基于扩张观测器的欠驱动船舶轨迹跟踪低频学习自适应动态面输出反馈控制

针对速度不可测的三自由度欠驱动船舶轨迹跟踪控制问题,考虑船舶存在模型参数不确定项以及外界环境干扰未知情况,提出一种基于扩张观测器的欠驱动船舶轨迹跟踪低频学习自适应动态面输出反馈控制策略.该策略构造扩张观测器估计船舶速度向量,利用神经网络算法逼近模型参数不确定项,然后采用动态面控制技术避免对虚拟控制律直接求导,简化控制律计算过程,并引入低频增益学习技术消除外界扰动导致控制信号产生高频振荡,最后选取李雅普诺夫函数证明该控制律能够保证船舶跟踪闭环系统中所有误差信号一致最终有界.仿真结果表明,本文所设计控制器对船舶模型参数不确定项及外界环境干扰具有较强的鲁棒性,能够实现对船舶轨迹的有效跟踪.     

离散非线性系统的迭代学习轨迹跟踪鲁棒算法优化及应用

针对一类存在随机输入状态扰动、输出扰动及系统初值与给定期望值不严格一致的离散非线性重复系统,提出了一种P型开闭环鲁棒迭代学习轨迹跟踪控制算法．基于λ范数理论证明了算法的严格鲁棒稳定性,并通过多目标函数性能指标优化P型开闭环迭代学习控制律的增益矩阵参数,保证了优化算法下系统输出期望轨迹跟踪误差的单调收敛性,达到提高学习算法收敛速度和跟踪精度的目的．最后应用于二维运动移动机器人的实例仿真,验证了本文算法的可行性和有效性．     

一类MISO 最小相位系统的执行器故障自适应容错控制

针对一类具有执行器卡死或/和变执行器故障的多输入单输出(MISO)非线性最小相位系统,提出一种自适应容错跟踪控制方案.采用自适应算法估计系统的不确定性,利用神经网络逼近执行器未知故障函数,以完成执行器组合故障状态下的跟踪控制.所设计的控制律不仅保证了闭环系统稳定,而且所有状态均有界,跟踪误差一致最终有界.仿真结果表明了所提出方法的有效性.     

存在环境约束的机器人自适应迭代学习控制

针对存在不确定性和外界干扰的受限机器人系统提出一种自适应迭代学习控制律.不确定性参数被估计在时间域内,同时重复性外界干扰在迭代域内得到补偿.通过引入饱和学习函数,保证了闭环系统所有信号有界.借助Lyapunov复合能量函数法,证明了系统渐进收敛到期望轨迹的同时,能够保证力跟踪误差有界可调.     

基于非线性增益递归滑模的船舶轨迹跟踪动态面自适应控制

针对三自由度全驱动船舶存在模型不确定和未知外部环境扰动的情况,设计出一种基于非线性增益递归滑模的船舶轨迹跟踪动态面自适应神经网络控制方法.该方法综合考虑船舶位置和速度误差之间关系设计递归滑模面,引入神经网络对船舶模型不确定部分进行逼近,设计带σ-修正泄露项的自适应律对神经网络逼近误差与外界环境扰动总和的界进行估计,并应用一种非线性增益函数构造动态面控制律,选取李雅普诺夫函数证明了该控制律能够保证轨迹跟踪闭环系统内所有信号的一致最终有界性.最后,基于一艘供给船进行仿真验证,结果表明,船舶轨迹跟踪响应速度快、精度高,所设计控制器对系统模型参数摄动及外界扰动具有较强的鲁棒性.     

基于观测器的网络化多智能体预测控制

研究含模型不确定性的刚性航天器输入受限时的姿态跟踪控制设计问题.针对修改的罗德里格斯姿态参数描述的航天器姿态跟踪动力学模型,基于一种有界非线性连续函数和修改的罗德里格斯姿态参数自身有界性,设计鲁棒自适应状态反馈受限控制器,不确定参数的自适应更新律可保证在线估计参数的有界性.通过所提出的输入受限控制设计方法给出输入受限幅值、期望轨迹上界、模型不确定性上界与控制器增益之间的定量关系,并采用李亚普诺夫方法证明通过选择合适的控制器参数可以保证闭环系统角速度误差渐近收敛到零,且姿态跟踪误差收敛到原点小邻域,同时保证控制量始终在预先给定的受限范围内.仿真结果验证了所设计的控制器可在输入受限的情况下完成控制目标并有效抑制模型的不确定性影响.     

