A multi-range vision strategy for autonomous offroad navigation

Abstract

Vision-based navigation and obstacle detection must be sophisticated in order to perform well in complicated and diverse terrain, but that complexity comes at the expense of increased system latency between image capture and actuator signals. Increased latency, or a longer control loop, degrades the reactivity of the robot. We present a navigational framework that uses a self-supervised, learning-based obstacle detector without paying a price in latency and reactivity. A long-range obstacle detector uses online learning to accurately see paths and obstacles at ranges up to 30 meters, while a fast, short-range obstacle detector avoids obstacles at up to 5 meters. The learning-based long-range module is discussed in detail, and field experiments are described which demonstrate the success of the overall system.