Navigation of micro aerial vehicle in unknown environments

Abstract

This paper presents a methodology to navigate a micro aerial vehicle (MAV) flying through a cluttered and unknown environment. Due to the limited payload and power availability of the MAV platform, the guidance algorithm should be simple enough to be performed onboard with low computation and power consumption. The approach proposed here employs a miniature scanning laser rangefinder as a sensor to learn continuously about the surroundings, and a reflexive method is used to plan a collision-free path based on the information given by the sensor. To track the path, a simple difference controller is implemented on the MAV platform. The proposed methodology was tested in several typical scenarios. Simulation results show that the designed system provides an effective and robust solution to the navigation problem. The vehicle reached the destination point without collision to encountered obstacles.