无人直升机的姿态增强学习控制设计与验证

针对小型无人直升机的姿态控制问题,考虑到现有基于模型的控制方法对直升机动力学模型的先验信息依赖较大,以及未建模动态系统的影响等问题,设计了一种基于增强学习(RL)的飞行控制算法.仅利用直升机的在线飞行数据,补偿了未建模不确定性的影响.同时为了抑制外界扰动,提高系统的鲁棒性,设计了一种基于误差符号函数积分的鲁棒(RISE)控制算法.将两种算法结合,并利用基于Lyapunov分析的方法,证明了无人机姿态控制误差的半全局渐近收敛.最后在无人直升机飞行控制实验平台上,进行了姿态控制的实时实验验证.实验结果表明,本文提出的控制方法具有良好的控制效果,对系统不确定性和外界风扰具有良好鲁棒性.     

基于神经网络的小型无人直升机非线性鲁棒控制设计

针对小型无人直升机的姿态控制问题,为补偿系统参数不确定性和外界扰动的影响,设计一种连续的非线性鲁棒控制器.首先,利用神经网络在线估计系统不确定性,采用基于误差符号函数积分的鲁棒控制算法抑制外界扰动,同时补偿神经网络估计误差; 然后,利用基于Lyapunov函数的分析方法,证明所设计控制器的闭环稳定性,确保无人直升机姿态误差的半全局渐近收敛;最后,在无人直升机飞行实验平台上,进行无人机抗风扰控制实验.实验结果表明,所提出的控制方法具有良好的控制效果,对系统不确定性和外界扰动具有良好的鲁棒性.     

Trajectory control of a non-linear one-link flexible arm

The trajectory-tracking control problem is considered for a one-link flexible arm described by a non-linear model. Two meaningful system outputs are chosen; namely, the joint angle and the angular position of a suitable point along the link. The common goal is to stiffen the behaviour of the flexible link with respect to the chosen output. Based on the input-output inversion algorithm, a state-feedback control law is designed that enables exact tracking of any smooth trajectory specified for the output. In the closed loop an unobservable dynamics naturally arises, related to the variables describing the arm's distributed flexibility. Joint-based design is shown to be always stable, whereas in the link-point design the closed-loop dynamics may become unstable depending on the location of the output along the link. Open- versus closed-loop strategies are developed and compared. Extensive simulation results are included.     

Robust neural control for robotic manipulators

A robust neural control scheme for mechanical manipulators is presented. The design basically consists of an adaptive neural controller which implements a feedback linearization control law for a generic manipulator with unknown parameters, and a sliding-mode control which robustifies the design and compensates for the neural approximation errors. It is proved that the resulting closed-loop system is stable and that the trajectory-tracking control objective is achieved. Some simulation results are also provided to evaluate the design.     

An output feedback sliding mode control strategy for MIMO systems of arbitrary relative degree

This paper develops an output feedback sliding mode controller for multi‐input multi‐output (MIMO) systems of any relative degree. A minimal set of outputs and output derivatives are identified to determine an augmented system which is relative degree 1, and a robust sliding mode differentiator is presented as the means to construct the extended output signal. It is shown that the transmission zeroes of the original plant appear directly in the reduced order sliding mode dynamics relating to the augmented system. A super twisting control algorithm is shown to provide robust control performance. Simulation results for a rationalized helicopter model taken from the literature are used to demonstrate the attraction of the approach. Copyright © 2010 John Wiley & Sons, Ltd.     

Observer-based robust optimal control for helicopter with uncertainties and disturbances

Unknown model uncertainties and external disturbances widely exist in helicopter dynamics and bring adverse effects on control performance. Optimal control techniques have been extensively studied for helicopters, but these methods cannot effectively handle flight control problems since they are sensitive to uncertainties and disturbances. This paper proposes an observer-based robust optimal control scheme that enables a helicopter to fly optimally and reduce the influence of unknown model uncertainties and external disturbances. A control Lyapunov function (CLF) is firstly constructed using the backstepping method, then Sontag's formula is utilized to design an inverse optimal controller to stabilize the nominal system. Furthermore, it is stressed that the radial basis function (RBF) neural network is introduced to establish an observer with adaptive laws, approximating and compensating for the unknown model uncertainties and external disturbances to enhance the robustness of the closed-loop system. The uniform ultimate boundedness of the closed-loop system is ensured using the presented control approach via Lyapunov stability analysis. Finally, simulation results are presented to demonstrate the effectiveness of the proposed control strategy.     

