Optical Navigation System for Robotics Application

摘要

Frequently, robots find themselves asking this question. Knowing your location?? And being able to navigate to other locations is extremely important for autonomous robots. The act of finding one's location against a map is known as localization. So how can a robot localize itself? The most common form of intelligent localization and navigation is to use a map, combined with sensor readings and some form of closed-loop motion feedback. But a robot needs to do so much more than just localizing itself. Often, we'd like our robots to be able to build their own maps, since map building by hand is tedious, boring, and error-prone. The field of robotics that studies localization and mapping is typically called SLAM (simultaneous localization and mapping) [1]. Robot cists around the globe have been working to find a solution to localization for more than 20 years; however, only in the past 4-5 years we have seen some promising results. In this work, we describe a first-of-a-kind, breakthrough technology for pose finding for localization that requires only one low-cost camera and a powerful ARM based micro-controller. Because of its low-cost and robust performance in realistic environments, this technology is particularly well-suited for use in consumer and commercial applications.