Quasi-Static Modelling and Computer Simulation of Hanging Cables on Robots

This paper presents a computer method to simulate the quasi-static motion of hanging cables on robots. The shape of the flexible cable is changing during motion and the finite segment method is applied to determine its configuration. The cable is modeled as a series of rigid segments segments connected together through revolute joints in 2-D case and spherical joints in 3-D case. The elasticity of cable is represented by torsional springs at the joints. In both cases, a set of highly nonlinear equations are derived based on force equilibrium and the Newton-Raphson method is applied to calculate the solution. In order to assure convergence and improve computational efficiency, the parameter perturbation method is applied together with the Newton-Raphson method. Also, some computational strategies are developed to simplify the three dimensional problem. Finally, the developed methods are demonstrated in displaying the motion of a hanging cable which is attached to a revolute joint, a prismatic joint and a three degrees of freedom robot.

Optimal Robotic Assembly Planning Using Dijkstra’s Algorithm

Robots play an important role in the modern factory and are used in a manufacturing cell for several functions such as assembly, material handling, robotic welding, etc. One of the principal problems faced by robots while performing their tasks is the presence of obstacles such as fixtures, tools, and objects in the robot workspace. Such objects could result in a collision with one of the arms of the robots. Fast collision-free motion planning algorithms are therefore necessary for robotic manipulators to operate in a wide variety of changing environments. The configuration space approach is one of the widely used methods for collision-free robotic path planning. This paper presents a novel graph-based method of searching the configuration space for a collision-free path in a robotic assembly operation. Dijkstra’s graph search algorithm is used for optimizing the joint displacements of the robot while performing the assembly task. The methodology is illustrated using a simple robotic assembly planning task.

Fuzzy Logic in Hole Drilling of Composites

Experimental and theoretical research data was utilized in building a Fuzzy Logic Controller model applied to simulate the drilling process of composite materials. The objective is to have a better understanding and control of delamination of composites during the drilling process and at the same time to improve the hole finish by controlling fraying and splintering. By controlling the main issues in the drilling process such as feed rate, cutting speed, thrust force, and torque generated in addition to the tool geometry, it is possible to optimize the drilling process avoiding the conventionally encountered problems.

Approximate Degree Reduction of NURBS Curves and Surfaces: An Integrated Approach

Data exchange between different CAD systems usually requires conversion between different representations of free-form curves and surfaces. Also, trimmed surfaces give rise to high degree boundary curves. Accurate conversion of these forms becomes necessary for reliable data transfer. Also important is the issue of shape control, specially in the aircraft industry. The objective of this paper is to investigate conversion methods and effect of shape control on the design and choice of such methods.

Solution to the Forward and Inverse Kinematic Equations Using Multilayer Back-Propagation Neural Network

The problem of inverse kinematics in Robotics, is a nonlinear mapping from a given cartesian coordinates to the desirable joint coordinates of the robot arm. It is found that an appropriately designed neural network can be trained to learn the non-linearity of the Inverse Kinematic Equation (IKE). We present an approach for solving the Forward Kinematic Equation (FKE) and the IKE by means of a Multi Layer Back-Propagation Neural Network (Rumelhart et al., 1986). The neural network approach is applied to a Two Degrees-of-Freedom (DOF) robot manipulator and the results are compared with those obtained using the analytical solution. The results obtained from the simulation of the neural network indicate a fairly accurate learning of the FKE and IKE by the Multi Layer Back-Propagation Neural Network.

Control Systems in Manufacturing

This paper reviews different control methodologies applied in manufacturing environment. Since comparatively newer control methodologies like — Neural networks, Fuzzy logic, and Cerebellar model articulation controller have gained more research interests in recent years, they have been dealt in more detail. With this, we have presented application of neural network for endeffector positioning of three degree planar robot and results have been evaluated. Finally the future research trends in these areas have been discussed.

A Machine Vision System for Inspecting Hydraulic Hose Assemblies

This paper describes a machine vision system for inspecting hydraulic hose assemblies. At present the inspection in this industry is done manually and is prone to human error. A specially designed hose gripping mechanism, a mandrel system, and a camera and lighting fixture allows the system to be integrated on a shop floor and is able to inspect various parameters of a fitting. The system allows the inspection to be done more accurately and improves the quality control process.

Virtual Reality Promises New Design Capabilities

Virtual reality has received much attention in the past few years in relation to entertainment and simulation. Although there is much hype surrounding virtual reality, the underlying technologies and concepts could have a significant impact on computer-based design engineering tools. This paper explores the possible applications of virtual reality to the engineering design process. The objective of the described research is to explore opportunities to create tools, environments, and systems which will increase the designer’s productivity through the application of virtual reality (VR) technologies and concepts. Background on design, CAD and VR is presented to give a context for later proposals. Previous research in the application of VR to CAD is reviewed. Specific VR capabilities are presented to show their possible application in mechanical design and CAD systems specifically. The potential applications of VR are explored in a structured manner with examples for each category. The paper concludes with recommendations on further research and development directions.

Form Feature Based Design of Turned Parts

A form feature based modeller implemented on a personal computer is developed to design turned parts. The interactive environment provides the designer with a means of conveying the design intent and functionality without manipulating low-level graphics. Geometric defaults, tolerances, and warnings are embedded in the system and may be expanded for effective design for manufacture. The implementation of the modeller and usefulness of feature based design is discussed. This modeller is implemented in AutoLisp to run within AutoCAD.

Interference Detection of Non-Convex Solids for Manipulators

When performing a computer simulation on analytical study of robot motions it is possible to unwittingly require a part of the robot (e.g. the hand) to interpenetrate (i.e. to interfere with) another part (e.g. an arm). It is therefore important to be able to predict in advance whether self interference or collision of any type occurs. This problem arises in fields of interest other than robotics, e.g. computer aided design and computer graphics. In this report, we have developed a computational method which predicts interference of moving objects in space. The method works for non-convex solids and multiply-connected solids (solids containing holes). The method checks the boundaries of surfaces enveloping solids for interference. Every pair of surfaces (one on each body) are examined for points of intersection. Points of interest are then studied to determine whether any two solids do interfere. The theory is developed for planar, ruled, and double curved surfaces.