Hyper Redundant Manipulator Using Compound Three-Bar Linkages
KoichiKoganezawa, TakashiKinoshita, Tokai University, Japan
A new mechanism for hyper redundant manipulator (HRM) is presented, which comprises of serially assembled compound three-bar linkages (CTL). The CTL mechanism has some unique properties. This paper presents the forward and inverse kinematics of this mechanism and shows the simulation of the HRM with 12 CTL units. The recursive algorithm of the inverse kinematics that the author originally developed is employed. It also presents the method and the simulation of the dynamical analysis. Holonomic constraints are sufficiently sustained by the constraint stabilization method that the author developed. The mechanical structure of the HRM having some CTL units that is under construction is shown.
09:30―09:50
A Web-based Telerobotic System for Research and Education at Essex
Lixiang Yu, Pui Wo Tsui, Quan Zhou, Huosheng Hu, Univ. of Essex, UK
This paper describes steps toward building a web-based telerobotic system for both research and teaching in the Essex University. The system has standard network protocol and interactive human-machine interface. Using a Web browser, a remote operator can control and/or program a mobile robot to navigate in our laboratory while receiving visual feedback and simulated local perceptual map via the Internet. The employment of an intuitive user interface enables both researchers and students to control and program mobile robots and to do some experiments remotely.
The 'Elephant Trunk' Manipulator, Design and Implementation
M.W. Hannan, I.D. Walker, Clemson University, USA
Over that last several years, research in the area of hyper-redundant robots and those robots based on tentacle/trunk type manipulators has become of great interest.Though much progress has been obtained in both the areas of robot construction and robot kinematics, there has been little progress in the implementation of a kinematic model for motion on a physical robot. In this paper, we review both our ‘elephant trunk’ robot’s construction and our previously designed kinematic model. We provide experimental results verifying not only the legitimacy of our kinematic model, but also the real time implementation of the model on the physical robot.
09:50―10:10
Internet Based Teleoperation using Wave Variables with Prediction
Saghir Munir, Wayne J. Book, Georgia Institute of Technology, USA
Wave-based teleoperation has been previously attempted over the Internet, however performance rapidly deteriorates with increasing delay. This paper focuses on the use of a modified Smith predictor, a Kalman filter and an energy regulator to improve the performance of a wave-based teleoperator. This technique is further extended for use over the Internet, where the time delay is varying and unpredictable. It is shown that the resulting system is stable even if there are large uncertainties in the model of the remote system (used in prediction). Successful experimental results using this technique for teleoperation in a master-slave arrangement over the Internet, where the control signal is streamed between Atlanta (Georgia) and Tokyo (Japan), are also given.
The Frenet-Serret Manipulator
Hiromi Mochiyama, Shin'ichirou Hiramatsu, Yasuchika Mori, National Defense Academy, Japan
The Frenet-Serret manipulators are roboticarms for the whole-arm manipulation. This type of manipulator has the kinematic structure same as Frenet-Serret apparatus of a spatial curve which is a tool for expressing the geometric features of the curve in classical differential geometry. We developed a Frenet-Serret manipulator with three joint units each of which has roll-pitch two degree-of- freedom. In this paper, we show the design concept, specification, mechanical structure and driving mechanism of the developed Frenet-Serret manipulator. We also show a recursive regressor form of the dynamics which will be utilized for system identification and control of the manipulator.
10:10―10:30
Man-Machine Interface using Steering Wheel and Pedals for a Quadruped Walking Robot
For an easier and natural teleoperation of the quadruped walking robots, a user interface is developed with a steering wheel and an assembly of pedals. In order to change the speed and the direction of the quadruped walking robots according to the operation of the steering wheel and the pedals, an on-line gait transition method is developed. Using this human-machine interface and on-line gait transition, the operator can control the quadruped walking robots almost in the same way as a driver drives a car. A quadruped walking robot named JROB-2 is used in the experiments. As the steering wheel is also equipped with various control buttons, it is possible to assign various single command tasks, like “stand up” and …
The Design of High Performance Manipulators
H.M.J.R. Soemers, Philips Center for Industrial Technology, The Netherlands
Two case studies are given on the design of high performance manipulators both performed at the Philps CFT (Centre for Industrial Technology). The first case deals with a double scara type manipulator the second with a linear motor based gantry robot for SMD placement. Emphasis is laid on a design approach aimed at achieving outstanding performance in a predictable way. The predictability was achieved by modelling geared to support design decisions. Also the designs themselves had a great deal of predictable properties thanks to the kinematic principles followed.
