共查询到20条相似文献,搜索用时 171 毫秒
1.
2.
3.
4.
5.
6.
针对深度强化学习在高维机器人行为控制中训练效率低下和策略不可解释等问题,提出一种基于结构交互驱动的机器人深度强化学习方法(structure-motivated interactive deep reinforcement learning, SMILE).首先,利用结构分解方法将高维的单机器人控制问题转化为低维的多关节控制器协同学习问题,从而缓解连续运动控制的维度灾难难题;其次,通过两种协同图模型(ATTENTION和PODT)动态推理控制器之间的关联关系,实现机器人内部关节的信息交互和协同学习;最后,为了平衡ATTENTION和PODT协同图模型的计算复杂度和信息冗余度,进一步提出两种协同图模型更新方法 APDODT和PATTENTION,实现控制器之间长期关联关系和短期关联关系的动态自适应调整.实验结果表明,基于结构驱动的机器人强化学习方法能显著提升机器人控制策略学习效率.此外,基于协同图模型的关系推理及协同机制,可为最终学习策略提供更为直观和有效的解释. 相似文献
7.
面向未知环境基于智能预测的模糊控制器研究 总被引:4,自引:0,他引:4
提出了一种新的面向未知环境的智能预测算法,并将此算法应用于机器人力跟踪控制中.该方法利用机械手末端与未知受限环境产生的接触轨迹,通过模糊推理智能地预测阻抗控制模型中的参考轨迹,并根据力误差变化用参考比例因子对其进行调节,以适应未知环境刚度的变化.通过对阻抗模型参数进行模糊调节减少受限运动中的力误差,提高了全局的力控制效果.仿真结果证明了此算法的有效性. 相似文献
8.
9.
10.
考虑机器人间的通信受限约束,将机器人抽象为微粒,提出基于微粒群优化的多机器人气味寻源方法.首先,采用结合斥力函数的策略,引导机器人快速搜索烟羽;然后,基于无线信号对数距离损耗模型,估计机器人间的通讯范围,据此形成微粒群的动态拓扑结构,并确定微粒的全局极值;最后,将传感器的采样/恢复时间融入微粒更新公式,以跟踪烟羽.将所提出方法应用于3个不同场景的气味寻源,实验结果验证了该方法的有效性. 相似文献
11.
Based on a combination of a PD controller and a switching type two-parameter compensation force, an iterative learning controller with a projection-free adaptive algorithm is presented in this paper for repetitive control of uncertain robot manipulators. The adaptive iterative learning controller is designed without any a priori knowledge of robot parameters under certain properties on the dynamics of robot manipulators with revolute joints only. This new adaptive algorithm uses a combined time-domain and iteration-domain adaptation law allowing to guarantee the boundedness of the tracking error and the control input, in the sense of the infinity norm, as well as the convergence of the tracking error to zero, without any a priori knowledge of robot parameters. Simulation results are provided to illustrate the effectiveness of the learning controller. 相似文献
12.
Suguru Arimoto 《国际强度与非线性控制杂志
》1995,5(4):269-284
》1995,5(4):269-284
A principle of ‘joint-space orthogonalization’ is proposed as an extended notion of hybrid (force and position) control for robot manipulators under geometric constraints. The principle realizes the hybrid control in a strict sense by letting position feedback signals be orthogonal in joint space to the contact force vector whose components exert at corresponding joints. This orthogonalization is executed via a projection matrix computed in real-time from a Jacobian matrix of the constraint equation in joint coordinates. To show the important role of the principle in control of robot manipulators, two basic set-point control problems are analysed. One is a hybrid PID control problem for robot manipulators under geometric endpoint constraint and another is a coordinated control problem of two arms. It is shown that passivity properties of residual dynamics of robots follow from the introduction of a quasi-natural potential and the joint-space orthogonalization. Various stability problems of PID-type feedback control schemes without compensating for the gravity force and with or without use of a force sensor are discussed from passivity properties of robot dynamics with the aid of the hyper-stability theory. 相似文献
13.
《Advanced Robotics》2013,27(6):621-636
This paper proposes a decentralized position/internal force hybrid control approach for multiple robot manipulators to cooperatively manipulate an unknown dynamic object. In this approach, each autonomous robot has its own controller and uses its own sensor information in performing the fast cooperation. This approach eliminates a lot of information communications between each robot and reduces numerous computations. The influences of the position and the internal force estimation errors to the overall control system is analyzed. A cooperative identification method for each autonomous robot to identify the object's complex dynamics, cooperatively, is presented. In addition, the trade-off between the unilateral force constraint and the robots' position response is studied. Experiments show the effectiveness of this control approach. 相似文献
14.
Due to task kinematic modelling inaccuracy, constraint functions imposed on robot manipulators may not be known exactly. In this article, a variable structure control (VSC) method is developed for robust motion and constrained force control of robot manipulators in the presence of parametric uncertainties, external disturbances, and constraint function uncertainties. The method is based on a particular structure of the constrained robot, in which motion control and force control are treated together. The proposed VSC controller provides the sliding mode and reaching transient response with prescribed qualities. A sufficient condition to guarantee the robot does not lose contact with the constraint surface is given. Detailed simulation results illustrate the proposed method. © 1994 John Wiley & Sons, Inc. 相似文献
15.