Multidimensional Taylor network adaptive control for MIMO time‐varying uncertain nonlinear systems with noises

In this paper, an adaptive control approach based on the multidimensional Taylor network (MTN) is proposed here for the real‐time tracking control of multiple‐input–multiple‐output (MIMO) time‐varying uncertain nonlinear systems with noises. Two MTNs are used to formulate the optimum control and adaptive filtering approaches. The feed‐forward MTN controller (MTNC) is developed to realize the precise tracking control. The closed‐loop errors between the filtered outputs and expected values are directly chosen as the MTNC's inputs. A valid initial value selection scheme for the weights of the MTNC, which can ensure the initial stability of adaptive process, is introduced. The proposed MTNC can update its weights online according to errors caused by system's uncertain factors, based on stable learning rate. The resilient backpropagation algorithm and the adaptive variable step size algorithm via linear reinforcement are utilized to update the MTNC's weights. The MTN filter (MTNF) is developed to eliminate measurement noises and other stochastic factors. The proposed adaptive MTN filtering system possesses the distinctive properties of the Lyapunov theory–based adaptive filtering system and MTN. Lyapunov function of the filtering errors between the measured values and MTNF's outputs is defined. By properly choosing the weights update law in the Lyapunov sense, the MTNF's outputs can asymptotically converge to the desired signals. The design is independent of the stochastic properties of the input disturbances. Simulation of the MTN‐based control is conducted to test the effectiveness of the presented results.     

LPV model identification for gain scheduling control: An application to rotating stall and surge control problem

We approach the problem of identifying a nonlinear plant by parameterizing its dynamics as a linear parameter varying (LPV) model. The system under consideration is the Moore–Greitzer model which captures surge and stall phenomena in compressors. The control task is formulated as a problem of output regulation at various set points (stable and unstable) of the system under inputs and states constraints. We assume that inputs, outputs and scheduling parameters are measurable. It is worth pointing out that the adopted technique allows for identification of an LPV model's coefficients without the requirements of slow variations amongst set points. An example of combined identification, feedback control design and subsequent validation is presented.     

Model based feedforward control of an industrial glass feeder

This article presents the automation of set point changes of an industrial glass feeder in container glass production. A model is proposed consisting of multiple first order partial differential equations (PDEs). Based on the derived model a feedforward control approach is presented. The approach allows for the calculation of control inputs out of reference trajectories of the system outputs and is used to perform automated set point changes with short transition time. Finally, the approach is implemented at an industrial glass feeder. Measurement results from the Thüringer Behälterglas GmbH (Schleusingen, Germany) are included.     

Sliding mode control assisted by GPI observers for tracking tasks of a nonlinear multivariable Twin-Rotor aerodynamical system

This paper explores the techniques based on Generalized Proportional Integral Observers (GPI), under the approach of active disturbance rejection, applied to the control of multivariable nonlinear systems with the same number of inputs and outputs (square). As a central axis, a sliding mode control scheme assisted by GPI observers is proposed. These observers are responsible for estimating the disturbances associated with the dynamics of the sliding surfaces produced by non-linearities, not modeled elements, parameter uncertainty and external disturbances. This process allows the creation of sliding regimes under the point of view of decoupled control. As a case study, the position control of a small-scale 2-DOF helicopter affected by adverse operating conditions, such as unknown external disturbances and actuator faults, is considered. Finally, experimental results that validate the proposed control strategy and a comparison with a linear GPI observer-based active disturbance rejection control scheme are presented.     

