This thesis report describes the process of developing an autonomous non-holonomic robot that can navigate and find targets in unknown environments. The goal was to create a system for a robot that was good enough to reach an honorable position in the SICK Robot Day competition 2009. The challenge of the competition was to build a autonomous vehicle that as fast as possible, found and marked number plates in a predefined order. To achieve this goal, we combined old and well known research with algorithms developed for this very purpose.The outcome of this project was an AI that could process sensor data and use it to plan smooth and safe routes to the destination. Behaviors for exploring unknown environments, detecting dead-ends, and avoiding collisions with other vehicles was also implemented. The report further explains and describes techniques such as SLAM and motion planning, which could have been applied to achieve a better and more accurate result.
Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:umu-48272 |
Date | January 2011 |
Creators | Vilson, Gunnar, Ängalid, Kim |
Publisher | Umeå universitet, Institutionen för datavetenskap, Umeå universitet, Institutionen för datavetenskap |
Source Sets | DiVA Archive at Upsalla University |
Language | English |
Detected Language | English |
Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
Format | application/pdf |
Rights | info:eu-repo/semantics/openAccess |
Relation | UMNAD ; 883 |
Page generated in 0.0012 seconds