基于深度强化学习的欠驱动仿生机器鳗鱼控制研究

水下仿生机器人具有高效率、高机动性、低噪声等优点,针对仿生机器鳗鱼存在设计复杂、控制难度大等问题,该文提出了一种新型欠驱动机器鳗鱼的控制方法。首先,基于主动加被动的仿生机构推进原理,设计了两段主动体与两段被动顺从体相结合的机器鳗鱼仿生机构;然后,在仿真环境中进行建模,利用深度强化学习算法进行数据收集和训练,选择表现良好的神经网络在仿真环境中进行控制测试,从而得到机器鳗鱼的控制函数;最后,通过对比实验,验证了该文设计方法的可行性以及控制函数的有效性,实现了对机器鳗鱼的控制。     

基于干扰观测器的机械臂广义模型预测轨迹跟踪控制

为实现对多自由度机械臂关节运动精确轨迹跟踪,提出一种基于非线性干扰观测器的广义模型预测轨迹跟踪控制方法。针对机械臂轨迹跟踪运动学子系统,采用广义预测控制（Generalized Predictive Control,GPC）方法设计期望的虚拟关节角速度。对于机械臂轨迹跟踪动力学子系统,考虑机械臂的参数不确定性和未知外界扰动,利用GPC方法设计关节力矩控制输入,基于非线性干扰观测器方法实时估计和补偿系统模型中的不确定性。在李雅普诺夫稳定性理论框架下证明了机械臂关节角位置和角速度的跟踪误差最终收敛于零的小邻域。数值仿真验证了所提出控制方法的有效性和优越性。     

基于一阶不确定项估计律的机械臂鲁棒控制研究

针对多自由度机械臂控制系统的模型参数误差、关节摩擦力以及外部输入扰动等不确定项,设计了一类一阶误差估计律;结合基于机构动力学名义模型的输入输出反馈线性化控制算法,对六自由度刚性机械臂的时变轨迹跟踪控制进行了研究,理论上证明了设计的鲁棒控制器是全局渐进稳定的.仿真结果表明该控制策略对系统的各类不确定项具有很好的鲁棒性,能够实现高精度的轨迹跟踪控制.     

Robust fuzzy tracking control for robotic manipulators

In this paper, a stable adaptive fuzzy-based tracking control is developed for robot systems with parameter uncertainties and external disturbance. First, a fuzzy logic system is introduced to approximate the unknown robotic dynamics by using adaptive algorithm. Next, the effect of system uncertainties and external disturbance is removed by employing an integral sliding mode control algorithm. Consequently, a hybrid fuzzy adaptive robust controller is developed such that the resulting closed-loop robot system is stable and the trajectory tracking performance is guaranteed. The proposed controller is appropriate for the robust tracking of robotic systems with system uncertainties. The validity of the control scheme is shown by computer simulation of a two-link robotic manipulator.     

Optimal control of manipulators with any number of passive joints

This article discusses the problem of controlling robot manipulators with passive joints, when the number of passive joints is larger than the number of active joints. Assuming that brakes and position sensors are available at each passive joint, we investigate the following issues: (1) what is a sufficient condition for controllability of the passive joints via dynamic coupling with the active joints and how can we quantify the controllability at a given configuration; (2) what is the optimal control and locking sequence of the passive joints; and (3) how can we control both passive and active joints to an equilibrium point in joint space. We propose an optimal control method and demonstrate its validity with both simulation and experimental results. The work presented here is significant because it provides a better understanding and a guideline for utilizing manipulators with passive joints for energy efficiency and fault-tolerant design in applications such as space robotics, hyperredundant robots, and sport mechanics. © 1998 John Wiley & Sons, Inc. 15: 115–129, 1998     

考虑柔性关节的机械臂自适应运动控制策略研究

具有柔性关节的轻型机械臂因其自重轻、响应迅速、操作灵活等优点,取得了广泛应用;针对具有柔性关节的机械臂系统的关节空间轨迹跟踪控制系统动力学参数不精确的问题,提出一种结合滑模变结构设计的自适应控制器算法;通过自适应控制的思想对系统动力学参数进行在线辨识,并采用Lyapunov方法证明了闭环系统的稳定性;仿真结果表明,该控制策略保证了机械臂系统对期望轨迹的快速跟踪,具有良好的跟踪精度,系统具有稳定性。     

A computational framework for simulation of Underwater Robotic Vehicle systems

This paper presents a computational framework for efficiently simulating the dynamics and hydrodynamics of Underwater Robotic Vehicle (URV) systems. Through the use of object-oriented mechanisms, a very general yet efficient version of the Articulated-Body (AB) algorithm has been implemented. An efficient solution to branching within chains is developed in the paper so that the algorithm can be used to compute the dynamics for the entire class of open-chain, tree-structured mechanisms. By including compliant contacts with the environment, most closed-chain systems can also be modeled. URV systems with an extended set of topologies can be simulated including proposed underwater walking machines with intra-body powered articulations. Using the encapsulation inherent in C++, the hydrodynamics code has been confined to a single class, thereby explicitly defining this framework and providing an environment for readily implementing desired hydrodynamics algorithms. Resulting simulations are very efficient and can be used in a number of applications both in the development and use of URV systems.     

