Publication

Abstract : In this work, a six-degree-of-freedom (6-DOF) mathematical model of the Crazyflie nano quadcopter is developed to investigate its dynamic behavior and control performance. The model is developed using MATLAB/Simulink and is tested with a PID controller for different trajectory tracking scenarios. In order to evaluate the accuracy of the mathematical model developed, the authors compared its behaviour with that of a detailed physics model of the Crazyflie from the Gazebo robotics simulator. The same PID controller is applied to the Gazebo-based model via ROS-enabled Simulink integration, which ensured a fair comparison under the same trajectory-tracking conditions. The comparative study unveils differences in position and attitude responses, and motor command characteristics between the mathematical and physics-based models. The observed discrepancies in the mathematical model response are due to the model inaccuracy in capturing the actuator dynamics or aerodynamic drag effects. On the other hand, the Gazebo physics model integrates real-world actuator limitations and rotor aerodynamics, thus allowing controller performance assessment under more realistic operating environment. Nevertheless the mathematical model is good to be used for the initial stage controller development, tuning and the preliminary analysis. Novelty of the work is the control-centric integration of MATLAB/Simulink with ROS and the Gazebo physics engine. Such integration makes possible controller design and analysis, and incorporation of higher-level tasks such as multi-quadcopter deployment, and obstacle-aware navigation all within a unified MATLAB, ROS, Gazebo environment. An obstacle avoidance case study is presented to illustrate how the framework can be used for developing higher level tasks. The presented modeling and control framework can be a tool for pedagogy and research in autonomous aerial systems, contributing to technological innovation.