Quadcopter Modeling
The Pennsylvania State University - Dr. Puneet Singla
Solving a Quadcopter's Equations of Motion
April 2022
The objective of this project was to use rigid-body dynamics and Lagrangian mechanics to model the motion of a square-frame quadcopter. With six degrees of freedom, but only four independent rotors, the system is under-actuated and proved to be unable to be controlled properly.
Background
AERSP 304 was a junior-level System Dynamics and Control course that covered analytical and computational techniques for solving mathematical models of multiple degree of freedom engineering systems. This is achieved by exposure to topics such as model development, approximating systems with linear ODEs, linear feedback control systems, and stability analysis.
My Involvement
Given initial conditions and equations for translational acceleration; Euler angle change over time; body angular accelerations; roll, pitch and yaw moments; and angular velocities, ODE45 was utilized to numerically determine the motion of the quadcopter over a given simulation time. Since ODE45 requires first-order ODEs, several equations of motion had to be reduced from second-order. This, in addition to the fact that many variables show up across multiple equations, a state space was created. The velocities of the four motors changed over the simulation time, resulting in different rotations, so each time frame had to be calculated separately. The position, velocity, and rotation of the quadcopter was then graphed over time. The case in which the forces of the rotors were described by different equations as per an autopilot was also analyzed. This was used to help the unstable quadcopter stabilize over time.