Development of cooperation between flying robot, ground robot and ground station with fuzzy logic and image processing

The aerial robots represent an interested and rich area of research because they are very useful to perform complex tasks such as image acquisition of unreachable areas, localization and tracking targets. To develop blimp system that is appropriate in diversity scenarios, an intelligent control with high autonomy degree is required. Thus, the main challenge in context
of designing and navigation is the combination between intelligent control framework and small-light hardware components. Therefore, we design blimp robot based on embedded system; then, we present several fuzzy sets models that should deal with autonomous,navigation and visual tracking problems. These models are empirically designed by combining the possibilities distributions theory with fuzzy logic.
On the other hand, the use of cooperative robots system dramatically increases since it is appropriate solution in relation to performance, efficiency and reliability. The development of such cooperation system is one of the most demanding goals in artificial intelligence research and the most challenging issues here are the localization and tracking targets. Thus, this
dissertation addresses the problem of tracking robots in parallel with achieving the cooperative behavior based on computer vision system and artificial intelligent control to improve the performance and efficiency of such system. Moreover, the thesis considers the use of computer vision system that can be applied to navigation and tracking missions. This system has been
designed to detect and track either ground robots or 3D aerial object. In addition, considering use of wireless sensor network for estimation multi-targets locations and the distances between them is presented. Finally, the approaches that were developed in the thesis have been carried out and evaluated on the blimp robot. The extensive experiments have been done to demonstrate, analyze and validate the capabilities of such system as well as the properties of the proposed algorithms.