2001 ieee/asme international Conference on



Download 0.91 Mb.
Page29/38
Date18.10.2016
Size0.91 Mb.
#2673
1   ...   25   26   27   28   29   30   31   32   ...   38

W1A

Flexible Manipulators

Walking Machines

W1B

SALA PLATEA

09:30―11:10

SALA PASTA

Jurek Sasiadek, Canada

CHAIR

Michael McCarthy, USA

Luigi Villani, Italy

CO-CHAIR

Raffaele Di Gregorio, Italy

Rest-to-Rest Motion of a One-link Flexible Arm


Alessandro De Luca, Giandomenico Di Giovanni, Università di Roma "La Sapienza", Italy

We present a solution to the problem of finding the torque command that provides rest-to-rest motion in a given time for a one-link flexible arm. The basic idea is to design an auxiliary output such that the associated transfer function has no zeros. Planning a smooth interpolating trajectory for this output imposes a unique rest-to-rest motion to the whole arm with automatically bounded link deformation. The nominal torque is then obtained by inverse dynamics computation in the time domain. The method is presented for a linear model based on the Euler-Bernoulli beam description of the flexible link with dynamic boundary conditions. This approach lends itself to nonlinear extensions and feedback solutions.

09:30―09:50

Dynamically Stable Trajectory Synthesis for a Biped Robot during the Single-Support Phase


S. Marchese, G. Muscato, Università di Catania, Italy, G.S. Virk, University of Portsmouth, UK
The aim of this paper is to develop an algorithm to generate dynamically stable walking trajectories for a biped robot during the single-support phase, which can be used as reference trajectories for control. The biped is modeled as an anthropomorphic planar mechanical system of 7 links connected by 6 actuated rotoidal joints. We want the biped robot to move with a human-like behavior. In order to achieve this goal the kinematics and dynamics aspects have been treated separately. To obtain a human-like motion the biped has been regarded as a redundant manipulator, under the assumption that the supporting foot was fixed to the floor. In particular, the concept of task priority in relation to the inverse …

Rest-to-Rest Motion of a Two-link Robot with a Flexible Forearm


Alessandro De Luca, Giandomenico Di Giovanni, Università di Roma "La Sapienza", Italy

We consider the problem of finding the torque commands that provide rest-to-rest motion in a given time for the FLEXARM a two-link planar manipulator with a flexible forearm and nonlinear dynamics. The basic idea is to design a set of two outputs with respect to which the system has no zero dynamics. Planning smooth interpolating trajectories for these outputs imposes a unique rest-to-rest motion to the whole robot with bounded link deformations. The nominal rest-to-rest torque is obtained by standard inverse dynamics computation. In the multi-input nonlinear case this approach requires in general the use of a dynamic linearizing extension. Numerical results are presented and possible extensions discussed.

09:50―10:10

Locomotion of LAURON III in Rough Terrain


Bernd Gaßmann, Kay-Ulrich Scholl, Karsten Berns, Forschungszentrum Informatik, Germany

In particular the advantages of legged compared to wheeled locomotion arise in unstructured, uneven and rough terrain. Based on the sensor system the behaviour control of LAURONIII for walking in rough terrain is presented in this paper. To cope with terrain irregularities local behaviours like collision reaction and searching for ground are realised by an adaptation of the basic leg trajectories. Several global behaviours like the control of the body height and the centre of gravity influence the trajectories of all legs simultaneously.

End-Point Control of a Flexible-Link via Optimal Dynamic Inversion


Aurelio Piazzi, Università di Parma, Antonio Visioli, Università di Brescia, Italy
In this paper we propose a new method for the end-point control of a single flexible link. As a distinguished feature the methodology relies on an exact stable dynamic inversion analytically performed on a family of output functions. In this context the choice of ``transition polynomials'' as output functions permits to optimize the motion time subject to limits on the velocity and acceleration of the motor. Furthermore the technique is robust to unmodelled dynamics. Simulation results show the effectiveness of the approach.

