Given the wide-ranging applications of quadcopters, their use has increased in recent years, pro‎mp‎ting numerous studies in this field. A key research area related to quadcopters involves examining cooperation between these devices within a multi-agent system. This research focuses on the design and control of a system comprising multiple quadcopters and a rigid load plane connected to the center of mass of the quadcopters by cables. The primary goal of this system is to control the force generated by the impact of a rigid object colliding with the load plate surface. First, the system kinematic and dynamic modeling is developed. The system is divided into two main sections: the carrier agents, which include the cables and quadcopters, and the load agent, which consists solely of the load plate. This division increases the system flexibility and scalability, allowing the number of carrier agents to be adjusted without increasing kinematic and dynamic complexities. The Newton-Euler method, which is used in multi-body system dynamics, is employed to model the carrier agent, enabling integrated analysis of the dynamic behavior of the cables and quadcopters. To model the impact on the load plate, an impulsive model is used, enhancing the ability to model sudden events such as impacts. Furthermore, the conservation of momentum principle is applied to calculate the force resulting from the impact. The resulting force is three-dimensional and acts in three different directions on the load agent and, consequently, on the quadcopters. The magnitude of this force depends on factors such as the velocity of the falling object, the velocity of the load at the moment of collision, the masses of both objects, and the coefficient of restitution. In this study, the impact is considered fully inelastic. The impact point is modeled in both central and non-central scenarios relative to the center of mass of the load agent to achieve a more comprehensive simulation of a 3D rigid object striking the load plate. In the next section of the study, controllers based on the LQR control method are designed. To use these controllers, the system equations of motion are linearized using feedback linearization. An LQR controller with an integrator is designed for each carrier agent. The system uses a rigid configuration for load transportation. Based on the defined reference path for the load, reference paths are established for each carrier agent and transmitted to the quadcopters to guide the load along its designated path. This distribution of control tasks enables decentralized control within the system. Finally, simulations of the system are conducted under various conditions to assess the performance of the designed controllers in handling impact forces and ensuring the load agent path-following. The simulation results demonstrate the high capability of the designed controllers in managing impact forces and maintaining system stability.

محمد دانش

ايمان ايزدي نجف آبادي , مرضيه كمالي