Javascript must be enabled for the correct page display

Threshold Detection

Alves, H. (2010) Threshold Detection. Bachelor's Thesis, Artificial Intelligence.

AI-BA-2010-HAlves.pdf - Published Version

Download (2MB) | Preview


When an autonomous robot drives around in a house, it will encounter obstacles. Some of these obstacles, like furniture or walls, should be avoided to prevent the robot being damaged. Other obstacles however, have to be passed when a robot has to leave a room, for example a doorstep or a tapestry. In this research we tried to determine which obstacles can be passed, and which cannot. To make this classification, we used a Time Delay Neural Network, an Echo State Network and a Naive Bayes Classifier. The robot we used was equipped with four infra-red sensors, two of which were pointed downwards so even small obstacles could be detected. Our research shows that, of the algorithms tested by us, The Naive Bayes Classifier gives the best results, and that the four used sensors might not always give enough information to make a correct classification.

Item Type: Thesis (Bachelor's Thesis)
Supervisor name: Jong, and Wiering, M. and Goeree, B.
Degree programme: Artificial Intelligence
Thesis type: Bachelor's Thesis
Language: English
Date Deposited: 15 Feb 2018 07:44
Last Modified: 02 May 2019 11:35

Actions (login required)

View Item View Item