10:10―10:30

Biomimetic and Smart Technology Principles of Humanoid Design


Darwin G. Caldwell, N. Tsagarakis, P. Artrit, J. Canderle, S. Davis, G.A. Medrano-Cerda, University of Salford, UK

Humanoid and bipedal robots are complex mechatronic systems requiring developments in mechanical/kinematic design, sensory systems, Artificial Intelligence, computing, control, communications and actuation/power systems. This complexity makes their onstruction a substantial challenge for workers in many branches of science and engineering and has prompted research inspired by the success of biological creations. Traditionally such mechanism design has avoided the use of flexible materials and structures because of the difficulties associated with operation and control. However, it is clear that the light flexible structures and designs common in nature, might offer many advantages if they could be used effectively. This paper will initially study the actuation requirements for a …

Adaptive Open-Loop Control Method for a Hydraulically Driven Flexible Manipulator


Janne Kovanen, Heikki Handroos, Lappeenranta University of Technology, Finland
A novel method applicable for the open-loop control of hydraulic driven flexible manipulators is presented. By using a new kind of adaptive control function instead of the traditional open-loop control strategies the dynamic behaviour of such systems can be significantly improved. The adaptive control function is developed for the open-loop control of a log crane equipped with an electrically controlled proportional directional valves and digital control system. The proposed method utilizes the existing hydraulic components and control hardware of commercially used log crane. In addition to this, the cylinder position and pressure transducers are used. The dynamics of manipulator can be determined by studying the …

10:30―10:50

Biomimetic Small Walking Machine


N. Kagawa, H. Kazerooni, University of California at Berkeley, USA

The goal is to design and construct a biologically inspired, small, robust, and fast walking machine. This paper explains how two important observations from cockroaches can be employed as design guidelines to design and build small walking machines. The first observation is the fact that cockroaches use very little feedback and control results from the properties of their parts and their morphological arrangement. This observation leads to our hypothesis that a robust and simple electromechanical walking machine should have implicit feedback control only within its hardware. The second observation is that the optimal walking frequency of a cockroach is close to the natural frequency of the cockroach. This observation leads us to hypothesize that the walking frequency of an electromechanical cockroach also needs to be the natural frequency of the …

Gradient-Descent Based Trajectory Planning for Regulation of a Two-Link Flexible Robotic Arm


Gülay Öke, Yorgo İstefanopulos, Boðaziçi University, Turkey

In the control of flexible manipulators most work in the literature concentrates on the suppresion of link deformations. In this study the aim is not to suppress the deflections but to calculate joint angles to minimize the tip position error. A gradient descent method is utilized off-line for the trajectory planning of a two-link flexible manipulator under gravity. Fuzzy logic control with gravity compensation is applied for the position control of the manipulator where the reference inputs are the values of the joint angles as determined by the trajectory planning procedure. Simulations have been carried out to show the performance of both the trajectory planning method and of the fuzzy controller in a regulation problem. Results are illustrated both in the joint space and in the operational space.

10:50―11:10

An Integrated Approach for Locomotion and Visual Control of a Service Hexapod via CNNs


P. Arena, Università di Catania, M. Branciforte, R. Caponetto, STMicroelectronics, A. Rizzo, G. Peschiera, Università di Messina, Italy

In this paper Cellular Neural Networks (CNN’s) are widely used, both for imaging processing (classical application) and for generation of locomotion patterns. In particular, a hexapod walking robot is presented. The locomotion patterns are generated by a RD-CNN and a visual serving approach exploiting an external camera is adopted to control the trajectory.




W1C

Mechatronics Curricula

DES’s in Automated Transportation

W1D

SALA BIANCA

09:30―11:10

SALA TURCA

Wayne Book, USA

CHAIR

Bart De Schutter, The Netherlands

Paolo Dario, Italy

CO-CHAIR

Luca Ferrarini, Italy

The Role of Mechanical Design in Mechatronics Education


John F. Gardner, Boise State University, USA

A survey of Mechatronics course offerings around the United States shows a rich variety of approaches to this topic. Indeed, it appears that a majority of Mechanical Engineering programs across the country offer some formal offering in this area and it seems clear that the vast majority of Mechatronics courses are offered through Mechanical Engineering departments. With the traditional emphasis on design skill in mechanical engineering, it would seem obvious that mechatronics courses would feature a major design component. Surprisingly, very few existing mechatronics courses (at least those reported in the literature) feature a strong emphasis on mechanical design. That’s is not to say that the creative process of design is not featured in Mechatronics courses, because it clearly is. It seems equally clear, however, that that accepted definition of mechatronics (“…the synergistic …

09:30―09:50

Modeling and Specification of Discrete Event Systems using Combined Process Algebra and Petri Nets


