Tracking Control of an Aerial Blimp Robot Based on Image Information

Abstract

An autonomous surveillance blimp system is provided with only one camera as a sensor. Motion segmentation is used to improve automatic detection and tracking of point features on a target object. Inverse optimal tracking control is applied to an aerial blimp which is one of the underactuated systems. The controller minimizes some meaningful cost dependent on the controller, which is not given in advance, and the designed system has a stability margin which guarantees robustness with respect to input uncertainties. Some experiments are performed by using an indoor blimp.