A layered approach to design autopilots

UAVs (Unmanned Aerial Vehicles) have been used for several applications which include re mote sensing and atmospheric data monitoring. These aircraft have had major improvements taking advantage of the miniaturization of electronic components and modules such as GPS receivers, digital cameras, wireless communication equipment and sensors. The main component in this class of vehicles is the autopilot, a module with the capability to control the aircraft and keep it in a predefined route. The absence of a human pilot allows a reduction of costs and enables its use for missions that would normally be too risky to be performed by a human being. The project ARARA (Autonomous and Radio-Assisted Reconnaissance Aircraft) aims the development of UAVs for agricultural and environmental monitoring. Its main objective is the replacement of conventional aircraft used to obtain aerial imagery for the monitoring of crops and areas under environmental stress. This paper describes the autopilot used in the aircraft of the ARARA project. It includes a control module and a navigation module, stacked in a layered organization allowing easy evolutive maintenance. This autopilot allows fully autonomous operation of the aircraft and the results so far indicate the usefulness of the adopted approach each time a software improvement is made to the autopilot system.