Petter Falkman, Bengt Lennartson, Michael Tittus, Chalmers University of Technology, Sweden
A modeling framework for general routing and resource booking problems is presented. The task is to specify desired routes for individual objects (products data packets vehicles) which are to be served by a number of shared resources (machines computers communication links). Based on simple booking models for the resources and routing specifications for the objects a controller that synchronizes the objects utilization of the available resources is automatically generated. A high level language is presented in order to simplify the specification of desired routes. This modeling language combines Petri nets and process algebra. Process operators are introduced for specification of alternative and …

Mechatronics and Early Mechanical Design


Thomas R. Kurfess, Georgia Institute of Technology, USA

The issue targeted in this paper is the development of a set of studio projects for a sophomore level course that integrate mechanical and electrical systems. The paper discusses the types of projects that can be run and how they are used to foster mechanical and electronic design. For example issues such as sensor specification are critical and teach the students how important and sometimes difficult «specing out» a design can be. From an electronics perspective a simple and an inexpensive micro-controller (the BASIC Stamp) is used. A variety of pre-designed modules are supplied to the students as well as code snippets to run these devices. The most interesting part of the paper is the design challenges that are afforded to the student. These challenges are in the form of mechanical system design and implementation in our small scale machine shop located in the studio …

09:50―10:10

Deadlock Avoidance in Automated Guided Vehicle Systems


Maria Pia Fanti, Biagio Turchiano, Politecnico di Bari, Italy

Automated Guided Vehicle Systems (AGVSs) are the most flexible means to transport pieces among workstations of an Automated Manufacturing System (AMS). The control of such systems must be in charge of avoiding collisions and deadlock conditions. The paper formulates a control scheme to face this problem. In particular, some algorithms are proposed to control in real time path assignments to the vehicles and their moves from zone to zone in the system. A final discussion compares the proposed approach to other methods offered by the recent literature.

Postgraduate and Undergraduate Mechatronics' Courses at the University of Wollongong


C.D. Cook, F. Naghdy, F. DeBoer, University of Wollongong, Australia
This paper outlines the University’s involvement with Industry based manufacturing projects, and how this has lead to the recent establishment of postgraduate and undergraduate Mechatronics degrees. The nature of the industrial projects will be described with examples of specific problems, test equipment and experimental rigs given. This work will be used to explain the reasons for the design of the mechatronics courses at Wollongong. A new teaching methodology particularly suitable for mechatronc’s education is also discussed.

10:10―10:30

Modelling Automated Material Handling in Intermodal Terminals


Cristiana Degano, Angela Di Febbraro, Politecnico di Torino, Italy
A model of an intermodal terminal in which freight only arrives and departs stored into

containers is presented in this paper. The container terminal is thought of as divided into areas each one associated with a kind of container handling operation. The complexity of the system to represent is further increased by the fact that the material transportation system inside the terminal is (at least) partially automated. For such a transportation system made up of different kinds of (unmanned) vehicles a Petri-net model is proposed with the objective of regulating its faulty behaviours. An example relevant to a case study is reported to show a practical application of the proposed model.

Development of Mechatronics Tertiary Education in Six Different Thai Universities


M. Yousef Ibrahim, Monash University, Australia
This paper presents the experience gained from Mechatronics course development in both Australia and Thailand. It also presents the educational philosophy adopted to produce a new breed of engineers. The educational philosophy was based on engaging industry in course development and subjects delivery. This was to meet the new Millennium's demands for more intelligent products. This paper outlines the mechanism adopted to achieve the course objectives through close industrial collaboration.

10:30―10:50

Modeling of AS/RS via Coloured Petri Nets


F. Basile, C. Carbone, P. Chiacchio, Università di Napoli Federico II

In this paper a colored timed Petri net approach to automated storage/retrieval systems (AS/RSs) modeling is presented. The AS/RS considered here consists in a number of rack structures with cranes and shuttles used to transfer pallets from main buffer stations to rack locations. The model is designed so that it can be used not only for off-line evaluation of a given management strategy based on simulation but also for on-line system monitoring and dynamic dispatching control. The plant model has been decomposed in six modules which communicates via fusion places. Token colors has been properly defined in order to include information which completely characterize the pallet and the mission in which it is involved so as to take into account the communication with a higher level mission dispatcher. In addition this fact allow us to render completely local any decision about …

A New Approach to Mechatronics Systems Design Using Hardware in the Loop Simulation


Devdas Shetty, Univ. of Hartford, Richard A. Kolk, United Technologies Corp., Jun Kondo, Claudio Campana, Univ. of Hartford, USA

