Abstract: | Designing a mobile robot which can navigate a novel environment quickly and robustly is generally very difficult. Indeed, much robotic research is devoted to building an artificial creature that explores its world as effectively as a common rat or gerbil. This dissertation examines in-depth a well-documented computational model of the rat hippocampal navigation system, and discusses its implementation in software. It also presents experimental results with the model on both simulated and real-life robotic platforms, and compared performance with that of rodents and a more traditional robot-guidance algorithm.
|