Flying robots have been gaining popularity recently due to their ability to travel to places where ground robots cannot reach. In this work, we examine one of the popular yet important class of flying robot – a quadrotor system. It consists of four rotors attached with helicopter blades. The configuration allows easier and more agile positional control, and its dynamic model is much simpler than a conventional helicopter. In this work, we explore more advanced control method to achieve finite time (state) convergence for smoother trajectory control. We rely on the on-board sensor and the external positional (Vicon) tracking system to provide the full state information.
Our controller has a number of advantages. It was built from the prior work in Parra-Vega et. al., 2003. The most significant advantage is that the controller need minimal model information of the plant. The other controllers may require very precise aerodynamic model, but our control works well without such information. Furthermore, the control gains are very easy to tune. Here is the video of our successful hardware implementation:
Here is the earlier video of our calibration process and some simple results:
A. Sanchez, V. Parra-Vega, C. P. Tang, F. Oliva-Palomo, C. Izaguirre-Espinosa, “Continuous Reactive-based Position-Attitude Control of Quadrotors,” Proceedings of the American Control Conference, Montreal, Quebec, Canada, June 27-29, 2012.[Preprint]