基于神经网络观测器的船舶轨迹跟踪递归滑模 动态面输出反馈控制

针对三自由度全驱动船舶速度向量不可测问题,考虑船舶模型参数和外部环境扰动均未知的情况,提出一种基于神经网络观测器的船舶轨迹跟踪递归滑模动态面输出反馈控制方法.该方法设计神经网络自适应观测器估计船舶速度向量,且利用神经网络逼近模型参数不确定项,综合考虑船舶位置和速度误差之间关系构造递归滑模面,再采用动态面控制技术设计轨迹跟踪控制律和参数自适应律,并引入低频增益学习方法消除外界扰动导致的高频振荡控制信号.选取李雅普诺夫函数证明了该控制律能够保证轨迹跟踪闭环系统内所有信号的一致最终有界性.最后,基于一艘供给船进行仿真验证,结果表明,船舶轨迹跟踪响应速度快,所设计控制器对系统模型参数摄动及外界扰动具有较强的鲁棒性.     

输入输出受限船舶的轨迹跟踪自适应递归滑模控制

针对输入输出受限, 模型部分不确定和受到未知海洋干扰的全驱动船舶的轨迹跟踪问题, 提出一种基于时 变非对称障碍李雅普诺夫函数的最小参数自适应递归滑模控制策略. 该策略首先设计障碍李雅普诺夫函数约束船 舶轨迹在有限区域内, 利用最小参数法神经网络逼近模型不确定项, 降低系统的计算复杂度, 然后采用指令滤波器 对输入信号进行幅值约束, 同时避免对因反步法导致的微分爆炸问题, 综合考虑船舶位置以及速度误差间的关系设 计递归滑模控制律, 提高系统的鲁棒性, 采用双曲正切函数和Nussbaum函数补偿由输入饱和引起的非线性项, 提高 系统稳定性. 最后通过Lyapunov理论分析证明了全驱动船舶闭环系统中所有信号是一致最终有界的. 仿真结果表 明, 本文所设计的船舶轨迹跟踪控制方案能有效处理船舶模型不确定部分以及未知外界干扰的问题, 能够实现船舶 在输入受限的情况下在有限区域内航行并准确的跟踪期望轨迹, 具有较强的鲁棒性.     

基于S函数调节的非线性自适应动态面控制

针对一类控制增益未知的多输入多输出（MIMO）非线性系统,提出了一种基于神经网络的鲁棒自适应动态面控制方法．利用动态面控制解决反推法的计算膨胀问题;同时在参数自适应律中引入S（Sigmoid）函数,动态调节神经网络的收敛速度,解决了自适应初始阶段的抖振现象．利用李亚普诺夫稳定性定理,证明了闭环系统所有信号最终有界,系统的跟踪误差最终收敛到有界紧集内．仿真结果表明了该方法的有效性．     

执行器有故障的多输入单输出系统的自适应输出反馈控制

对一类控制增益符号未知且执行器有故障的输出反馈多输入单输出非线性系统,提出了一种后推容错控制方案.该方案在系统状态不可量测的情况下,利用Nussbaum函数处理符号未知的常数增益,并通过构造K-滤波器来估计了系统不可量测的状态.在容错控制器设计过程中,引入变能量函数来处理利用虚拟控制律所无法抵消的部分.与现有研宄成果相比,放宽了未知增益需要上下界均为已知的假设条件.最后,通过选取合适的李雅普诺夫函数,证明了闭环系统所有信号半全局一致终结有界,且跟踪误差收敛到原点的一个小邻域内.仿真结果表明了所提控制方法的有效性.     