Hierarchical intelligent control for robotic motion

This paper presents a new scheme for intelligent control of robotic manipulators. This scheme is a hierarchically integrated approach to neuromorphic and symbolic control of robotic manipulators. This includes an applied neural network for servo control and knowledge-based approximation. The neural network in the servo control level is based on a numerical manipulation, while the knowledge based part is symbolic manipulation. The knowledge base part develops control strategies symbolically for the servo level. The neural network compensates for vagueness in the control strategies, nonlinearities of the system and uncertainties in its environment using neuromorphic control.     

Adaptive control of nonlinear systems: A case study of underwater robotic systems

The problem of controlling underwater mobile robots in 6 degrees of freedom (DOF) is addressed. Underwater mobile robots where the number of thrusters and control surfaces exceeds the number of controllable DOF are considered in detail. Unlike robotic manipulators underwater mobile robots should include a velocity dependent thruster configuration matrix B( q ), which modifies the standard manipulator equation to: Mq + C( q ) q + g(x) = B( q ) u where x = J( x ) q . Uncertainties in the thruster configuration matrix due to unmodeled nonlinearities and partly known thruster characteristics are modeled as multiplicative input uncertainty. This article proposes two methods to compensate for the model uncertainties: (1) an adaptive passivity-based control scheme and (2) deriving a hybrid (adaptive and sliding) controller. The hybrid controller combines the adaptive scheme where M, C, and g are estimated on-line with a switching term added to the controller to compensate for uncertainties in the input matrix B. Global stability is ensured by applying Barbalat's Lyapunov-like lemma. The hybrid controller is simulated for the horizontal motion of the Norwegian Experimental Remotely Operated Vehicle (NEROV).     

Nonlinear disturbance observer design for robotic manipulators

Robotic manipulators are highly nonlinear and coupled systems that are subject to different types of disturbances such as joint frictions, unknown payloads, varying contact points, and unmodeled dynamics. These disturbances, when unaccounted for, adversely affect the performance of the manipulator. Employing a disturbance observer is a common method to reject such disturbances. In addition to disturbance rejection, disturbance observers can be used in force control applications. Recently, research has been done regarding the design of nonlinear disturbance observers (NLDOs) for robotic manipulators. In spite of good results in terms of disturbance tracking, the previously designed nonlinear disturbance observers can merely be used for planar serial manipulators with revolute joints [Chen, W. H., Ballance, D. J., Gawthorp, P. J., O'Reilly, J. (2000). A nonlinear disturbance observer for robotic manipulators. IEEE Transactions on Industrial Electronics, 47 (August (4)), 932–938; Nikoobin, A., Haghighi, R. (2009). Lyapunov-based nonlinear disturbance observer for serial n-link manipulators. Journal of Intelligent & Robotic Systems, 55 (July (2–3)), 135–153]. In this paper, a general systematic approach is proposed to solve the disturbance observer design problem for robotic manipulators without restrictions on the number of degrees-of-freedom (DOFs), the types of joints, or the manipulator configuration. Moreover, this design method does not need the exact dynamic model of the serial robotic manipulator. This method also unifies the previously proposed linear and nonlinear disturbance observers in a general framework. Simulations are presented for a 4-DOF SCARA manipulator to show the effectiveness of the proposed disturbance observer design method. Experimental results using a PHANToM Omni haptic device further illustrate the effectiveness of the design method.     

Intelligent rotational direction control of passive robotic joint with embedded sensors

Passive compliant joints with springs and dampers ensure a smooth contact with the surroundings, especially if robots are in contact with humans, but the passive compliant joints cannot determine precisely the position of the members of the joint or direction of the collision force. In this paper was proposed the structure of a passive compliant robotic joint with conductive silicone rubber elements as internal embedded sensors. The sensors can operate as absorbers of excessive external collision force instead of springs and dampers and can be used for some measurements. Therefore, this joint presents one type of safe robotic mechanisms with an internally measuring system. The sensors were made by press-curing from carbon-black filled silicone rubber which is an electro active material. Various compression tests of the sensors were done. The main task of this study is to investigate the application of a control algorithm for detecting the direction of the robotic joint angular rotation when subjected to an external collision force. Soft computing methodology, adaptive neuro fuzzy inference strategy (ANFIS), was used for the controller development. The simulation results presented in this paper show the effectiveness of the developed method.     