一种具有噪声干扰的MIMO非线性时变系统自适应控制

针对含噪声多输入多输出不确定非线性时变系统,提出一种基于多维泰勒网(MTN)的自适应控制方案,其中两个MTN分别用来实现优化控制和非线性滤波.首先,提出多维泰勒网控制器(MTNC)以实现实时跟踪控制.将滤波输出与期望值之间的闭环误差作为MTNC的输入,根据系统不确定因素引起的误差,基于稳定的学习率,设计线性再励的自适应变步长算法以快速更新MTNC权值.其次,提出多维泰勒网滤波器(MTNF)以消除测量噪声.由于定义了测量值与MTNF输出之间误差的Lyapunov函数,自适应MTN滤波系统兼具基于Lyapunov理论的自适应滤波(LAF)和MTN的特有性质.最后,通过在Lyapunov意义下选取适当的权值更新律,可使MTNF输出渐近地收敛到期望信号,并证明了滤波器的收敛性和稳定性.仿真结果验证了所提出方案的有效性.     

Design of robot control systems

An expository discussion on practical aspects of the design problems of robot control systems is presented. The first section discusses the present status of robot control methodology based on the so-called 'teaching and playback' control scheme. It is pointed out that PTP (point to point) control is still central in practice because only a sort of pulse-incremental servo controller is implemented for each joint actuator in actual industrial robots. The second section points out that servo controllers of this kind perform approximately as a PD or PID controller, and demonstrates that such PD and PID control schemes can work well even if the robot dynamics are non-linear and have strong couplings between the joint variables. The third section deals with path-tracking and trajectory-tracking control problems when teaching by human operators is not possible. This is then followed by a final but substantial section on recent results on learning control and adaptive control. An example of learning control for robot motions is given and its potential applicability in robotic systems is discussed.     

Adaptive consensus control of nonlinear systems with unknown control directions

In this paper, we investigate the adaptive consensus control for a class of nonlinear systems with different unknown control directions where communications among the agents are represented by a directed graph. Based on the backstepping technique, a fully distributed adaptive control approach is proposed without using global information of the topology. Meanwhile, a novel Nussbaum-type function is proposed to address the consensus control with unknown control directions. It is proved that boundedness of all closed-loop signals and asymptotic consensus tracking for all the agents' outputs are ensured. In simulation studies, a numerical example is illustrated to show the effectiveness of the control scheme.     

Designing approach on trajectory-tracking control of mobile robot

Based on differential geometry theory, applying the dynamic extension approach of relative degree, the exact feedback linearization on the kinematic error model of mobile robot is realized. The trajectory-tracking controllers are designed by pole-assignment approach. When angle speed of mobile robot is permanently nonzero, the local asymptotically stable controller is designed. When angle speed of mobile robot is not permanently nonzero, the trajectory-tracking control strategy with globally tracking bound is given. The algorithm is simple and applied easily. Simulation results show their effectiveness.     

Discrete-time output feedback sliding mode control for spatial control of a large PHWR

This paper presents a novel approach to design a sliding mode control using a new computationally efficient formulation of Multirate Output Feedback (MROF), that applies to models in a special form in which some of the states of the system are directly available as outputs. The computation of the remaining states of the system requires matrices of lower dimension and thus the method is less susceptible to ill-conditioning. The proposed method has been successfully applied to the spatial control problem of a large Pressurized Heavy Water Reactor (PHWR), in which the outputs are the 14 zonal power levels. Also, the zonal power levels are selected as state variables along with other variables. Thereby, the PHWR model is perfectly suitable for the application of this new formulation. The non-linear model of PHWR including xenon and iodine dynamics is characterized by 70 state variables and 14 inputs and outputs each. Linear nodal model is obtained by linearizing the non-linear dynamic equations of the reactor about the full power operating point. The non-linear simulation results in representative transients produced by the proposed method are found to be superior to other methods.     

