Function-based design process for an intelligent ground Function-based design process for an intelligent ground

摘要

An engineering design framework for an autonomousnground vehicle vision system is discussed. We present both the con-nceptual and physical design by following the design process, devel-nopment and testing of an intelligent ground vehicle vision systemcon-nstructed for the 2008 Intelligent Ground Vehicle Competition. Duringnconceptual design, the requirements for the vision system are ex-nplored via functional and process analysis considering the flows intonthe vehicle and the transformations of those flows. The conceptualndesign phase concludes with a vision system design that is modularnin both hardware and software and is based on a laser range findernand camera for visual perception. During physical design, prototypesnare developed and tested independently, following the modular in-nterfaces identified during conceptual design. Prototype models, oncenfunctional, are implemented into the final design. The final vision sys-ntem design uses a ray-casting algorithm to process camera and lasernrange finder data and identify potential paths. The ray-casting algo-nrithm is a single thread of the robot’s multithreaded application. Othernthreads control motion, provide feedback, and process sensory data.nOnce integrated, both hardware and software testing are performednon the robot. We discuss the robot’s performance and the lessonsnlearned.