Iman Sadeghzadeh

Active Fault Tolerant Control of a quadrotor UAV based on gainscheduled PID control

In this paper, an Active Fault-Tolerant Control (AFTC) technique is developed and applied to an unmanned quadrotor helicopter UAV (Unmanned Aerial Vehicle, known also as Qball-X4) with 6 degrees of freedom based on a Gain-Scheduled Proportional-Integral Derivative (GS-PID) control technique. For implementing such an AFTC system, a Fault-Detection and Diagnosis (FDD) block is essential and implemented to detect and identify the actuator fault. The FDD block is implemented based on the OptiTrack visual feedback for providing information needed by GS-PID to switch from one set of pre-tuned controller gains for normal (pre-fault) condition to another set of controller gains tuned for faulty (post-fault) conditions in the presence of an actuator fault in the Qball-X4 UAV. Finally, experimental testing results are presented to demonstrate the effectiveness of the proposed active fault-tolerant control strategy based on the GS-PID control technique.

Fault/Damage Tolerant Control of a Quadrotor Helicopter UAV using Model Reference Adaptive Control and Gain- Scheduled PID

In this paper, two useful approaches to Fault Tolerant Control (FTC) for a quadrotor helicopter Unmanned Aerial Vehicle (UAV) in the presence of fault(s) in one or more actuators during flight have been investigated and experimentally tested based on a Model Reference Adaptive Control (MRAC) and a Gain-Scheduled Proportional-IntegralDerivative (GS-PID) control. A Linear Quadratic Regulator (LQR) controller is used in cooperation with the MRAC and the GS-PID to control the pitch and roll attitudes of the helicopter. Unlike the MRAC, the GS-PID is used only to control the helicopter in height control mode. MRAC is used to control the helicopter in both height control as well as trajectory control. For damage tolerant control the MRAC is evaluated based on partial damage of one of propellers during flight. Finally, the experimental flight testing results of both controllers are presented for the fault tolerant control performance comparison in the presence of actuator faults in the quadrotor UAV.

Control Allocation and Re-allocation for a Modified Quadrotor Helicopter against Actuator Faults

Abstract This paper investigates the modification of a conventional quadrotor helicopter Unmanned Aerial Vehicle (UAV) by adding two actuators to the helicopter for increased hardware redundancy and enhanced Fault-Tolerant Control (FTC) capability. On one hand, this redundancy improves the helicopters safety margin and on the other hand, it extends the FTC methods that can be flying-tested on the quadrotor with more severe faults/damages. For illustration, control re-allocation is employed as a fault-tolerant control method in the presence of actuator faults. Some experimental results using the quadrotor testbed at the Networked Autonomous Vehicles (NAV) Lab of the Department of Mechanical and Industrial Engineering of Concordia University are given to illustrate the capabilities of the modified UAV system.

Payload Drop Application Using an Unmanned Quadrotor Helicopter Based on Gain-Scheduled PID and Model Predictive Control

Two useful control techniques are investigated and applied experimentally to an unmanned quadrotor helicopter for a practical and important scenario of using an Unmanned Aerial Vehicle (UAV) for dropping a payload in circumstances where search and rescue and delivery of supplies and goods is dangerous and difficult to reach environments such as forest or high building fires fighting, rescue in earthquake, flood and nuclear disaster situations. The two considered control techniques for such applications are the Gain-Scheduled Proportional-Integral-Derivative (GS-PID) control and the Model Predictive Control (MPC). Both the model-free (GS-PID) and model-based (MPC) algorithms show a very promising performance with application to taking-off, height holding, payload dropping, and landing periods in a payload dropping mission. Finally, both algorithms are successfully implemented on an unmanned quadrotor helicopter testbed (known as Qball-X4) available at the Networked Autonomous Vehicles Lab (NAVL) of Concordia University for payload dropping tests to illustrate the effectiveness and performance comparison of the two control techniques.

Optimal reliability design for over-actuated systems based on the MIT rule: Application to an octocopter helicopter testbed

This paper addresses the problem of optimal reliability in over-actuated systems. Overloading an actuator decreases its overall lifetime and reduces its average performance over a long time. Therefore, performance and reliability are two conflicting requirements. While appropriate reliability is related to average loads, good performance is related to fast response and sufficient loads generated by actuators. Actuator redundancy allows us to address both performance and reliability at the same time by properly allocating desired loads among redundant actuators. The main contribution of this paper is the on-line optimization of the overall plant reliability according to performance objective using an MIT (Massachusetts Institute of Technology) rule-based method. The effectiveness of the proposed method is illustrated through an experimental application to an octocopter helicopter testbed.

Payload drop application of unmanned quadrotor helicopter using gain-scheduled PID and model predictive control techniques

In this paper two useful control techniques are applied to a Quadrotor helicopter to control the height and while carrying a payload weighing one-fourth of its total weight as well as dropping the payload at a predetermined time. The first technique used in this paper is the Gain Scheduled Proportional Integral Derivative (GS-PID) control and the Model Predictive Control (MPC) algorithm is studied secondly. Both algorithms showed a very promising performance. Finally, both algorithms are successfully implemented on an unmanned quadrotor helicopter testbed (known as Qball-X4) available at the Networked Autonomous Vehicles Lab (NAVL) of Concordia University for height control to demonstrate effectiveness and stability of the two techniques. The results are presented and compared at the last section of this paper.

Control Allocation for a Modified Quadrotor Helicopter Based on Reliability Analysis

This paper considers the problem of control allocation for a modi ed quadrotor heli- copter unmanned system based on reliability analysis. As one of the objectives is to increase the overall systems reliability, the conventional quadrotor helicopter which is equipped with four actuators is rst upgraded by adding four additional actuators. Second, control allocation is applied to the upgraded system for a better redundancy management while considering information on actuator failure rates. With respect to such a reliability con- sideration, more control duties are allocated to the most reliable actuators and less control duties are allocated to the least reliable ones. Experimental ight tests show how the consideration of reliability analysis in the control allocation a ects the generated control inputs for a better control e orts distribution with consideration of a higher reliability in actuator components and also overall quadrotor system. Moreover, it is shown how the systems global reliability and consequently its availability are improved.

Fault Tolerant Control of a Quadrotor Helicopter Using Model Reference Adaptive Control

This paper proposes a useful approach to Fault Tolerant Control (FTC) based on the Model Reference Adaptive Control (MRAC) technique with application to a quadrotor helicopter Unmanned Aerial Vehicle (UAV) in hovering as well as trajectory tracking flight in order to control and keep the desired height and trajectory of the quadrotor helicopter in both normal conditions and in the presence of faults in one or more actuators. A Linear Quadratic Regulator (LQR) controller is used in cooperation with the MRAC to control the pitch and roll attitude of the helicopter. Three cases of fault are considered: 1) simulated fault in all the four actuators; 2) simulated fault in back and right motors; 3) a physical damage of 23% of one of the four propellers during autonomous flight. It can be seen from the test results that under the faulty and damage conditions MRAC controller provided a good response of the quadrotor UAV and result in safe landings of the quadrotor.Copyright

Linear Parameter Varying control synthesis: State feedback versus H∞ technique with application to quadrotor UA V

This paper focuses on the synthesis of Linear Parameter Varying (LPV) control based on two different LPV control structures. In the first structure the H∞ self-Gain-Scheduling (GS) control technique is used to obtain the LPV controller and in the second method, the composite quadratic Lyapunov function and the quadratic cost function are used to find the optimal state feedback gain. Finally, a six-degree of freedom quadrotor helicopter is used as an illustrative plant to compare the results of both LPV control structures.