10:30―10:50
Micro Teleoperation with Parallel Manipulator
Noriaki Ando, Masahito Ohta, Kouhei Gonda, University of Tokyo, Hideki Hashimoto, University of Tokyo & Presto JST, Japan
This paper describes the haptic interface based micromanipulation systems developer for micro robot fabrication. It is based on fabrications, micro process handling and micro components inspection. It is based on scaled bilateral tele-operation system between different structures. These systems are composed of an original 6DOF parallel link manipulator to carry out micromanipulation and a haptic interface with force feedback. These systems aim to make the micromanipulation more productive constructing a better human interface through the micro envoroment force and scale expansion. In this paper, new micromanipulation systemsis proposed and evaluated the analysis is conducted performing fundamental experiments to evaluate the precision and manipulability.
A special class of platform manipulators is the subject of this paper. These manipulators comprise two platforms connected by three legs, each being composed of three revolute (R) and one spherical (S) joints, which gives the manipulator six degrees of freedom. Hence, two actuators are required per leg. Under the assumption that the two R joints proximal to the fixed platform are actuated, we derive the differential kinematic relations between actutor joint rates and mobile-platform twist. This model comprises two Jacobian matrices, the forward and the inverse-kinematics Jacobians. These relations are then applied to the singularity analysis of the parallel manipulator developed at Singapore's Gintic Institute of Manufacturing Technology and Nanyang Technological University.
10:50―11:10
A Generalized Control Approach for Ideal Teleoperation
The master-slave manipulator is capable of extending human dexterous skills to the distant or dangerous environment. In order to maintain its operational efficiency, working conditions at the slave side need to be honestly transferred to the master side. This paper presents a generalized control strategy for the ideal teleoperation of a master-slave manipulator. The proposed control framework fully applies position and force signals at both the master and the slave ends, and can bring about transparent property of ideal teleoperation. Better position tracking performance is demonstrated by teleoperation experiments on a single-link master-slave telemanipulator interacting with environment.
M1C
Motion and Imaging
Transportation Systems
M1D
SALA BIANCA
09:30―11:10
SALA TURCA
Takanori Shibata, Japan
CHAIR
Manfred Hiller, Germany
Antonello Cutolo, Italy
CO-CHAIR
Angela Di Febbraro, Italy
Design and Analysis of an Absolute Non-Contact Orientation Sensor for Wrist Motion Control
Harry Garner, Georgia Institute of Technology, Martin Klement, Lexmark Inc., Kok-Meng Lee, Georgia Institute of Technology, USA
In this paper, aspects of the design and analysis of a vision-based non-contact absolute orientation sensor with application to wrist-like actuator motion control will be presented. This work is primarily motivated by the need for the measurement of the absolute orientation of a spherical rotor present in wrist-like actuators for the motion control of such systems. Current orientation measurement approaches increase the complexity of the mechanical structure through additional contact with the spherical rotor. To eliminate this additional complexity, a non-contact vision-based technique that realizes an absolute orientation measurement using a specially designed grid pattern that allows the absolute orientation formation …
09:30―09:50
Safeness-Enforcing Supervisory Control for Railway Networks
Fabrizio Diana, Alessandro Giua, Carla Seatzu, Università di Cagliari, Italy
In this paper we deal with the problem of modeling railway networks with Petri nets so as to apply the theory of supervisory control for discrete event systems to automatically design the system controller. We provide a modular representation of railway networks in terms of stations and tracks including sensors and semaphores. We ensure safeness and local liveness imposing both Generalized Mutual Exclusion Constraints and constraints also involving the firing vector.
