CAD Modeling and Simulation of a Large Quadcopter with a Flexible Frame

Abstract

In this paper, we present the design issues, dynamic modeling, and feedback control problems for quadcopters having a large frame with mechanical flexibility. This situation occurs for example for the case of solar quadcopters that require a large structure to have onboard PV panels for the power supply and battery recharging. Flexibility may manifests itself as mechanical oscillations and static deflections, greatly complicating the motion control of a quadcopter platform. If the time to settle the oscillations is significant relative to the cycle time of the overall task, flexibility will be a major consideration in the flight control design, and a degradation of the overall expected system performance typically occurs. For quadcopters with flexible frames, it is difficult to use the Euler-Lagrange or Newtonian approach to derive the dynamic model as is done with small and rigid quadrotors. Many relevant factors that lead to the consideration of the distributed flexibility should be analyzed. To deal with this challenge, an optimal approach for sizing the frame and platform structure is followed. This is achieved by using a modern approach to computer modeling of quadcopters through the integration process of SolidWorks CAD modeling and MATLAB/Simulink environments. This is followed by identifying the resonant frequencies of the model using ANSYS Workbench. Initially, a SolidWorks model is created and then imported into MATLAB Simscape Multibody. Then, the transfer functions of the quadrotor model are derived using system identification. Subsequently, the equivalent transfer functions from the Simulink model are obtained and used for the PID controllers design. The validity of the quadcopter model represents an essential step to simulate the model before flying the quadcopter in the real world. It brings confidence that when the commands are run on the actual quadrotor, they will bring the same results. The design procedure for modeling a dynamical model of a large quadcopter and the tuning of its PID-based control strategies has been successfully implemented for a solar-quadcopter prototype, creating a reliable and effective automatic navigation and control system.