Robotic systems have experienced exponential growth thanks to their incredible adaptability. Modern robots require an increasing level of autonomy, safety and reliability. This book addresses the challenges of increasing and ensuring reliability and safety of modern robotic and autonomous systems. The book provides an overview of research in this field to-date, and addresses advanced topics including fault diagnosis and fault-tolerant control, and the challenging technologies and applications in industrial robotics, robotic manipulators, mobile robots, and autonomous and semi-autonomous vehicles. Chapters cover the following topics: fault diagnosis and fault-tolerant control of unmanned aerial vehicles; control techniques to deal with the damage of a quadrotor propeller; observer-based LPV control design of quad-TRUAV under rotor-tilt axle stuck fault; an unknown input observer based framework for fault and icing detection and accommodations in overactuated unmanned aerial vehicles; actuator fault tolerance for a WAM-V catamaran with azimuth thrusters; fault-tolerant control of a service robot; distributed fault detection and isolation strategy for a team of cooperative mobile manipulators; nonlinear optimal control for aerial robotic manipulators; fault diagnosis and fault-tolerant control techniques for aircraft Systems; fault-tolerant trajectory tracking control of in-wheel motor vehicles with energy efficient steering and torque distribution; nullspace-based input reconfiguration architecture for over-actuated aerial vehicles; data-driven approaches to fault-tolerant control of industrial robotic systems.
Inspec keywords: manipulators; autonomous aerial vehicles; helicopters; nonlinear control systems; fault tolerance; mobile robots; optimal control
Other keywords: service robots; industrial robotic systems; nonlinear optimal control; fault diagnosis; fault tolerant control; unmanned aerial vehicles; robot manipulators; quadrotors; observer based control
Subjects: General topics in manufacturing and production engineering; Nonlinear control systems; Control technology and theory (production); Optimal control; Mobile robots; Robot and manipulator mechanics; Aerospace control; General and management topics
This chapter aims to design and develop an active fault-tolerant control (FTC) scheme for UAVs to ensure their safety and reliability. First, a cost-effective fault estimation scheme with a parallel bank of recurrent neural networks (RNNs) is proposed to accurately estimate actuator fault magnitude. Then, an adaptive sliding mode control (SMC) is proposed to guarantee system-tracking performance in the presence of model uncertainties without stimulating control chattering. Finally, a reconfigurable active FTC framework is established for a closed-loop unmanned quadrotor helicopter system by combing the developed fault estimation scheme with the proposed adaptive SMC.
This chapter can be considered a tutorial to guide the readers toward the implementation of active fault-tolerant control systems dealing with the damage of a propeller of an unmanned aerial vehicle. The addressed aerial device is a quadrotor with fixed propellers. The presented methodology also supposes to turn off the motor, the opposite of the broken one. In this way, a birotor configuration with fixed propellers is achieved. State-of-the-art approaches, using a PID-based controller and a backstepping controller, are presented in a tutorial form, thus neglecting the stability proofs, to leave room to a fast and concise description of the implementation procedures.
This chapter introduces the control method of the quad-TRUAV in the transition procedure under rotor -tilt axle stuck fault. By introducing virtual control inputs, the nonlinear model of the quad-TRUAV is transformed into the LPV form, and observer based LPV control is proposed to ensure the closed -loop stability of controlled plant. For applied control inputs, the inverse procedure is designed further, and the stability of transition procedure is ensured asymptotically in theory. After rotor -tilt axle stuck fault, the inverse procedure is reconstructed with degraded dynamics by ignoring velocity control. With the redesigned reference value of angle of attack, the cruise ability of TRUAV under stuck fault is still kept.
The use of unmanned aerial vehicles (UAVs) as support to operations in remote areas and harsh environments, such as marine operations in the Arctic, is becoming more and more crucial. These systems are expected to face very critical weather conditions, and for this reason they are naturally prone to the occurrence of icing. The formation of ice layers on airfoils decreases the lift and, simultaneously, increases the drag and the mass of the vehicle, thus requiring additional engine power and implying a premature stall angle. By adopting some tools from the control allocation framework, this chapter aims at presenting an unknown input observer (UIO) approach for fault and icing diagnosis and accommodation in UAVs equipped with a redundant suite of effectors and actuators. The chapter is structured as follows. At first, the UAV model and the basic setups are given, and then the fault and icing effects on the UAV dynamics are discussed. A short overview on the design ofUIOs is proposed, and a main section is dedicated to present the icing diagnosis and accommodation tasks. Furthermore, by means of appropriate scheduling parameters, the UIO-based diagnosis scheme can be extended to the nonlinear aircraft dynamics by considering the framework of linear parameter varying (LPV) systems. A collection of simulation examples concludes the chapter and illustrates the practical application of the proposed architecture.
In this chapter, we present a fault-tolerant control scheme for an over-actuated unmanned surface vehicle (USV) equipped with two azimuth thrusters. The scheme manages the most common actuator faults, i.e., loss of efficiency of the thruster and lock-in-place of the azimuth angle. The scheme is based on a three-layer architecture: a heuristic-based control policy for proper reference generation, a control law for the vehicle dynamics to achieve speed tracking of the generated reference, and a control allocation level for optimally redistributing the control efforts among the thrusters even in presence of actuator faults and failures. The control allocation and the control policies are the main focus of the chapter, since their reconfiguration capabilities allow tolerance with respect to actuator failures. On the contrary, the control law does not depend on the health status of the system. The scheme is then tested in simulation, using a nonlinear model of a wave adaptive modular vessel catamaran.