This paper presents an approach to mechatronics system design course, which is primarily based on a combination of mechanical, electrical and software elements along with the design and construction of an intelligent mechatronics technology demonstrator. Topics covered in this course are principles of transducers and sensors and how to interface them with an external process in a computer environment. Methods of system modeling, real time interfacing and rapid prototyping are addressed. Computer interfacing work includes signal conversation, interface components, and real time application of computer systems to problems in manufacturing. The goals of this paper are to describe (1) the University of …

10:50―11:10

Machine Layout Problem in FMS Design


Imen Chaieb, Ouajdi Korbaa, Jean-Claude Gentina, Ecole Centrale de Lille, France
This paper deals with the transport system design in automated FMS. We describe the different phases for the design of such a system and explain more in details the step consisting in laying out machines in a manufacturing cell. The machine layout problem was treated under several assumptions in the literature, and our contribution consists in introducing flexibility in the production system illustrated by the consideration of transformation resources existing in many exemplars. The positioning of machines is evaluated by the criterion of the minimization of transportation tasks duration, and formulated as a mixed quadratic problem resolved by a mathematical solver (CPLEX).

W2A

Service Robots

Human―Machine Interfaces I

W2B

SALA PLATEA

11:30―13:10

SALA PASTA

Alessandro De Luca, Italy

CHAIR

Akitoshi Itoh, Japan

Peter Putz, The Netherlands

CO-CHAIR

Darwin Caldwell, UK

CleaningAssistant — A Service Robot Designed for Cleaning Tasks


Fabrizio Marrone, Matthias Strobel, FAW Ulm, Germany

A new service robot designed for cleaning tasks in home environments is introduced. Robot systems will work directly with people in these areas thus placing a central importance on making interactions between people and machines as natural as possible. The main focus of this paper is twofold: First an introduction to the system's design and to an intuitive programming approach which allows the robot to be easily used by non-experts is given. The approach is based on human gesture recognition and context sensitive interpretation. Second in order to transform full coverage paths planned in workspace to paths in the robot's configuration space a closed form solution of the robot's inverse kinematics

11:30―11:50

Dynamic Modeling of Primary Commands for a Car Simulator


A. Frisoli, C.A. Avizzano, M. Bergamasco, Scuola Superiore Sant’Anna, S. Data, C. Santi, Centro Ricerche Fiat, Italy
Simplified dynamic analytical models of primary commands of a car i.e. steer wheel and gearshift are identified and developed in the paper. The dynamic models are used to design the control law of force feedback devices which will be integrated in a car simulator. The model simulation results match satisfactorily with the experimental available data. An experimental partial assessment of control law for the gearshift simulation has been performed with a commercially available force-feedback joystick. The gearshift simulation control is implemented with an hybrid model based on a state machine. The results are presented and discussed.

A Robotic System for Total Knee Replacement


Andrea Malvisi, Maurilio Marcacci, Sandra Martelli, Istituti Ospedalieri Rizzoli, Gianni Campion, Paolo Fiorini, Università di Verona, Italy

This paper describes a new robot developed specifically for invasive orthopedic surgery. The design has been a cooperative development among physicians and engineers to achieve the necessary functionality. Target surgery is total knee replacement and the procedure is integrated with computerized data acquisition and path planning for off-line pre-operative planning, and with on-line intra-operative real-time controls for adjustments during surgery. The paper describes the main features of the robotic system in terms of me chanical, programming and functional characteristics. The rationale of the design decisions is presented and current experimental plans are discussed.

11:50―12:10

Modeling of Drivers' Longitudinal Behavior


Johan Bengtsson, Rolf Johansson, Lund Institute of Technology, Agneta Sjögren, Volvo Technological Development Corporation, Sweden

In the last years, many vehicle manufacturers have introduced advance driver support in some of their automobiles. One of those new features is Adaptive Cruise Control (ACC), which extends the conventional cruise control system to control of relative speed and distance to other vehicles. In order to design an ACC controller it is suitable to have a model on drivers’ behavior. Our approach to find dynamical models of the drivers’ behavior was to use system identification. Basic data analysis was made by means of system identification methodology, and several models of drivers’ longitudinal behavior are proposed, including both linear regression models and subspace based models. Detection when a driver is changing his behavior in various situations to a deviant behavior is useful. To that purpose a GARCH …


Download 0.91 Mb.

Share with your friends:
1   ...   25   26   27   28   29   30   31   32   ...   38




The database is protected by copyright ©ininet.org 2024
send message

    Main page