Visual Fuzzy Control for Blimp Robot to Follow 3D Aerial Object

Abstract

This works presents a novel visual servoing system in order to follow a 3D aerial moving object by blimp robot and to estimate the metric distances between both of them. To realize the autonomous aerial target following, an efficient vision-based object detection and localization algorithm is proposed by using Speeded Up Robust Features technique and Inverse Perspective Mapping which allows the blimp robot to obtain a bird's eye view. The fuzzy control system is relies on the visual information given by the computer vision algorithm. The fuzzy sets model were introduced imperially based on possibilities distributions and frequency analysis of the empirical data. The system is focused on continuously following the aerial target and maintaining it within a fixed safe distance. The algorithm showing robustness against illumination changes, rotation invariance as well as size invariance. The results indicate that the proposed algorithm is suitable for complex control missions. © Springer International Publishing Switzerland 2014.