Feedback stabilization and tracking of constrained robots 总被引:1,自引:0,他引:1
Mathematical models for constrained robot dynamics, incorporating the effects of constraint force required to maintain satisfaction of the constraints, are used to develop explicit conditions for stabilization and tracking using feedback. The control structure allows feedback of generalized robot displacements, velocities, and the constraint forces. Global conditions for tracking, based on a modified computed-torque controller and local conditions for feedback stabilization, using a linear controller, are presented. The framework is also used to investigate the closed-loop properties if there are force disturbances, dynamics in the force feedback loops, or uncertainty in the constraint functions 相似文献
16.
De Queiroz M.S. Jun Hu Dawson D.M. Burg T. Donepudi S.R. 《IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics》1997,27(5):796-809
In this paper, we design an adaptive position/force controller for robot manipulators during constrained motion. The proposed controller can compensate for parametric uncertainty while only requiring measurements of link position and end-effector force. A filtering technique is utilized to produce a pseudo-velocity error signal and thus, eliminate the need for link velocity measurements. The control strategy provides semiglobal asymptotic tracking performance for the end-effector position and the interaction force between the constraint and the end-effector. An experimental implementation of the proposed controller on a two-link planar robot is also presented. 相似文献
17.
Robust adaptive motion/force tracking control design for uncertain constrained robot manipulators 总被引:3,自引:0,他引:3
In the presence of uncertain constraint and robot model, an adaptive controller with robust motion/force tracking performance for constrained robot manipulators is proposed. First, robust motion and force tracking is considered, where a performance criterion containing disturbance and estimated parameter attenuations is presented. Then the proposed controller utilizes an adaptive scheme and an auxiliary control law to deal with the uncertain environmental constraint, disturbances, and robotic modeling uncertainties. After solving a simple linear matrix inequality for gain conditions, the effect from disturbance and estimated parameter errors to motion/force errors is attenuated to an arbitrary prescribed level. Moreover, if the disturbance and estimated parameter errors are square-integrable, then an asymptotic motion tracking is achieved while the force error is as small as the inversion of control gain. Finally, numerical simulation results for a constrained planar robot illustrate the expected performance. 相似文献
18.
This article considers the question of position and force control of three-link elastic robotic systems on a constraint surface in the presence of robot parameter and environmental constraint geometry uncertainties. The approach of this article is applicable to any multi-link elastic robot. A sliding mode control law is derived for the position and force trajectory control of manipulator. Unlike the rigid robots, sliding mode control of an end point gives rise to unstable zero dynamics. Instability of the zero dynamics is avoided by Controlling a point that lies in the neighborhood of the actual end point position. The sliding mode controller accomplishes tracking of the end-effector and force trajectories on the constrained surface; however, the maneuver of the arm causes elastic mode excitation. For point-to-point control on the constraint surface, a stabilizer is designed for the final capture of the terminal state and vibration suppression. Numerical results are presented to show that in the closed-loop system position and force control is accomplished in spite of payload and constraint surface geometry uncertainty. © 1995 John Wiley & Sons, Inc. 相似文献
19.
Mitsuhiro Yamano Jin-Soo Kim Atsushi Konno Masaru Uchiyama 《Journal of Intelligent and Robotic Systems》2004,39(1):1-15
This paper discusses cooperative control of a dual-flexible-arm robot to handle a rigid object in three-dimensional space. The proposed control scheme integrates hybrid position/force control and vibration suppression control. To derive the control scheme, kinematics and dynamics of the robot when it forms a closed kinematic chain is discussed. Kinematics is described using workspace force, velocity and position vectors, and hybrid position/force control is extended from that on dual-rigid-arm robots. Dynamics is derived from constraint conditions and the lumped-mass-spring model of the flexible robots and an object. The vibration suppression control is calculated from the deflections of the flexible links and the dynamics. Experiments on cooperative control are performed. The absolute positions/orientations and internal forces/moments are controlled using the robot, each arm of which has two flexible links, seven joints and a force/torque sensor. The results illustrate that the robot handled the rigid object damping links' vibration successfully in three-dimensional space. 相似文献
20.
Conventional robot control schemes are basically model-based methods. However, exact modeling of robot dynamics poses considerable problems and faces various uncertainties in task execution. This paper proposes a reinforcement learning control approach for overcoming such drawbacks. An artificial neural network (ANN) serves as the learning structure, and an applied stochastic real-valued (SRV) unit as the learning method. Initially, force tracking control of a two-link robot arm is simulated to verify the control design. The simulation results confirm that even without information related to the robot dynamic model and environment states, operation rules for simultaneous controlling force and velocity are achievable by repetitive exploration. Hitherto, however, an acceptable performance has demanded many learning iterations and the learning speed proved too slow for practical applications. The approach herein, therefore, improves the tracking performance by combining a conventional controller with a reinforcement learning strategy. Experimental results demonstrate improved trajectory tracking performance of a two-link direct-drive robot manipulator using the proposed method. 相似文献