Abstract:This paper focuses on neural learning from adaptive neural control (ANC) for a class of flexible joint manipulator under the output tracking constraint. To facilitate the design, a new transformed function is introduced to convert the constrained tracking error into unconstrained error variable. Then, a novel adaptive neural dynamic surface control scheme is proposed by combining the neural universal approximation. The proposed control scheme not only decreases the dimension of neural inputs but also reduces t… Show more
“…In a general framework of neural-learning control, NNs are used to either approximate the robot forward dynamics or inverse dynamics for controller design [13], and the control problem falls within the context of either reinforcement learning [14,15], adaptive control [16], or optimal control [17]. Radial basis function (RBF) NNs, multi-layer feedforward NNs, and recurrent neural networks (RNNs) have been explored for trajectory tracking control [18][19][20][21][22]. The concept of incorporating fuzzy logic into NN control has also grown into a popular research topic [23][24][25][26].…”
Fast and precise robot motion is needed in many industrial applications. Most industrial robot motion controllers allow externally commanded motion profiles, but the trajectory tracking performance is affected by the robot dynamics and joint servo controllers, to which users have no direct access and about which they have little information. The performance is further compromised by time delays in transmitting the external command as a setpoint to the inner control loop. This paper presents an approach for combining neural networks and iterative learning controls to improve the trajectory tracking performance for a multi-axis articulated industrial robot. For a given desired trajectory, the external command is iteratively refined using a high-fidelity dynamical simulator to compensate for the robot inner-loop dynamics. These desired trajectories and the corresponding refined input trajectories are then used to train multi-layer neural networks to emulate the dynamical inverse of the nonlinear inner-loop dynamics. We show that with a sufficiently rich training set, the trained neural networks generalize well to trajectories beyond the training set as tested in the simulator. In applying the trained neural networks to a physical robot, the tracking performance still improves but not as much as in the simulator. We show that transfer learning effectively bridges the gap between simulation and the physical robot. Finally, we test the trained neural networks on other robot models in simulation and demonstrate the possibility of a general purpose network. Development and evaluation of this methodology are based on the ABB IRB6640-180 industrial robot and ABB RobotStudio software packages.
“…In a general framework of neural-learning control, NNs are used to either approximate the robot forward dynamics or inverse dynamics for controller design [13], and the control problem falls within the context of either reinforcement learning [14,15], adaptive control [16], or optimal control [17]. Radial basis function (RBF) NNs, multi-layer feedforward NNs, and recurrent neural networks (RNNs) have been explored for trajectory tracking control [18][19][20][21][22]. The concept of incorporating fuzzy logic into NN control has also grown into a popular research topic [23][24][25][26].…”
Fast and precise robot motion is needed in many industrial applications. Most industrial robot motion controllers allow externally commanded motion profiles, but the trajectory tracking performance is affected by the robot dynamics and joint servo controllers, to which users have no direct access and about which they have little information. The performance is further compromised by time delays in transmitting the external command as a setpoint to the inner control loop. This paper presents an approach for combining neural networks and iterative learning controls to improve the trajectory tracking performance for a multi-axis articulated industrial robot. For a given desired trajectory, the external command is iteratively refined using a high-fidelity dynamical simulator to compensate for the robot inner-loop dynamics. These desired trajectories and the corresponding refined input trajectories are then used to train multi-layer neural networks to emulate the dynamical inverse of the nonlinear inner-loop dynamics. We show that with a sufficiently rich training set, the trained neural networks generalize well to trajectories beyond the training set as tested in the simulator. In applying the trained neural networks to a physical robot, the tracking performance still improves but not as much as in the simulator. We show that transfer learning effectively bridges the gap between simulation and the physical robot. Finally, we test the trained neural networks on other robot models in simulation and demonstrate the possibility of a general purpose network. Development and evaluation of this methodology are based on the ABB IRB6640-180 industrial robot and ABB RobotStudio software packages.
“…where the first two terms guarantee the constraints satisfaction, the third term improves the robustness to disturbance, and the last term is used to compensate for the dynamics uncertainties. Substituting (33), (34), (36) and (38) into (32), we have:…”
Section: Control Designmentioning
confidence: 99%
“…Therefore, considering the closed-loop system including the robot dynamics (22), the control torque (38) and the NN update law (44), the tracking error e 1 will converge asymptotically to the compact set:…”
In this paper, we propose a biologically-inspired framework for robot learning based on demonstrations. The dynamic movement primitive (DMP), which is motivated by neurobiology and human behavior, is employed to model a robotic motion that is generalizable. However, the DMP method can only be used to handle a single demonstration. To enable the robot to learn from multiple demonstrations, the DMP is combined with the Gaussian mixture model (GMM) to integrate the features of multiple demonstrations, where the conventional GMM is further replaced by the Fuzzy GMM (FGMM) to improve the fitting performance. Also, a novel regression algorithm for FGMM is derived to retrieve the nonlinear term of the DMP. Additionally, a neural network based controller is developed for the robot to track the generated motions. In this network, the cerebellar model articulation controller (CMAC) is employed to compensate for the unknown robot dynamics. The experiments have been performed on a Baxter robot to demonstrate the effectiveness of the proposed methods. Index Terms-Robot learning from demonstrations, dynamic movement primitive, fuzzy Gaussian mixture model, Gaussian mixture regression, cerebellar model articulation controller, neural control.
“…To address the learning and control problem of unknown cascaded nonlinear systems, a few elegant dynamic learning methods were proposed by combining a recursive design with a system decomposition strategy (Wang, Wang, Liu, & Hill, 2012;Wang & Wang, 2015a, 2015b. Furthermore, the learning mechanism was also applied into some physical systems such as marine surface vessels (Dai, Wang, & Luo, 2012;Dai, Zeng, & Wang, 2016), robot manipulators (Wang, Ye, and Chen, 2017b), and gait recognition (Zeng & Wang, 2016). For FJR with unknown system dynamics, two or three NN approximators will be used in a recursive design process.…”
This paper presents dynamic learning from adaptive neural control with prescribed tracking error performance for flexible joint robot (FJR) included unknown dynamics. Firstly, a system transformation method is introduced to convert the original FJR system into a normal system. As a result, only one neural network (NN) approximator is used to identify the uncertain system nonlinear dynamics and the verification on the convergence of neural weights is simplified extremely. To further solve the predefined performance issue, a performance function is introduced to describe a tracking error constraint and a error transformation technique is used to convert the constrained tracking control problem into the unconstrained stabilization of error system. By combining a high-gain observer and the backsteppig method, the adaptive neural controller is presented to stabilize the unconstrained error system. Under the satisfaction of the partial persistent excitation condition, the adaptive neural controller is shown to be capable of achieving unknown dynamics acquisition, expression and storage. Furthermore, a neural learning control with using the stored NN weights is proposed for the same or similar control task so that a time-consuming NN online adjustment process can be avoided and a better control performance can be obtained. Simulation results demonstrate the effectiveness of the proposed control method.
ARTICLE HISTORY
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.