Modified Transpose Effective Jacobian Law for Control of Underactuated Manipulators

Underactuated manipulators consist of active and passive joints, and developing a control technique that can manage such systems is an attractive, challenging problem. Most works in this area present model-based control laws that require a full dynamics model, and are consequently affected from uncertainties and time delays due to massive computations. Non-model-based control approaches provide an efficient alternative for practical implementation. The Modified Transpose Jacobian (MTJ) algorithm is one of these controllers that has been recently proposed for fully actuated manipulators with a square matrix Jacobian. Based on an approximated feedback linearization approach, the MTJ does not need a priori knowledge of the plant dynamics. In this paper, this scheme is extended to the complicated control problem of underactuated robots in Cartesian space. To this end, the notion of the Transpose Effective Jacobian (TEJ) is presented and so the proposed algorithm is called the Modified TEJ (MTEJ) algorithm. The MTEJ control law employs stored data of the control command in the previous time step, as a learning tool to yield an improved performance. Therefore, the proposed law needs just to a portion of mass matrix that corresponds to passive joint(s), and it is much less affected by inaccuracies in system properties. The gains of the proposed MTEJ can be selected more systematically and do not need to be large; hence, the noise rejection characteristics of the algorithm are improved. Also, no need for the pseudo-inversion of the Jacobian matrix in the proposed controller makes further convenience in the underactuated cases. In addition, the relationship between kinematic and dynamic manipulability measures is discussed for underactuated manipulators. Obtained results show its superior performance even compared to that of the model-based algorithms that need full dynamics models, while the proposed MTEJ requires much lower computation effort.     

Vehicle-manipulator system dynamic modeling and control for underwater autonomous manipulation

Underwater autonomous manipulation is a challenging task, which not only includes a complicated multibody dynamic and hydrodynamic process, but also involves the limited observation environment. This study systematically investigates the dynamic modeling and control of the underwater vehicle-manipulator multibody system. The dynamic model of underwater vehicle-manipulator system has been established on the basis of the Newton–Euler recursive algorithm. On the basis of dynamic analysis, a motion planning optimization algorithm has been designed in order to realize the coordinate motions between AUV and manipulator through reducing the restoring forces and saving the electric power. On the other hand, a disturbance force observer including the coupling and restoring forces has been designed. An observer-based dynamic control scheme has been established in combination with kinematic and dynamic controller. Furthermore, from the simulations, although the disturbance forces such as restoring and coupling forces are time varying and great, the observer-based dynamic coordinate controller can maintain the AUV attitude stable during the manipulator swing and pitch motions. During the precise manipulation simulation, the stable AUV attitude and minimization of disturbance forces have been realized through combination of optimal motion planning and the observer-based dynamic coordinate controller.     

机器人的鲁棒预测控制

本文提出基于误差预测的机器人鲁棒控制器。考虑到机器人的动力学建模误差影响其控制性能,本文建立机器人的误差模型,给出预测建模误差对运动轨迹偏差的作用的有效方法,并提出建模误差的鲁棒性补偿。本文分别在关节空间和直角空间针对冗余机器人和非冗余机器人提出鲁棒预测控制器设计,其有效性由仿真例子检验。     

Robust control of robot manipulators based on uncertainty and disturbance estimation

In this work, uncertainty and disturbance estimation (UDE) based robust trajectory tracking controller for rigid link manipulators was proposed. The UDE was employed to estimate the composite uncertainty that comprises the effects of system nonlinearities, external disturbances, and parametric uncertainties. A feedback linearization based controller was designed for trajectory tracking, and the same was augmented by the UDE‐estimated uncertainties to achieve robustness. The resulting controller however required measurement of joint velocities apart from the joint positions. To address the issue, an observer that employed the UDE‐estimated uncertainties for robustness was proposed, giving rise to the UDE‐based controller–observer structure. Closed‐loop stability of the overall system was established. The notable feature of the proposed design was that it neither required accurate plant model nor any information about the uncertainty. Also, the design needed only joint position measurements for its implementation. To demonstrate the effectiveness, simulation results of the proposed approach as applied to the trajectory tracking control of two‐link robotic manipulator and comparison of its performance with some of the well‐known existing controllers were presented. Lastly, hardware implementation of the proposed design for trajectory control of Quanser's single‐link flexible joint module was carried out, and it was shown that the proposed strategy offered a viable approach for designing implementable robust controllers for robots. Copyright © 2011 John Wiley & Sons, Ltd.     