An experimental study of pilots' control characteristics for flight of an STOL aircraft in backside of drag curve at approach and landing

Abstract

In general, most vehicles can be modelled by a multi-variable system which has interactive variables. It can be clearly shown that there is an interactive response in an aircraft's velocity and altitude obtained by stick control and/or throttle control. In particular, if the flight conditions fall to backside of drag curve in the flight of an STOL aircraft at approach and landing then the ratio of drag variation to velocity change has a negative value (ΔD/Δu<0) and the system of motion presents a non-minimum phase. Therefore, the interaction between velocity and altitude response becomes so complicated that it affects to pilot's control actions and it may be difficult to control the STOL aircraft at approach and landing.

In this paper, experimental results of a pilot's ability to control the STOL aircraft are presented for a multi-variable manual control system using a fixed ground base simulator and the pilot's control ability is discussed for the flight of an STOL aircraft at backside of drag curve at approach and landing.     

Prescribed performance control approaches,applications and challenges: A comprehensive survey

As to control systems, transient performance is as important as steady-state performance. For some special dynamic systems, transient performance is a more prior index in comparison with the steady-state one. Prescribed performance control (PPC) has been proved to be a powerful tool that guarantees control system outputs/errors with desired transient performance as well as steady-state performance. The purpose of this paper is to give a comprehensive review on the latest developments of PPC theories and applications. The existing performance functions are classified into five different categories, and their features are comprehensively compared, providing a useful guidance for further applications. Then, the latest developments of PPC's applications in all kinds of control systems are recalled. Specially, the faced challenges and theoretical defects of PPC are discussed, which is expected to point out the further research direction for PPC.     

小型无人直升机的姿态与高度自适应反步控制

针对小型无人直升机的姿态与高度控制问题, 本文提出了一种基于反步法的自适应控制策略. 具体而言, 首先对小型无人直升机的运动学模型进行了等效变换, 使系统中未知参数满足线性参数化条件, 然后应用反步法设计了包含主旋翼挥舞模型的姿态与高度自适应控制器, 并借助Lyapunov方法和芭芭拉定理对闭环系统的稳定性进行了严格的数学分析. 最后, 对该控制器的性能进行了仿真验证, 结果表明在直升机质量和惯性矩阵存在不确定性(未知)的情况下, 该控制算法依然能够取得良好的控制效果.     

生物启发神经动力学模型的自治水下机器人反步跟踪控制

本文提出一种生物启发神经动力学模型的自治水下机器人(autonomous underwater vehicles,AUV)三维轨迹跟踪控制算法.将AUV在三维空间的运动分为水平面运动和垂直面运动,针对传统反步轨迹跟踪控制器出现的速度跳变问题,采用生物启发神经动力学模型,通过构造一种简单的中间虚拟变量,同时结合Lyapunov函数设计出轨迹跟踪控制律,使其控制效果在水平面和垂直面都能够达到全局渐近稳定并且有平滑连续的输出结果,克服AUV反步轨迹跟踪控制的速度跳变问题.仿真结果证明了所提控制律的有效性.     

A new trajectory control of a flexible-link robot based on a distributed-parameter dynamic model

The objective of this study is to develop a new trajectory-tracking control method, free from the so-called spillover instability, for a flexible-link robot. Based on a distributed-parameter dynamic model (a partial differential equation), a new moment-feedback trajectory-tracking control scheme is designed for a one-link flexible robot having a payload at the free end, in which zero geometric boundary conditions at the hub end and non-zero dynamic boundary conditions at the free end are taken into account. The proposed control is then extended to an adaptive scheme to cope with parametric uncertainties. The proposed control is stable for trajectory-tracking control and asymptotically stable at desired goal positions, which is proven by using the Lyapunov stability theorem. In addition, the proposed trajectory-tracking control, based on a distributed-parameter dynamic model, does not have the spillover problem. Furthermore, the control performance is guaranteed regardless of the magnitude of desired angle of rotation, which does not require any additional actuators such as piezoelectric actuators on the link and boundary force or moment actuators at the free end. The effectiveness of the proposed control has been shown by experiments.