Simulated Robot Arm Research Articles

Control of the trajectory of a planar robot using recurrent hybrid networks

This paper describes the use of recurrent neural networks in the control of a simulated planar two-jointed robot arm. Recurrent networks have feedback connections and thus an inherent memory for dynamics which makes them suitable for dynamic system modelling. A feature of the networks adopted is their hybrid hidden layer which includes both linear and non-linear neurons. This facilitates learning of the inverse dynamics model of the robot which can be thought of as comprising a linear and a non-linear part. Following a brief description of the control problem and alternative PID and computed-torque control schemes, the proposed neural network and neural controller will be detailed. The results presented show the superior ability of the proposed neural control scheme at adapting to changes in the dynamics parameters of the robot.

OPTIMAL HIDDEN UNITS FOR TWO-LAYER NONLINEAR FEEDFORWARD NEURAL NETWORKS

The output layer of a feedforward neural network approximates nonlinear functions as a linear combination of a fixed set of basis functions, or "features". These features are learned by the hidden-layer units, often by a supervised algorithm such as a back-propagation algorithm. This paper investigates features which are optimal for computing desired output functions from a given distribution of input data, and which must therefore be learned using a mixed supervised and unsupervised algorithm. A definition is proposed for optimal nonlinear features, and a constructive method, which has an iterative implementation, is derived for finding them. The learning algorithm always converges to a global optimum and the resulting network uses two layers to compute the hidden units. The general form of the features is derived for the case of continuous signal input, and this result is related to transmission of information through a bandlimited channel. The results of other algorithms can he compared to the optimal features, which in some cases have easily computed closed-form solutions. The application of this technique to the inverse kinematics problem for a simulated planar two-joint robot arm is demonstrated here.

A tree-structured adaptive network for function approximation in high-dimensional spaces

Nonlinear function approximation is often solved by finding a set of coefficients for a finite number of fixed nonlinear basis functions. However, if the input data are drawn from a high-dimensional space, the number of required basis functions grows exponentially with dimension, leading many to suggest the use of adaptive nonlinear basis functions whose parameters can be determined by iterative methods. The author proposes a technique based on the idea that for most of the data, only a few dimensions of the input may be necessary to compute the desired output function. Additional input dimensions are incorporated only where needed. The learning procedure grows a tree whose structure depends upon the input data and the function to be approximated. This technique has a fast learning algorithm with no local minima once the network shape is fixed, and it can be used to reduce the number of required measurements in situations where there is a cost associated with sensing. Three examples are given: controlling the dynamics of a simulated planar two-joint robot arm, predicting the dynamics of the chaotic Mackey-Glass equation, and predicting pixel values in real images from pixel values above and to the left.

Three-dimensional neural net for learning visuomotor coordination of a robot arm

An extension of T. Kohonen's (1982) self-organizing mapping algorithm together with an error-correction scheme based on the Widrow-Hoff learning rule is applied to develop a learning algorithm for the visuomotor coordination of a simulated robot arm. Learning occurs by a sequence of trial movements without the need for an external teacher. Using input signals from a pair of cameras, the closed robot arm system is able to reduce its positioning error to about 0.3% of the linear dimensions of its work space. This is achieved by choosing the connectivity of a three-dimensional lattice consisting of the units of the neural net.

Topology-conserving maps for learning visuo-motor-coordination

We investigate the application of an extension of Kohonen's self-organizing mapping algorithm to the learning of visuo-motor-coordination of a simulated robot arm. We show that both arm kinematics and arm dynamics can be learned, if a suitable representation for the map output is used. Due to the topology-conserving property of the map spatially neighboring neurons can learn cooperatively, which greatly improves the robustness and the convergence properties of the algorithm.

Robot Motion and Task Planning:Simulation and Programming of a Robot Arm

ABSTRACT A robot simulation system has been developed to assist in robot task planning and control data programming. The system allows simulation of the robot manipulator on the CRT by taking into consideration the kinematics and dynamics of the robot arm. The architecture of the robot simulation system is presented. Description of robot positional kinematics is given by the use of geometrical model and its extension to homogeneous transformation. Several simulated robot arm positions are shown to illustrate the kinematic model.

Pole Placement Methods for Multivariable Control of Robotic Manipulators

Efficient methods for multivariable digital control of robotic manipulators based upon pole placement by state feedback are presented as an improved method of path control. Implementation of these methods by both gain scheduling and on-line computation is discussed. An example of the computations for a three-link robot arm and computer simulation experiments are given.

Simulating Motion Elements of General-Purpose Robot Arms

Robot arm displacement curves are constructed, assuming that all links in the arm are rigid. This can be accomplished by using familiar, cam-motion profile techniques whenever possible. The benefits of these considerations can help achieve smoother motion while minimizing wear on and effort by the robot actuators. A set of motion primitives within the General Robot Arm Simulation Program ( GRASP) are presented as examples.

On the Fitness of Behavior Sequences

Our aim is to demonstrate a method of determining the extent to which behavior maximizes fitness. We believe the temporal organization of behavior to be in part dependent on the animal's genetic makeup and subject to natural selection and that behavioral strategies may be as adaptive as structural characters. The question of the adaptiveness of behavior has been found hard to study in the past (Tinbergen 1963) because of the difficulty of quantitative verification of hypotheses. The method presented here tests specifically whether deployment of behavioral options is optimally related to environmental conditions.