Vision-Based Recovery and Guidance for Small Unmanned Air Vehicles

Abstract

Current methods to enable the automatic recovery of unmanned air vehicles require that additional equipment be installed on the aircraft. This equipment in many cases can be costly, large, or require substantial power limiting these methods to large vehicles. This paper describes a technique for using the existing video camera and transmitter typically found on small unmanned air vehicles as the source for additional guidance information to facilitate automatic recovery of these small vehicles on board ships. The techique couples information from the vehicle autopilot with machine vision to generate guidance commands which the vehicle will attempt to track. A description of the machine vision algorithm and target along with the guidance law is presented. Simulation examples show that the performance of the guidance system in presence of wind and other sensor errors is sufficient to enable recovery of these vehicles with accuracy within one wingspan of error.