Visual Sensing In Autonomous Robots

dc.contributor.authorKamemoto, Seth
dc.contributor.departmentElectrical Engineering
dc.date.accessioned2014-01-15T20:27:29Z
dc.date.available2014-01-15T20:27:29Z
dc.date.issued2014-01-15
dc.description.abstractThis paper defines a visual sensing system that can be used in autonomous robots. Autonomous robots require better decision-making skills to expand their abilities. As the tasks become more complex, choosing the appropriate decision becomes more difficult and requires the analysis of more environmental factors. An improved sensory system allows autonomous robots to obtain the additional environmental information they need to make the appropriate decision in a complex environment. To demonstrate the visual system, a digital camera is connected to a personal digital assistant (PDA), which handles the image processing and decision-making. This system finds the shortest path through a maze by taking a birds-eye picture of the maze. Visual sensing depends on image processing to convert raw image data into useful information regarding the environment. Thresholding is used to find the walls of the maze in the image. From these walls, a graph is constructed representing the squares in the maze and the paths between squares. Dijkstra's Algorithm is applied to the graph to find the shortest path.
dc.format.extent56 pages
dc.identifier.urihttp://hdl.handle.net/10125/32269
dc.publisherUniversity of Hawaii at Manoa
dc.rightsAll UHM Honors Projects are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission from the copyright owner.
dc.titleVisual Sensing In Autonomous Robots
dc.typeTerm Project
dc.type.dcmiText

Files

Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
Kamemoto_Seth.PDF
Size:
1.07 MB
Format:
Adobe Portable Document Format