Alan S. Morris

Dynamic modelling, simulation and control of a manipulator with flexible links and joints

Abstract The paper presents a dynamic modelling technique for a manipulator with multiple flexible links and flexible joints, based on a combined Euler–Lagrange formulation and assumed modes method. The resulting generalised model is validated through computer simulations by considering a simplified case study of a two-link flexible manipulator with joint elasticity. Controlling such a manipulator is more complex than controlling one with rigid joints because only a single actuation signal can be applied at each joint and this has to control the flexure of both the joint itself and the link attached to it. To resolve the control complexities associated with such an under-actuated flexible link/flexible joint manipulator, a singularly perturbed model has been formulated and used to design a reduced-order controller. This is shown to stabilise the link and joint vibrations effectively while maintaining good tracking performance.

Soft computing methods applied to the control of a flexible robot manipulator

The paper describes use of soft computing methods (fuzzy logic and neural network techniques) in the development of a hybrid fuzzy neural control (HFNC) scheme for a multi-link flexible manipulator. A manipulator with multiple flexible links is a multivariable system of considerable complexity due to the inter-link coupling effects that are present in both rigid and flexible motions. Modelling and controlling the dynamics of such manipulators is therefore difficult. The proposed HFNC scheme generates control actions combining contributions form both a fuzzy controller and a neural controller. The primary loop of the proposed HFNC contains a fuzzy controller and a neural network controller in the secondary loop to compensate for the coupling effects due to the rigid and flexible motion along with the inter-link coupling. It has been ascertained from the present investigation that the proposed soft-computing-based controller works effectively in the tracking control of such a multi-link flexible manipulator. The results are extendable to other multivariable systems of similar complexity.

A fuzzy expert system for fault detection in statistical process control of industrial processes

Little work has previously been reported on the use of fuzzy logic within statistical process control when this is used for fault detection as part of quality control systems in industrial manufacturing processes. Therefore, the paper investigates the potential use of fuzzy logic to enhance the performance of statistical process control (SPC). The cumulative sum of the deviation in the monitored parameter is combined with the deviation in an attempt to discriminate between false alarms and real faults and, consequently, to improve the quality of the solution. Combinations of control rules are utilized and trained to cope with different inputs such that rejection of false alarms is achieved and quick detection of real faults is obtained. The design and implementation of this fuzzy expert system (FES) are presented, and a comparative rule based study is performed.

Using short-time Fourier transform and wavelet packet filter banks for improved frequency measurement in a Doppler robot tracking system

Previous research has shown that an ultrasonic tracking system using the Doppler effect can potentially track high-speed robot motion very accurately. However, ultrasound energy that is reflected by obstacles in the robot workspace can cause significant distortions in the frequency measurement. To reduce the distortions, a novel method for frequency measurement is described in this paper. This uses the short-time Fourier transform (STFT) to estimate the frequency of the interference signal. Then, the estimated frequencies are analyzed with an adaptive filter constructed with wavelet packet filter banks. Results are given that demonstrate great improvements in frequency measurement.

Neuro-adaptive control of robotic manipulators

The need to meet demanding control requirements in increasingly complex dynamical control systems under significant uncertainties makes neural networks very attractive, because of their ability to learn, to approximate functions, to classify patterns and because of their potential for massively parallel hardware implementation. This paper proposes the use of artificial neural networks (ANN) as a novel approach to the control of robot manipulators. These are part of the general class of non-linear dynamic systems where non-linear compensators are required in the controller.

Introduction to Measurement

This chapter provides an introduction to the various measurement techniques. The instrument choice is a compromise between performance characteristics, ruggedness and durability, maintenance requirements and purchase cost. Measurement techniques have been of immense importance ever since the start of human civilization, when measurements were first needed to regulate the transfer of goods in barter trade to ensure that exchanges were fair. The very first measurement units were those used in barter trade to quantify the amounts being exchanged, and to establish clear rules about the relative values of different commodities. Today, the techniques of measurement are of immense importance in most facets of human civilization. Measuring system exists to provide information about the physical value of some variable being measured. In simple cases, the system can consist of only a single unit that gives an output reading or signal according to the magnitude of the unknown variable applied to it. The starting point in choosing the most suitable instrument for the measurement of a particular quantity in a manufacturing plant or other system is the specification of the instrument characteristics required, especially parameters like the desired measurement accuracy, resolution, sensitivity, and dynamic performance. To carry out such an evaluation properly, the instrument engineer must have a wide knowledge of the range of instruments available for measuring particular physical quantities; and he/she must also have a deep understanding of how instrument characteristics are affected by particular measurement situations and operating conditions.

Singular perturbation approach to trajectory tracking of flexible robot with joint elasticity

The control problem of a robot manipulator with flexures both in the links and joints was investigated using the singular perturbation technique. Owing to the combined efects of the link and joint jlexibilities, the dynamics of this type of manipulator become more complex and under-actuated leading to a challenging control task. The singular perturbation being a successful technique for solving control problems with under-actuation was exploited to obtain simpler subsystems with two-time-scale separation, thus enabling easier design of subcontrollers. Furthermore, simultaneous tracking and suppression of vibration of the link andjoint of the manipulator is possible by application of the composite controller, i.e. the superposition of both subcontrol actions. In the first instance, the design of a composite controller was based on a computed torque control for slow dynamics and linear-quadratic fast control. Later, to obtain an improved control performance under model uncertainty, the composite control action was achieved using the radial basis function neural network for the slow control and a linear-quadratic fast control. It was confirmed through numerical simulations that the proposed singular perturbation controllers suppress the joint and link vibrations of the manipulator satisfactorily while a perfect trajectory tracking was achieved.

Singular perturbation based neuro-H ∞ control scheme for a manipulator with flexible links and joints

A novel composite control scheme for a manipulator with flexible links and joints is presented that uses the singular perturbation technique (SPT) to divide the manipulator dynamics into reduced order slow and fast subsystems. A neural network controller is then applied for the slow subsystem and a state-feedback H∞ controller for the fast subsystem. Results are presented that demonstrate improved performance over an alternative SPT-based controller that uses inverse dynamics and LQR controllers.

Finding the inverse kinematics of manipulator arm using artificial neural network with lookup table

Neural networks were used to find the inverse kinematics of a two-link planar and three-link manipulator arms. The neural networks utilised were multi-layered perceptions with a back-propagation training algorithm. Because of the redundancy in the manipulators studied, this work used lookup tables for the different configurations of the manipulator arm.

Genetic Algorithm Tuned Fuzzy Logic Controller for a Robot Arm with Two-link Flexibility and Two-joint Elasticity

Many recent contributions on flexible link and elastic joint robotic arms focus on how to solve path tracking and vibration damping problems in slow and fast mode control, respectively. For slow mode control, the problem has been dealt with previously by soft computing tools in which some parameters are designed manually. As a result, system performances are often tiresome and intractable. This paper introduces a scheme to improve the system performance by applying genetic algorithms (GAs) to tune the membership function parameters of a fuzzy logic controller for the slow mode of a two-flexible-link and two-elastic-joint robotic manipulator. The system with the new controller is simulated and its behaviour is compared with that provided by conventional and expert-designed fuzzy logic controllers.