A Mosaic of Eyes

摘要

Autonomous navigation is a traditional research topic in intelligent robotics and vehicles, which requires a robot to perceive its environment through onboard sensors such as cameras or laser scanners, to enable it to drive to its goal. Most research to date has focused on the development of a large and smart brain to gain autonomous capability for robots. There are three fundamental questions to be answered by an autonomous mobile robot: 1) Where am I going? 2) Where am I? and 3) How do I get there? To answer these basic questions, a robot requires a massive spatial memory and considerable computational resources to accomplish perception, localization, path planning, and control. It is not yet possible to deliver the centralized intelligence required for our real-life applications, such as autonomous ground vehicles and wheelchairs in care centers. In fact, most autonomous robots try to mimic how humans navigate, interpreting images taken by cameras and then taking decisions accordingly. They may encounter the following difficulties.