Low-Cost arduino based architecture for simulation and control of quadrotors

Abstract

Quadrotor or popularly known as quadcopter is an unmanned flying robotic platform that is driven by four fixed rotors. Quadrotors serve as a perfect flying platform in an anonymous and intricate environments eliminating the need to jeopardize human life in such hazardous situations. Apart from serving as a stable flight stage, these can do breathtaking acrobatic and complex maneuvers that are beyond the scope of the conventional aerial vehicles. There has been a tremendous surge in funding for research and development of such Unmanned Aerial Vehicles (UAV) as the investors are very well aware of the influence these robots have in military as well as civilian market. However the flip side to these overwhelming abilities the quadrotors have is the need for a robust and an efficient control model. The system designing process is a complex integration of aero-modeling and apt control techniques for stability. Multiple parameters need to be co-ordinated in real time for a smooth control of a multi rotor vehicle. This generally results in complex algorithms which increase computation requirements that can be costly. On the other hand cost reduction implies limitations in performance or poor control. This paper focuses to solve these challenges by dramatically simplifying the complexity of control algorithms without compromising on performance parameters of the quadrotor following strict cost constraints. We streamline the mathematical model simulations and the use of PD and PID control techniques by constraint based optimization that results in a simple, cost-effective and a robust control model. Through rigorous simulations and testing we not only obtain a faithful control model but also execute a self-leveling feature making the quadrotor smart. We also discuss the future possibilities for low-cost autonomous navigation in indoor environments.