A Master of Science thesis in Mechatronics Engineering by Mohammad Khaled Salameh Al-Sharman entitled, "Auto Takeoff and Precision Landing using Integrated GPS/INS/Optical Flow Solution," submitted in January 2015. Thesis advisor is Dr. Mohammad Amin Al-Jarrah and co-advisor is Dr. Mamoun Abdelhafez. Available are both soft and hard copies of the thesis.
Auto takeoff and landing has been considered as the most challenging part in performing a flight with a high degree of autonomy. Hence, many researchers have addressed the problem of developing a precise auto takeoff and landing system for unmanned miniature helicopters. The work on enhancing autonomous takeoff and landing for unmanned aerial vehicles can be categorized into two groups. The first group works on designing a robust control algorithm in which the controller performs auto takeoff and auto landing. The second group focuses on utilizing sensors with high accuracy to get accurate state measurements. As a result, the performance of both the estimator and the controller would be improved. The present research addresses the use of optical flow sensors to augment the Global Positioning System/ Inertial measurement unit (GPS/INS) solution in the terminal phases of the flight (i.e., takeoff and landing). The GPS/INS unit has an internal estimator to estimate the vehicle's state vector which is not accurate enough to perform precision landing. The GPS/INS estimated position using Commercial-Off-The-Shelf COTS components is inaccurate with a few meters error, which is called the radius of uncertainty (ROU). To perform a precise landing, an optical flow sensor is used to augment the GPS/INS readout while performing the takeoff and landing phases of the flight. In this research, we use a sensor fusion algorithm between the optical flow sensor measurements of the location of a predefined pattern within the ROU of the GPS/INS and the GPS/INS location measurements. This estimator is used to output the helicopter's position and velocity during takeoff and landing. The proposed estimator has succeeded in performing an auto takeoff with a maximum error of 0.26 m and a precise landing with a maximum error of 0.27 m in Z-position. The novelty in this study is in the use of GPS/INS/Optical Flow fusion algorithm to perform a precise auto takeoff and landing for a small-scale helicopter. In addition, an accurate model for the OF sensor is used for developing the control laws for autonomous takeoff and landing for Vertical takeoff and landing vehicles (VTOL).