Motion Tracking of a Part on a Vibratory Feeder
Winncy Y. Du, San José State University, USA
This paper addresses the motion tracking of a part on a vibrating plate feeder. A fast boundary extraction routine is first introduced to search a part of interest by investigating several scan lines. Once the first three boundary points of the part – forming a seed segment – are located, the boundary tracing and growing process can be quickly and accurately implemented inside a sequence of small regions. The pose vector [x,y,] that represents the location and orientation of the part is estimated based on the part’s boundary information. The motion tracking procedure consists of the four phases: 1) predict the next region that the part will appear using the previous and current pose vectors; 2) extract the part's boundary within the predicted region using the same boundary-extraction routine; 3) update the pose vector by calculating the new centroid and orientation; and 4) feed the new pose …
09:50―10:10
Model Predictive Control for Railway Networks
B. De Schutter, T. Van den Boom, Delft University of Technology, The Netherlands
Model predictive control (MPC) is a very popular controller design method in the process industry. Usually MPC uses linear discrete-time models. In this paper we extend MPC to a class of discrete-event systems with both hard and soft synchronization constraints. Typical examples of such systems are railway networks, subway networks, and other logistic operations. In general the MPC control design problem for these systems leads to a nonlinear non-convex optimization problem. We also show that the optimal MPC strategy can be computed using an extended linear complementarity problem.
Cheng-Lun Chen, George T.-C. Chiu, Purdue University, USA
This paper proposed a new process control strategy for reducing banding artifacts in electrophotographic (EP) processes. EP banding artifact is shown to correlate to the fluctuation of the organic photoconductive (OPC) drum angular velocity. Improved regulation of the OPC drum rotational velocity under various process uncertainty and variations will significantly improve EP process stability and reduce the appearance of banding. The proposed control strategy includes two levels of OPC drum speed regulation. The first level utilizes a loop shaping technique to incorporate a human visual system (HVS) model into the control loop to eliminate low frequency and non-periodic drum velocity fluctuation. The second level uses an internal model based repetitive controller to reduce the effect of periodic velocity fluctuations. The HVS based loop shaping design is intended to address …
10:10―10:30
Control of Transient Phase for Discrete Event Simulation using Computational Steering
T. Kesavadas, Abhishek Sudhir, SUNY Buffalo, USA
Traditional steps in a discrete event manufacturing simulation are to prepare input variables, select simulation parameters, run the simulation and review the results after the execution is completed. In a new approach proposed here, we have utilized an interactive simulation to obtain improved solutions and instant feedback from the proposed production system. A Computational Steering tool is provided to give the user increased control of the simulation. This interactive tool gives the user power to view the simulation results dynamically and to control the simulation in real-time. A case study that uses Computational Steering to reduce the length of the transient phase in steady-state analysis of simulation is presented.
