01-02-2013, 09:22 AM
Nonlinear and Fault-tolerant Control Techniques for a Quadrotor Unmanned Aerial Vehicle
1Nonlinear and Fault-tolerant.pdf (Size: 4.46 MB / Downloads: 48)
ABSTRACT
Unmanned Aerial Vehicles (UAVs) have become more and more popular, and how to control them has become crucial. Although there are many different control methods that can be applied to the control of UAVs, nonlinear control techniques are more practical since the nonlinear features of most UAVs. In this thesis, as the first main contribution, three widely used nonlinear control techniques including Feedback Linearization Control (FLC), Sliding Mode Control (SMC), and Backstepping Control (BSC) are discussed, investigated, and designed in details and flight-tested on a unique quadrotor UAV (Qball-X4) test-bed available at the Networked Autonomous Vehicles (NAV) Lab in Concordia University. Each of these three control algorithms has its own features. The advantages and disadvantages are revealed through both simulation and experimental tests. Sliding mode control is well known for its capability of handling uncertainty, and is expected to be a robust controller on Qball-X4 UAV. Feedback linearization control and backstepping control are considered a bit weaker than sliding mode control. A comparison of these three controllers is carried out in both theoretical analysis and experimental results under same fault-free flight conditions. Testing results and comparison show the different features of different control methods, and provide a view on how to choose controller under a specific condition. Besides, safety and reliability of UAVs have been and will always be a critical issue in the aviation industry.
Motivation
Due to the recent advances in sensing, communication, computing, and control technologies, unmanned vehicles have become vitally important in the engineering applications and our life. Among many other types of unmanned systems, there are two kinds of most widely investigated and developed unmanned vehicles, UAVs (Unmanned Aerial Vehicles) and UGVs (Unmanned Ground Vehicles). UGVs can be used as ground monitoring robots, and also as a replacement of human force. However there are certain limitations. Since UGVs can only be used on the ground, in some difficult terrain conditions, ground vehicles cannot reach the desired location.
Compared to UGVs, UAVs have greater capabilities. Aerial vehicles can be used to perform a large amount of tasks, such as monitoring forest fires and volcanic activities. They can also support military surveillance and air pollution control etc. There are different types of UAVs: fixed-wing airplanes, conventional helicopters and quadrotor helicopters. Fixed-wing airplanes require special runways to take off from. Both regular helicopters and quadrotors can overcome this flaw and are more flexible. Between these two types, quadrotor helicopters have four rotors more than regular ones, which means that they are more convenient and simpler to be built and to fly, and can possibly take more payload than the conventional helicopters.
Feedback Linearization Control
Feedback Linearization Control (FLC) is one of the most commonly used nonlinear control approaches and can be explained as linearization of a nonlinear system through feedback. Unlike the state feedback control, FLC can be applied directly to a nonlinear system without linear approximation. This approach transforms the states and the dynamics of the nonlinear system into linear ones. Therefore, after such a transformation, many linear control algorithms can also be used to make the control problem simpler.
Sliding Mode Control
Sliding Mode Control (SMC) is another advanced nonlinear control technique, with also strong robust abilities as the main feature of such a controller compared with the previous FLC algorithms. Sliding mode control has a sliding surface, which shows how the system converges. By adding a sign function, complexity can be reduced to a minimum so as to increase the stability of control system. For a rather complicated model with some uncertain parameters or dynamics, using controllers such as feedback linearization control which requires a precise model, will be inappropriate and inaccurate. Hence, the sliding mode control is chosen instead. SMC shows a strong capability of dealing with modeling errors, system uncertainties and external disturbances, as long as the sliding condition is satisfied.
Fault-Tolerant Control
The time of travelling to different places could be much shorter than before, due to the advanced aviation technology. However, if the system fails, the consequences also could be catastrophic. System faults occur rarely, but unpredictably and mostly suddenly. Therefore, Fault Tolerant Control (FTC) has become more important than ever.
A recent comprehensive overview on FTC is presented in [39] which classifies FTC strategies as Passive Fault Tolerant Control (PFTC), reconfigurable or Active Fault Tolerant Control (AFTC) which makes use of the information from the Fault Detection and Diagnosis (FDD) during operation of the FTC system (FTCS). Safety, reliability and reconfigurability analysis are also included in the paper to make a link for the currently individual research works between control engineering and safety engineering. Some key points in FTCS were also summarized in an early review paper [40] for summarizing control design methods developed up to 1997. Zhang [41] summarized a fault modeling method in FTCS for three different situations on sensor faults, actuator faults, and system dynamic faults. Fekih et al [42] presented a passive fault-tolerant control methodology using sliding surface and Lyapunov function to eliminate the pre-specified faults for the model of an F-18 aircraft. The results show that the design is effective. Reference [43] presented an integrated design procedure for fault detection, diagnosis, and reconfigurable control. A two-stage adaptive Kalman filter is used in fault detection and diagnosis scheme. The reconfigurable feedback and feedforward controllers are developed in details as well. Milhim et al [44] designed a gain scheduling based PID controller for FTC of a quadrotor UAV under simulation environment.
Thesis Contribution and Organization
In this thesis, the first goal is to design and implement three nonlinear controllers based on three different strategies: feedback linearization, sliding mode, and backstepping controls, and to test and evaluate the three algorithms in the real Qball-X4 quadrotor UAV test-bed available at Concordia University. The second goal is to develop and test a passive fault-tolerant control and an active fault-tolerant control strategy based on the developed sliding mode control technique for handling actuator faults and propeller damages in the UAV test-bed. To achieve the above goals, all these controllers are investigated and developed in details. Simulations are used to test if all the theoretically designed controllers function properly under different conditions, and experiments are the final proof of how they behave. Hence, each controller will be focused on practical usage, which means unnecessary assumptions will have to be reduced to the minimum in order to have a more realistic situation. Before experimental implementation of the controllers.
Summary
This chapter has reviewed the works that have been done on quadrotor using FLC, SMC, BSC or FTCS. All the studies have shown these three controllers are effective and
have good control performance. However, most of them are achieved in simulation environment, lack of practical proof on how well the controllers can behave in reality. Therefore, this thesis will redesign all theses controllers, FLC, SMC, BSC, and FTCS to control the quadrotor, Qball, and also implement the controllers in real environment to show the effectiveness of the control systems in practise as the final goal.