Figure 1 - uploaded by Wayne Book
Content may be subject to copyright.
Source publication
A single-link flexible manipulator with a rotary actuator at one
end and a mass at the other is modeled using the Lagrangian method
coupled with an assumed modes vibration model. A SIMO state space model
is developed by linearizing the equations of motion and simplified by
neglecting natural damping. Laplace domain pole-zero plots between
torque in...
Context in source publication
Citations
... A multi-layer perceptron controuer was constructed by Register et al. [24] to learn the inverse dynarnics model of a single-link manipulator for the tip position control. The simulation model was based on a lïnearized state-space model that has neglected natural damping. ...
This paper presents an intelligent-based control strategy for tip
position tracking control of a single-link flexible manipulator.
Motivated by the well-known inverse dynamics control strategy for
rigid-link manipulators, two feedforward neural networks (NNs) are
proposed to learn the nonlinearities of the flexible arm associated with
the inverse dynamics controller. The redefined output approach is used
by feeding back this output to guarantee the minimum phase behavior of
the resulting closed-loop system. No a priori knowledge about the
nonlinearities of the system is needed and the payload mass is also
assumed to be unknown. The network weights are adjusted using a modified
online error backpropagation algorithm that is based on the propagation
of output tracking error, derivative of that error and the tip
deflection of the manipulator. The real-time controller is implemented
on an experimental test bed. The results achieved by the proposed
NN-based controller are compared experimentally with conventional
proportional-plus-derivative-type and standard inverse dynamics controls
to substantiate and verify the advantages of our proposed scheme and its
promising potential in identification and control of nonlinear systems
... Mechanical vibration is a common phenomenon observed in the operation of an elastic manipulator that arises from the inertia and stiffness effect of machine parts in motion. Research on the problem of neural network control of an elastic manipulator has drawn wide attention [1][2][3][4][5][6][7][8][9]. This is due to the fact that compared to other control schemes, the neural networks do not need any mathematical model of the system, can process information in a parallel distributed manner, and have learning and self-organization capabilities. ...
... While most of the promising neural network method for the control of flexural manipulators does not appear to converge to a solution when the system is lightly damped [5], a new structure for the control of the elastic manipulators is presented in this paper. The control law is generated from the neural network models in order to control the system to track a desired trajectory. ...
Nonminimum phase property of a rotating elastic manipulator causes difficulties for both classical and neural network inverse model control. While most of the neural network methods for control of elastic manipulators do not appear to converge to a solution when the system is lightly damped, in this paper, an appropriate cost function for a neural controller is proposed. In the designed neural control system, there are only three-layer feedforward networks, consisting of an input layer with two nodes, one hidden layer, and output layer with one node. The number of units in the hidden layer and the value of the learning rate are robust to this designed network algorithm. In order to simulate the transient response of the rotating elastic manipulator system, a single-input, single-output state space representation is presented for the system nonlinear model. It can be seen from the simulation results, the designed neural controller can not only achieve very good tracking performance, zero steady-state errors, and strong robustness to system parameter uncertainty, but also reject the effects of the input torque disturbance.
... Comparison with a fixed-gain controller showed advantages for the multilayer perceptron. The research in [7] was shown to have certain limitations related to nonminimum phase characteristics [8]. In these referenced studies, nonminimum phase characteristics of the flexible link systems were not addressed during the design.of ...
... For flexible link manipulators, an assumed-modes method, coupled with a Lagrangian technique, yields a recursive, closed-form, dynamic solution suitable for control purposes [20]. Systematic application of the Lagrangian, a two mode vibration assumption and some simplifying assumptions, yield the undamped, linearized, inverse dynamic equation [8] used in this investigation. Substitution of the particular physical parameters from [8] s2(s2 +17)(s2 +182) (9) Eqn. 9, with zeros at s = ±5.5 and s = ±19.5, is clearly nonminimum phase. ...
... Systematic application of the Lagrangian, a two mode vibration assumption and some simplifying assumptions, yield the undamped, linearized, inverse dynamic equation [8] used in this investigation. Substitution of the particular physical parameters from [8] s2(s2 +17)(s2 +182) (9) Eqn. 9, with zeros at s = ±5.5 and s = ±19.5, is clearly nonminimum phase. By using the same procedure used with the simple system, the impulse response was determined. ...
A new approach for feedforward ANN control of nonminimum phase mechanical systems is proposed. A standard backpropagation-of-errors ANN is used to form an inverse model controller which is applied to simulated nonminimum phase sys-tems. Learning in the new approach is based on the convolution between a noncausal impulse response and a desired tip trajectory. Selection of the proper input set, input scaling and the ANN struc-ture are investigated. Once the input and structure are specified, the ANN is trained over a single trajectory. After training, the ANN is used to drive the system in an open-loop configuration. Plots of the system states resulting from the ideal excitation and from ANN excitation are compared. The results obtained by varying both the number of units and the input set are presented. The results demonstrate the effectiveness of the proposed ANN inverse model approach.
In this paper, the problem of tip position tracking control of a flexible-link manipulator is considered. Two neural network schemes are presented. In the first scheme, the controller is composed of a stabilizing joint PD controller and a neural network tracking controller. The objective is to simultaneously achieve hub-position tracking and control of the elastic deflections at the tip. In the second scheme, tracking control of a point along the arm is considered to avoid difficulties associated with the output feedback control of a non-minimum phase flexible manipulator. A separate neural network is employed for determining an appropriate output to be used for feedback. The controller is also composed of a neural network tracking controller and a stabilizing joint PD controller. Experimental results on a single-link flexible manipulator show that the proposed networks result in significant improvements in the system response with an increase in controller dynamic range despite changes in the desired trajectory.
This paper addresses the control of a manipulator with link
flexibilities. The increased complexity in its dynamics presents
challenges to controllers based on non-colocated sensing. In this paper
a nonlinear predictive control approach is presented using a discrete
time multilayer perceptron network model for the plant. The neural
network model is trained to predict future outputs based on the
available past measurements. At each sampling instant, the discrete time
control input is calculated by minimizing a performance criterion. The
method is compared to non-model based collocated PD control. Simulation
results are presented
Experimental evaluation of the performance of neural network-based
controllers for tip position tracking of flexible-link manipulators is
presented. A modified output re-definition approach is utilized to
overcome the problem caused by the non-minimum phase characteristic of
the flexible-link system. This modification is based on using minimum a
priori knowledge about the system dynamics. The modified output
redefinition approach requires a priori knowledge about the linear model
of the system and no a priori knowledge about the payload mass. Four
different neural network schemes are proposed. The neural networks are
trained and employed as online controllers. The four proposed neural
network controllers are implemented on a single flexible-link
experimental test-bed. Experimental and simulation results are presented
to illustrate the advantages and improved performance of the proposed
tip position tracking controllers over conventional PD-type controllers
in the presence of unmodeled dynamics
This paper proposes a control system for the flexible material
handling robot. In the past, there has been reported much research work
on flexible material handling robot systems. In most of the previous
work, control system parameters were treated as fixed. However, they
often change due to variation of the flexible payloads. In some cases,
control performance is degraded and the production process is terminated
in some situations. To deal with these cases, we propose an error
detection method and a replanning system. If system parameters change
and control performance is degraded, this system detects error by itself
and changes control parameters to recover from the error automatically.
Some experimental examples are shown to confirm effectiveness of this
proposed system