Autonomous takeoff and landing sequence of an RC helicopter (ATLas)

• Main Authors: Atienza, Patrick Von Angelo V., Lim, Hans Jefferson L., Lo, Patrick T., Mercadejas, Jan M.
• 格式: text
• 語言: English
• 出版: Animo Repository 2014
• 主題: Advances in Flight Control Systems; Artificial Intelligence and Robotics
• 在線閱讀: https://animorepository.dlsu.edu.ph/etd_bachelors/10138
• 機構: De La Salle University

A model (RC), single-rotor helicopter with six channels of control provides flight capabilities similar to real-world, cockpit-flown helicopters. With six degrees of freedom, the helicopter is both a highly versatile flyer, as well as a challenge to pilot. The Autonomous Takeoff and Landing Sequences of an RC Helicopter (ATLas) has developed a control system that can automate the takeoff and landing sequence, as well as maintain a low-altitutde hover of a 600-scale RC helicopter. These three areas of helicopter flight commonly involve vehicle operation in so-called Ground-effect region. The ground-effect region, which occurs up to an altitude of approximately the rotor diameter, is caused by rotor tip vortices being unable to form properly due to interference, and the rotor downwash being interrupted by the ground. This results in extra lift and speed, which, whilst being possibly seen as positive attributes, requires control system parameter returning. Without such compensation, the increased control sensitivity results in helicopter instability. It is therefore important that sufficiently accurate estimates of vehicle state (primarily: attitude and altitude) be available to the control system. These capabilities allow users to overcome the initial difficulties of RC helicopter operation by automating these procedures, effectively allowing the user to focus purely on flight. The ATLas uses an Arduino based controller to properly compensate for the ground effect both during takeoff and landing. A PI and PID controller was used in the implementation of this system, with the PI accepting the altitude error as input and converting it into the desired velocity, which is then fed into the PID controller that converts this into thrust. The system was able to demonstrate a takeoff, hover, and landing sequence, compensating for the ground effect without implementing complex aerodynamical models. For takeoff, the average rate of ascent was 0.4m/s, while for landing, the system produced an average descent rate of 0.7m/s.