U.A.V. Path Planning
We put the ‘A’ in U.A.V.
We have the opportunity to do what has never been done before. The autonomous vehicle is one step before the flying car. Current technologies are designed to be everything for everybody, and are forced to function however we see fit. We propose a radically new approach which builds on the decades of research in Artificial Intelligence that is designed for a car from the ground up. Come October we set the standard.
Software development began in January of 2005, eight months following the formation of Oregon State University’s team. As a senior design project this package is handle local path planning, that is to find optimal paths between way-points. We have considerably expanded our involvement to include global path planning and sensor collection.
As part of our distinction from the various software groups, we have gone under the name of Team KnightRider (which entirely appropriate given the situation). We are all seniors at OSU studying Computer Science, two-thirds of those being Information Systems (database and business applications). This project exists because we strongly believe this is the correct solution. We are here to show the world that the limitations in this field do not exist because of software.
This project is designed and tested using various BSD operating systems with the actual controller being either FreeBSD or Darwin. Any POSIX should work. The majority of code is written in Objective-C. In order to compile Objective-C you will need to add the objc library to GCC. To ensure memory management with retain/release statements we use the GNUstep foundation library. Any database sever will do as a buffer between the sensors and the software, we chose SQLite3 for its excellent performance and memory footprint (be sure to save any database file on a RAM disk—hard drive and rocky terrain spell disaster).