Robust H∞ Tracking Control of Uncertain Robotic Systems with Periodic Disturbances

This paper addresses the problem of designing robust tracking control for a large class of uncertain robotic systems. A more general model of the external disturbance is employed in the sense that the external disturbance can be expressed as the sum of a modeled disturbance and an unmodeled disturbance, for example, any periodic disturbance can be expressed in this general form. An adaptive neural network system is constructed to approximate the behavior of unknown robot dynamics. An adaptive control algorithm is designed to estimate the behavior of the modeled disturbance, and in turn the robust H_∞ control algorithm is required to attenuate the effects of the unmodeled disturbance only. Consequently, an intelligent adaptive/robust tracking control scheme is constructed such that an H_∞ tracking control is achieved in the sense that all the states and signals of the closed‐loop system are bounded and the effect due to the unmodeled disturbance on the tracking error can be attenuated to any preassigned level. Finally, simulations are provided to demonstrate the effectiveness and performance of the proposed control algorithm.     

Robust adaptive fuzzy sliding mode trajectory tracking control for serial robotic manipulators

The ever increasingly stringent performance requirements of industrial robotic applications highlight significant importance of advanced robust control designs for serial robots that are generally subject to various uncertainties and external disturbances. Therefore, this paper proposes and investigates the design and implementation of a robust adaptive fuzzy sliding mode controller in the task space for uncertain serial robotic manipulators. The sliding mode control is well known for its robustness to system parameter variations and external disturbances, and is thus a highly desirable and cost-effective approach to achieve high precision control task for serial robots. The proposed controller is designed based on a fuzzy logic approximation to accomplish trajectory tracking with high accuracy and simultaneously attenuate effects from uncertainties. In the controller, the high-frequency uncertain term is approximated by using a fuzzy logic system while the low-frequency term is adaptively updated in real time based on a parametric adaption law. The control efficacy and effectiveness of the proposed control algorithm are comparatively verified against a recently proposed conventional controller. The test results demonstrate that the proposed controller has better trajectory tracking performances and is more robust against large disturbances than the conventional controller under the same operating conditions.     

Neural-network-based robust fault diagnosis in robotic systems

Fault diagnosis plays an important role in the operation of modern robotic systems. A number of researchers have proposed fault diagnosis architectures for robotic manipulators using the model-based analytical redundancy approach. One of the key issues in the design of such fault diagnosis schemes is the effect of modeling uncertainties on their performance. This paper investigates the problem of fault diagnosis in rigid-link robotic manipulators with modeling uncertainties. A learning architecture with sigmoidal neural networks is used to monitor the robotic system for any off-nominal behavior due to faults. The robustness and stability properties of the fault diagnosis scheme are rigorously established. Simulation examples are presented to illustrate the ability of the neural-network-based robust fault diagnosis scheme to detect and accommodate faults in a two-link robotic manipulator.     

Robust learning control for robotic manipulators with an extension to a class of non-linear systems

A robust learning control (RLC) scheme is developed for robotic manipulators by a synthesis of learning control and robust control methods. The non-linear learning control strategy is applied directly to the structured system uncertainties that can be separated and expressed as products of unknown but repeatable (over iterations) state-independent time functions and known state-dependent functions. The non-linear uncertain terms in robotic dynamics such as centrifugal, Coriolis and gravitational forces belong to this category. For unstructured uncertainties which may have non-repeatable factors but are limited by a set of known bounding functions as the only a priori knowledge, e.g the frictions of a robotic manipulator, robust control strategies such as variable structure control strategy can be applied to ensure global asymptotic stability. By virtue of the learning and robust properties, the new control system can easily fulfil control objectives that are difficult for either learning control or variable structure control alone to achieve satisfactorily. The proposed RLC scheme is further shown to be applicable to certain classes of non-linear uncertain systems which include robotic dynamics as asubset. Various important properties concerning learning control, such as the need for a resetting condition and derivative signals, whether using iterative control mode or repetitive control mode, are also made clear in relation to different control objectives and plant dynamics.     

基于内模的机器人迭代学习控制

本文根据内模控制的概念，设计一个扰动控制器，使机器人系统表现为固定参数的解耦线性化系统．基于此线性系统，提出了一种迭代学习控制律，给出了算法收敛的充分条件．算法的参数选择非常简单，从而易于满足收敛条件．仿真结果表明了算法的有效性．