A novel robust adaptive control algorithm with finite-time online parameter estimation of a humanoid robot arm

A novel robust adaptive control algorithm is proposed and implemented in real-time on two degrees-of-freedom (DOF) of the humanoid Bristol-Elumotion-Robotic-Torso II (BERT II) arm in joint-space. In addition to having a significant robustness property for the tracking, the algorithm also features a sliding-mode term based adaptive law that captures directly the parameter estimation error. An auxiliary filtered regression vector and filtered computed torque is introduced. This allows the definition of another auxiliary matrix, a filtered regression matrix, which facilitates the introduction of a sliding mode term into the adaptation law. Parameter error convergence to zero can be guaranteed within finite-time with a Persistent-Excitation (PE) condition or Sufficient Richness condition for the demand. The proposed scheme also exhibits robustness both in the tracking and parameter estimation errors to any bounded additive disturbance. This theoretical result is then exemplified for the BERT II robot arm in simulation and for experiments.     

未知时变惯量航天器自适应姿态跟踪容错控制

针对存在未知时变惯量不确定性、执行机构衰退故障和外部干扰力矩的非刚体航天器系统,研究了航天器自适应姿态跟踪容错控制问题,结合非线性鲁棒控制方法、自适应方法、容错控制理论和参数估计方法,提出了一种鲁棒自适应姿态跟踪容错控制器。所设计的控制器克服了执行器故障、惯量不确定性以及外界干扰对系统稳定性的影响,保证了航天器姿态及角速度能够跟踪上时变的期望状态,实现了跟踪误差系统最终一致有界稳定。最后通过数字仿真验证了所提方法的有效性,并且与已有方法进行了对比,说明了所提方法的优越性。     

输入饱和非线性系统的周期自适应补偿学习控制

针对一类输入饱和不确定Brunovsky标准型非线性时滞系统,提出一种周期自适应跟踪补偿学习算法. 利用信号置换思想重组系统,基于最小公倍周期函数变换,将时滞时变项和不确定项合并为辅助参数,进而设计周期自适应学习律估计该辅助量,并利用饱和补偿器逼近和补偿超出饱和限的部分,由此构成综合控制器,以保证系统状态对有界期望值的跟踪,解决了饱和输入周期系统的重复迭代学习控制问题. 最后通过构造Lyapunov-Krasovskii复合能量函数的差分,计算证明了系统跟踪误差的收敛性和闭环信号值的有界性. 常见耦合非线性机械臂系统的力矩控制仿真,进一步验证了该算法的有效性.     

Redundant Input Safety Tracking for Omnidirectional Rehabilitative Training Walker with Control Constraints

This study proposes a nonlinear safety tracking method based on redundant degrees of freedom with control constraints to protect against actuator faults that may occur in robotic rehabilitative walkers. A redundant input model with uniform actuator faults is constructed by separating the corresponding columns of the control matrix, and an adaptive robust control method is presented to deal with the separated term that is considered in relation to the extrinsic bounded interference that occurs on robotic walker systems. Based on the backstepping technique, an adjustable control law can be designed to maintain stability in terms of solving linear matrix inequality. The trajectory tracking error performance and the velocity tracking error performance are derived simultaneously. As an application, simulation results confirm the effectiveness of the proposed method and verify that the walker can provide safe sequential motions even when one wheel actuator fails.     

Robust tracking control for a class of flexible-joint time-delay robots using only position measurements

In this paper, robust tracking control is investigated for a class of uncertain flexible-joint robots with time delays and time-varying perturbations. By employing the Lyapunov--Krasovskii functional technique and backstepping design technique, a novel robust tracking control scheme using only position measurements is developed such that all the states and signals of the closed-loop flexible-joint time-delay robot system remain bounded and the tracking error can asymptotically converge to a small neighbourhood around the origin. By appropriately choosing the weighting gains in the Lyapunov–Krasovskii functionals, the circular phenomenon in the controller design is overcome. Due to suitably designing the velocity observer and the virtual control input, the link-side dynamics does not need to be incorporated into the actuator-side tracking error dynamics, and so the complexity in the backstepping design is avoided. Consequently, we can easily construct the Lyapunov–Krasovskii functionals, and, in turn, the robust tracking control scheme developed here is a linear time-varying controller and can be simply implemented. Simulation examples are provided to verify the effectiveness of the proposed control algorithm.     

