Bo Han

An Application of Fuzzy PID Algorithm on Unmanned Helicopter Attitude Control

the design of an attitude controller for an unmanned helicopter is addressed. The controller proposed is based on fuzzy PID algorithm, and is used for stable and robust roll, pitch, and yaw control. The fuzzy logic rules are established through a combination of expert knowledge and experiment data which is obtained from a real unmanned helicopter platform. Both the simulation and the experimental results verify the validity of fuzzy PID algorithm which integrates the advantage of fuzzy control and PID control

A Hybrid Control Method for an Unmanned Scale Helicopter Attitude

A hybrid control method is designed and implemented on the attitude control of an unmanned scale helicopter. The hybrid controller is composed of a linear controller and a radial basis function neural network compensator. The compensator, which features in optimal approximation, can compensate the dynamical inversion error caused by the deficiency of full knowledge of helicopter dynamics. This system is applied to an unmanned scale helicopter mounted on a vertical platform. The application results show that the proposed controller cannot only cancel the dynamical error effectively but also improve tracking performance of the attitude control system. And the fine robustness for imprecise helicopter model is accounted for

An attitude estimation method for MAV based on the detection of vanishing point

For Micro Aerial vehicles (MAV) flying low in the city, an attitude estimation method based on machine vision is presented. First, the method detects the straight edge lines using Hough transform from the downward looking images captured. Then, a voting scheme that uses pixels as accumulators is applied to simultaneously evaluate the vertical edge lines of buildings and the vanishing point at which they converge. Finally, the attitude of UAV is calculated according to single view geometry. The experiment results indicate that the method detects the vanishing point reliably, and estimate the attitude precisely.

Intelligent Rotor Speed Controller for a Mini Autonomous Helicopter

Rotor is the most important component in mini autonomous helicopter. In order to simplify identification and control, it is desirable for mini autonomous helicopter to hold rotor speed constant, since varying rotor speed causes varying aerodynamics. An intelligent rotor speed controller has been developed, which adopts the control algorithms of feed forward and fuzzy tuned PI. The hardware design especially the tachometer design is described in this paper. To improve the performance of PI controller, a fuzzy logic supervisor has been proposed for tuning the gain parameters of PI controller online. The controller is then applied in a mini autonomous helicopter to verify its performance. The experimental results of real flights have shown that the intelligent rotor speed controller is superior in dynamic and steady state performance for its advantages of quick response and good robustness. Numerous flights have also proven its reliability

Landmark Recognition System for Mini Rotor Aerial Robot

Landmark recognition system is a very important part of mini rotor aerial robot. It recognizes landmarks, and supplies the navigation computer with location information of it. However, mini rotor aerial robots have low pay-load and are small in size. This means that the sensor like radar not fit in with this application and the small size and low power cost cameras may be the only sensors can be used. In this paper we focus on a landmark recognition system which uses an onboard camera as the sensor. Meanwhile, this paper presents an approach that is applied to real-time recognizing of landmark in video stream. This problem poses several challenges. Landmark in images captured from video stream by moving mini rotor robot vary in shape, size, scale and orientation. Unlike other approaches, this approach is based on classification algorithms. An extended set of Haar-like features is used, due to their simplicity. Landmarks are divided into several viewpoint categories according to their poses and each of these categories is processed by a cascade classifier which is trained by the Adaboosting algorithm. In the end, the experimental results verify the effectiveness of the system