The purpose of this paper is to describe optical flow‐based navigation of a very light fixed‐wing aircraft in flight between obstacles.
The optical flow information of two cameras mounted on the aircraft is used to detect the obstacle. It is assumed that the image processing has been completed and the optical flow vectors have been obtained beforehand. The optical flow is used to detect the obstacles and make a rapid turn manoeuvre for the aircraft.
It is shown that using the optical flow feedback by itself is unable to give a rapid turn to the aircraft and its rate should be employed into the control law. Six degree‐of‐freedom flight simulation showed that the proposed navigation and control strategy give satisfactory results in different flight environments like corridors with parallel and non‐parallel walls and in the L junctions. Simulations also showed that the aircraft flight velocity has little effect on collision avoidance performance.
This paper provides a theoretical framework to study the different parameters affect the obstacle detection and avoidance of an aircraft.
An analytical equation has been developed to relate the obstacle detection distance to the aircraft manoeuvrability parameters. In addition, an optical flow‐based controller also has been designed to provide rapid turn manoeuvres using the aileron control surface.