Predefined-time control for Euler–Lagrange systems with model uncertainties and actuator faults

A novel adaptive predefined-time tracking control algorithm is proposed for the Euler–Lagrange systems (ELSs) with model uncertainties and actuator faults. Compared with traditional finite-time and fixed-time studies, the system output tracking error under the proposed predefined-time controller converges to a small neighborhood of zero in finite time, whose upper bound is exactly a design parameter in the control algorithm. For the uncertain model, radial-based function neural network (RBFNN) is utilized to approximate the continuous uncertain dynamics. To deal with the actuator faults, an adaptive control law is involved in the fault-tolerant controller. In order to achieve the predefined-time bounded, a novel predefined-time sliding mode surface is designed. It is proved that the tracking error vector trajectory of closed-loop system is semi-globally uniformly ultimately predefined-time bounded, and the upper bounds of both the system settling time and the corresponding output tracking error can be adjusted with a simple parameter. Simulation examples finally demonstrate the effectiveness of the proposed control algorithm.     

DISCRETE‐TIME OBSERVER‐BASED OUTPUT TRACKING CONTROLLER DESIGN FOR NONMINIMUM PHASE SYSTEMS WITH PERTURBATIONS

In this paper, an observer‐based output tracking controller for an SISO nonminimum phase discrete‐time system is proposed. When the disturbances between two consecutive sampling instances do not vary significantly, the observer algorithm can simultaneously estimate the system states and the unknown perturbation, and can render the estimation errors of system states and perturbation constrained in a small bounded region. The control law, including a feedforward term and a feedback input, can make the tracking error constrained in a small bounded region with guaranteed system stability. A numerical example is presented to demonstrate the applicability of the proposed control scheme.     

含有广义饱和函数的全局渐近稳定非线性PID控制

为解决传统饱和函数条件比较苛刻的问题,本文提出了一种饱和函数,并将其应用于线性PD（比例－微分）+非线性PI（比例－积分）控制律．应用李亚普诺夫稳定性定理和拉萨尔不变原理推导了非线性PID控制律全局渐近稳定条件．为提高非线性PID控制精度,以位置跟踪误差绝对值时间积分和关节力矩输出误差绝对值时间积分为目标函数,以全局渐近稳定性条件以及关节额定驱动力矩为约束条件,应用多目标遗传算法NSGA-II（non-dominated sorted genetic algorithm-II）进行非线性PID控制律参数整定．选择优化后位置跟踪误差绝对值时间积分最小的饱和函数,并研究含有该饱和函数的非线性PID控制律对模型不确定性、输入干扰力矩和噪声的鲁棒性．与传统PID控制律和含有传统饱和函数的非线性PID控制律相比,本文方法的位置跟踪精度分别提高了近2个数量级和1个数量级．提出的饱和函数在靠近平衡点处具有较强反作用,使误差快速收敛于平衡点,有助于提高非线性PID控制律的位置跟踪精度以及鲁棒性．     

带有干扰观测器的线控转向系统复合自适应神经网络控制

考虑车辆线控转向(SbW)系统存在不确定动态特性以及外界干扰影响.本文提出一种带有干扰观测器的复合自适应神经网络实现SbW系统的精确建模与稳定控制.首先,利用神经网络在线逼近系统不确定动态,避免控制器设计中使用到系统模型的先验知识.然后,结合系统的跟踪误差与建模误差提出一种新的复合自适应学习率来更新神经网络的权值,从而...     

Robust variable structure and switching-Σ adaptive control ofsingle-arm dynamics

In this paper, the variable structure control (VSC) and switching-σ adaptive laws are used to design a new robust controller for single-arm rigid manipulators in joint space. The controller is shown to be robust with respect to bounded disturbance. More particularly, a bound on the tracking error is determined and shown to be smaller than those resulting from the VSC) law proposed in Slotine and Li as well as the robust adaptive control law derived in Reed and Ioannou. The simulations also show that the proposed control law has better tracking precision performance than the VSC law and switching-σ adaptive control law. It can also suppress chattering and maintain good tracking precision even if actuator unmodeled dynamics are considered