A Robust Machine Vision System Design to Facilitate the Automation of Surface Appearance Inspections
Johné M. Parker, University of Kentucky, USA
The performance demands on coated surfaces (e.g., automotive paints) are considerable. Though the primary purpose of automotive paints is to protect the car body from corrosion, it has been well established that the appearance of a painted surface greatly affects a customer’s perception of that product’s quality. Automotive manufacturers spend considerable sums on the painting process during manufacturing and, again, on warranty-covered claims of paint-related problems. Therefore, robust methods to both characterize and monitor surface quality are critical. Existing quality control methods using expert inspectors are generally effective in assessing perceived appearance; however, they are often labor intensive, time-consuming and can be subjective. Optical techniques conducted … are
H. Kobayashi, M. Higuchi, K. Kikuchi, Science University of Tokyo, Japan
In this paper, we present the traffic signal control method in order to avoid traffic jam by applying the simulator for the interpersonal relation by Egograms. Egograms that is consisted of 5 categories such as critical parents (CP), nurturing parents (NP), adult (A), free child (FC) and adapted child (AC), .indicates personality and is used for counseling by analyzing magnitudes of 5 categories and also for estimating the interpersonal relation. In this simulator, we prepare the computer agent which has Egograms obtained by a human subject. We provide food acquisition environment which is an instinct for survival, and then we arrange 11 actions for the agent that might be occurred around feeding stations. By analyzing actions selected by plural agents in simulation, we can diagnose the interpersonal relations. For the traffic signal control, we consider the traffic signal to be an agent and instead of …
A New Omnidirectional Vision Sensor for the Spatial Semantic Hierarchy
Emanuele Menegatti, Univ. Padova, Italy, Mark Wright, University of Edinburgh, UK, Enrico Pagello, Univ. Padova & LADSEB, Italy
In this paper we propose a new sensor for the Spatial Semantic Hierarchy created by Benjamin Kuipers. To prove the effectiveness of this new sensor it has been used as a sole sensor for a robot. The task of the robot is to build a topological map of an unknown environment using the Spatial Semantic Hierarchy. In the paper we present the strict link that it is possible to create between the Spatial Semantic Hierarchy and the omnidirectional images. We propose a set of topological events that it is possible to identify in the sequence of images acquired while the robot moves. These topological events can be used to pose a discrete set of places that will be the nodes of the topological map. The results of simulated experiments …
10:50―11:10
Generalization of the Cascade Principle in View of a Structured Form of Mechatronic Systems
Joachim Lückel, Thorsten Hestermeyer, Xiaobo Liu-Henke, Universität Paderborn, Germany
With the steady increase in microcontroller performance, mechatronic systems have become more and more complex. This requires new organizational structures both in system design and in controller-structure. For system design, a practical way of structuring mechanical systems vertically and horizontally has already been presented in a number of papers. This structure has to be reflected in the controller as well. The aim of this paper is to adapt and combine well-known controller design techniques to make up a new and well-structured way of controller design tailored to the organizational structure of mechatronic design. After a brief overview of the mechatronic design structure, this new controller design technique will be presented basis of the cascade structure of SISO controllers and the centralized controllers with state-vector feedback. For an example, the controller of a new modular railway system is presented, with special emphasis on the active suspension/tilt system.
M2A
Parallel Manipulators
Intelligent Systems I
M2B
SALA PLATEA
11:30―13:10
SALA PASTA
Jorge Angeles, Canada
CHAIR
Clarence de Silva, Canada
Rezia Molfino, Italy
CO-CHAIR
Michele Aicardi, Italy
Design of a New 2 DOF Parallel Mechanism
Young-Hoon Chung, Jae-Won Lee, Yeungnam University, Korea
We propose a new two-degree of freedom parallel mechanism in this paper. This mechanism was designed in order to improve the kinematic performace and to achieve static balance. We use the pantograph mechanisms in order to change the location of active joints which leads to transformation a direct kinematic singularities into a non-singularities. The direct kinematic singular configurations of the proposed parallel mechanism occur near the workspace boundary such as those of the serial manipulators which means that we do not any more suffer from the singular configuration which occurs at the inside of workspace. Hence the proposed mechanism can overcome disadvantage such as the singular configurations existing in the inside of workspace for a typical parallel mechanisms. Using the property that position vector of rigid body rotating about a fixed point is normal to the velocity …
11:30―11:50
Mimesis Embodiment and Proto-symbol Acquisition for Humanoids
Tetsunari Inamura, Japan Science and Technology Corporation, Yoshihiko Nakamura, Iwaki Toshima, University.of Tokyo, Hideaki Ezaki, Kawasaki Heavy Industries Co., Japan
Mimesis is the primitive skill of imitative learning, one of the methods for recognition of others' behavior and construction of self behavior. Mimesis is thought an origin of human intelligence because this function is observed not only at humans but also at animals. When the mimesis is adopted as learning method for humanoids, convenience for designing full body behavior increase because bottom-up learning approaches from robot side and top-down teaching approaches from user side involved each other. Therefore, we have developed abehavior acquisition and understanding system for humanoids based on the mimesis. This system is able to abstract observed others' behaviors into conceptual symbols, to recognize …
