Multi-agents quadrotors UAVs for search and rescue operations

Abstract

This thesis addresses the problems of using a multi-agents quadrotors UAVs for search and rescue operations. In the first part the quadrotors components are modeled, and the interaction between them as long as the physical limitations is taken into consideration. The effects of the environmental constraints represented by a wind gust model were also examined. For the optimal trajectory generation and control problem, a new scheme based on a multi-layer optimal system is proposed. The designed controller aims to improve the robustness, the performances, and energy optimization. Once the quadrotors is modeled and controlled, the formation control problem is then investigated. A double loop control structure based on a non-linear optimal backstepping controller is applied. The quadrotors UAVs have to follow the desired path, hold their formation, avoid the obstacles and eliminate the effects of the external wind disturbances. In the other hand, SAR operations using UAVs are often time critical and dangerous missions. The challenge behind this work is that a single quadrotors could track the SAR strategies pattern, search for survivals with high probability of detection and then rescue them. Finally this thesis presents the development of a quadrotors multi-agents system, in which a swarm of agents coordinate and collaborate their actions in order to search for survivals in a disaster area and rescue them. Furthermore this work proposes new searching strategies that speed up the searching process, and cover the integrity of the searching area in order to detect all the possible targets.