Fusaomi Nagata

3D robot sander for artistically designed furniture

This chapter introduces a robotic sanding system for attractively designed furniture with free-formed surfaces. The system is developed using an industrial robot with an open-architecture controller. The system has two novel features. One is that the surface-following controller developed for robotic sanding allows the robot to simultaneously realize a delicate polishing force control and smooth position/orientation control. The other is that a conventional and complicated teaching task is not required at all. The proposed robot sander directly handles CL data generated from 3D CAD/CAM, so that such a teaching process can be omitted. It is demonstrated that the proposed system can successfully sand the curved surface of attractive furniture with an extremely good quality.

Preliminary simulation of intelligent force control

In this chapter, simulations of an impedance model following force control using generalized learning-based fuzzy environment model (GFEM) are presented for articulated industrial robots with an open-architecture controller. The desired damping, which is one of the impedance parameters, has a substantial effect on force control performance, so that an important point for successfully using the force controller is how to suitably tune the desired damping according to each environment and task. We introduce an approach that produces the desired time-varying damping, giving the critical damping condition in contact with an object. However, the approach requires that the physical parameters of the object, such as viscosity and stiffness, are known beforehand. In order to deal with unknown environments, the proposed GFEM not only estimates the stiffness of unknown environments but also systematically yields the desired time-varying damping for stable force control. The effectiveness and promise of the proposed method are demonstrated through hybrid position/force control simulations using the dynamic model of a PUMA560 manipulator.

3D machining system for artistic wooden paint rollers

In this chapter, an intelligent machining system based on an NC machine tool with a rotary unit is introduced to efficiently and easily produce many kinds of artistically designed wooden paint rollers. A post-processor with the fuzzy feed rate generator is also presented to effectively make the machining system run. The post-processor allows the NC machine tool to easily transcribe a relief design from the surface on a flat model to the one on a cylindrical workpiece. The fuzzy feed rate generator generates suitable feed rate values according to the curvature of each relief design. Experimental results show that artistically designed wooden paint rollers can be successfully carved using the proposed machining system.

CAM system for articulated-type industrial robot

In this chapter, a CAM system for an articulated-type industrial robot RV1A is proposed in order to raise the relationship between a CAD/CAM and industrial robots spread to industrial manufacturing fields. A simple and flexible CAM system is designed from the viewpoint of the cooperation between CL data and robotic servo system. The CAM system required for an industrial robot is realized as an integrated system including not only a conventional main-processor of CAM but also a robotic servo system and kinematics, so that the RV1A can be directly controlled without any conventional teaching process. The basic design of the CAM system and experimental results are shown using the industrial robot RV1A. The effectiveness is demonstrated through a trajectory following a control experiment along multi-axis CL data consisting of position and orientation components and passive force control experiments.

Polishing robot for pet bottle blow molds

In this chapter, a CAD/CAM-based position/force controller in Cartesian space is presented for a mold polishing robot. CL data with normal vectors are referred to as not only the desired trajectory of tool translational motion but also the desired contact direction given to a mold surface, so that a complete non-taught operation of both position and contact direction can be realized. The controller also regulates the polishing force consisting of the contact force and kinetic friction forces. When the robot carries out polishing tasks, the position control loop delicately contributes to the force control loop to achieve both a regular pick feed control along CL data and a stable polishing force control on curved surface. The effectiveness and promise of the polishing robot is proved by actual polishing experiments using an industrial robot YASKAWA MOTOMAN UP6.

Velocity-based discrete-time control system with intelligent control concepts for openarchitecture industrial robots

A simulation technique of velocity-based discrete-time control system for open-architecture industrial robots is presented by giving and combining examples of intelligent controls such as genetic algorithms, fuzzy control and neural networks. In order to develop a novel control system for an open-architecture industrial robot, it is required with regard to safety, cost and simplicity to preliminarily examine and evaluate the characteristics and the performance. The proposed simulation technique has a high applicability.