In this chapter, the problem of fault -tolerant control of a service robot is addressed. The proposed approach is based on using a fault estimation scheme based on a robust unknown -input observer (RUI0) that allows one to estimate the fault as well as the robot state. This fault estimation scheme is integrated with the control algorithm that is based on observer -based state -feedback control. After the fault occurrence, from the fault estimation, a feedforward control action is added to the feedback control action to compensate the fault effect. To cope with the robot non -linearity, its non-linear model is transformed into a Takagi-Sugeno (TS) model. Then, the state feedback and RUI0 are designed using a linear matrix inequality (LMI)-based approach considering a gain -scheduling scheme. To illustrate the proposed fault -tolerant scheme a mobile service robot TIAGo, developed by PAL Robotics, is used.
Applications involving multi-robot systems have been increasing day by day, since they allow one to accomplish complex tasks otherwise impossible for a single unit. Common control approaches for these robot systems are based on distributed architecture, where each robot computes its own control input, only based on local information from onboard sensors or received from its neighbor robots. This means that the failure of one or more agents might jeopardize the task execution. For this reason, fault detection and isolation (FDI) strategies become crucial to accomplish the assigned task in the aforementioned case as well. This chapter presents a distributed fault diagnosis architecture aimed at detecting failures in a team of robots working in tight cooperation. The proposed approach relies on a distributed observer-controller scheme, where each robot estimates the overall system state by means of a local observer, and it uses such an estimate to compute the local control input to achieve a specific task. The local observer is also used to define a set of residual vectors aimed at detecting and isolating faults occurring on any robot of the team, even if there is no direct communication. The approach is validated through experiments where a heterogeneous team of three robots perform a cooperative task.
The chapter proposes a nonlinear optimal control approach for aerial manipulators, that is, multi-DOF unmanned aerial vehicles (UAVs) that comprise a robotic arm with flexible joints. The method has been successfully tested so far on the control problem of several types of unmanned vehicles and the present chapter shows that it can also provide an optimal solution to the control problem of a 5-DOF aerial manipulator. To implement this control scheme, the state-space model of the aerial manipulator undergoes first approximate linearization around a temporary operating point, through the first-order Taylor series expansion and through the computation of the associated Jacobian matrices. To select the feedback gains of the H-infinity controller, an algebraic Riccati equation is repetitively solved at each time-step of the control method. The global stability and the robustness properties of the control loop are proven through Lyapunov analysis. Finally, to implement state estimation-based feedback control, the H-infinity Kalman filter is used as a robust state estimator.
This chapter analyses and discusses an active fault-tolerant control (FTC) system for avionic applications. The approach applies to an aircraft longitudinal autopilot in the presence of faults affecting the system actuators. The key point of the developed FTC scheme consists of its active feature, since the fault diagnosis module provides a robust and reliable estimation of the fault signals, which are thus compensated. The design technique, relying on a nonlinear geometric approach (NLGA), i.e. a differential geometry tool, allows one to achieve adaptive filters (AFs), which provides both disturbance-decoupled fault estimates and fault isolation features. The chapter also shows how these fault estimates are thus exploited for control accommodation. In particular, by means of this NLGA, it is possible to obtain fault reconstructions decoupled from the wind components of the considered aircraft application. It is shown how this solution provides very good robustness characteristics and performances to the overall system. Finally, the effectiveness of the considered scheme is analysed by means of a high fidelity flight simulator, in different conditions and in the presence of actuator faults, turbulence, measurement noise and modelling errors.
In this chapter a reconfigurable trajectory -tracking control design method has been presented for autonomous in-wheel electric vehicles with independently controlled hub motors and the steer-by-wire steering system. The high-level control reconfiguration has been implemented through the design of a scheduling variable using the LPV framework in order to deal with fault events, while in normal operating conditions the objective of the reconfiguration is to maximize battery SOC, thus enhancing the range of the in-wheel electric vehicle. The energy optimal control reconfiguration has been designed based on the results of preliminary simulations with a high-fidelity vehicle and electrical models on different road conditions. Finally, the efficiency of the proposed method has been demonstrated in a real-data CarSim simulation, showing significant energy saving by the proposed method
This chapter focuses on the reconfiguration task in the case of actuator failures. The approach considered here is the control input reallocation, where the aim is to compensate the actuator failures by reconfiguring the remaining fl ight control surfaces such that the performance degradation caused by the failure is as small as possible. One possible approach to solve this problem is based on the nullspace (or kernel) of the aircraft dynamics. These algorithms can be applied only if control input redundancy is available in the system. As in aerospace applications this is often the case [9], this approach is a promising method for fault -tolerant fl ight control design.
Robotic systems in the modern industries are working in increasingly complex environment. The unexpected external disturbances and the abrupt change of working condition bring great challenges to the control of such systems. For some safetycritical applications, it is required to continuously stabilize the systems under faulty condition (including malfunctions and system failures). In this chapter, a data-driven fault-tolerant control (FTC) framework is introduced. As a semi-supervised machine learning technique that can learn from the rewards from external environment, reinforcement learning (RL) aims to maximize the long-term returns by, for instance, maintaining a value function. Based on the approximated value function, optimal control law can be derived. When using RL, although the convergence of the value function has been proved for some simple systems, such as the linear time invariant system, the internal stability of the close-loop system still remains unguaranteed. To deal with this problem, a novel framework for FTC system design based on Youla parameterization is introduced. It can be implemented in the modular and plug-and play manner. It should be noted that in this work RL acts as a supervising module that calculates optimized Youla matrix Qc in the design phase. Simulation results on a wheeled robot are provided to show the performance of the proposed approach in the continuously stabilizing framework. Some open questions and future work are summarized at the end of this chapter.
The aim of this book was to provide possible solutions to the problem of fault diagnosis and fault-tolerant control (FTC) of robotic and autonomous systems. A total of 44 international authors contributed to the book in the form of 12 chapters, describing both theoretical findings and challenging applications.