeRP is a miniature robotics platform with modular sensing capabilities. Prior generations of the TinyTeRP have experienced various problems in assembly process, materials selection, and their fundamental design. These problems are addressed by choosing 3D printing as the new manufacturing method and steel wire for the new axle. The TinyTeRP’s ability to travel in a straight line using open loop control is studied. After 1.37 m of travel in the x direction, the TinyTeRP was as close as 4.69 cm to or as far as 31.9 cm from the ideal ending position (a straight line), indicating that open loop control is a poor method for controlling a straight line trajectory. Comparing data on the angle of the trajectory collected from position data from the vision table to data collected from the gyroscope indicated that the gyroscope tracks the robot’s angle of motion well. Hence, using the gyroscope for closed loop control of the TinyTeRP’s motion is possible.Item REU: RSSI-based rendezvous on the tiny terrestrial robotic platform (TinyTeRP)(2013-08) Jang, Han Beol; Villalba, Roberto D.; Paley, Derek; Bergbreiter, SarahThe TinyTeRP is a centimeter-scale, modular wheeled robotic platform developed for the study of swarm- ing or collective behavior. This paper presents the use of TinyTeRPs to implement collective recruitment and ren- dezvous to a fixed location using several RSSI-based gradient ascent algorithms. We also present a redesign of the wheel- based module with tank treads and a wider base, improving the robot’s mobility over uneven terrain and overall robustness. Lastly, we present improvements to the open source C libraries that allow users to easily implement high-level functions and closed-loop control on the TinyTeRP.