1 March 18-21, 2013 Prof. Nobuyuki IWATSUKI Dept. of Mechanical Sciences and Engineering. Tokyo Institute of Technology Lecture at IIITDM Jabalpur EM624b: Design and Control of Redundant Robots (Lecture 1) 1. Introduction of lecturer Please visit my WEB-site: http://www.dynamics.mep.titech.ac.jp/english/index.html or Search by Google with a keyword ‘Iwatsuki-Okada lab’
16
Embed
March 18-21, 2013 Lecture at IIITDM Jabalpur EM624b ... · Lecture at IIITDM Jabalpur EM624b: Design and Control of ... ... (1)Structural/motion flexibility is …
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
March 18-21, 2013
Prof. Nobuyuki IWATSUKIDept. of Mechanical Sciences and Engineering.
Tokyo Institute of Technology
Lecture at IIITDM Jabalpur
EM624b:Design and Control of Redundant Robots
(Lecture 1)
1. Introduction of lecturer
Please visit my WEB-site:http://www.dynamics.mep.titech.ac.jp/english/index.html
orSearch by Google with a keyword ‘Iwatsuki-Okada lab’
2
Nobuyuki Iwatsuki, Professor, Dr. Eng.
Affiliation:Dynamics Engineering LaboratoryDepartment of Mechanical Sciences and EngineeringGraduate School of Science and EngineeringTokyo Institute of Technology
My policy for research and education:(1)Confront everything actively!
- Think after trying!- Make it by yourself if you can purchase it!
(2)Believe your infinite possibility!- The merit exists just near the demerit.- Always think about the conversion of conception.
(3)Develop gentle machinery to mankind- Not only for industrial application
but also for human daily life- Flexible and intelligent mechanical system
Research Themes:(A)Synthesis and Control of Robotic Mechanisms
Especially focused on ‘Mechanism Design and Motion Control of Hyper Redundant Robots’
(B)Silent EngineeringEspecially focused on ‘Estimation of Sound Power Radiating from Vibrating Structure and Structural Optimization to Reduce the Sound Power’
(C)Intelligent Laser MeasurementEspecially focused on 'Development of a Laser Speckle Interferometer to Measure Multi-dimensional Displacement with High Accuracy and Speed'
4
Several Examples of My Research Works:(1)Biped Walking Machine
– While undergraduate/Ph.D student (-1987)
Straight Walk of Biped Walking Machine(1982)
Knee joint
Ankle joint
Toe
Heel
Hip joint
Leg Mechanism with 1 DOF
Ankle path generator(Planar 6-bar link mech.)
Foot driving mechanism(Planar 8-bar link mech.)
Driving links
Planar 8-bar link mechanism
Counter weightDrawing 8 figure
Plane-cylinder joint
Slanting crank shaft
Attitude Stabilizing Mechanism
950m
m
A Biped Waking Machine with only 1 DOF (Mechanical synthesis for simple control)
Walking speed :120 steps/min (World record)
2nd prototype which can startand stop walking
Improvement for non steady walking
○Change stride○Independent drive of foot driving mechanism○Arbitrary control of attitude stabilizing mechanism
A new prototype with 5 DOFItems Prototype-1 Prototype-2
Degree-of-freedom 1 5Weight 20.6kgf 48.8kgf
Straight walk Max. 120step/min Max.15 steps/min
Start/stop to walk - Sucessful
Dynamics is important not only to stabilize Biped machine but also to drive actuators !
5
(2)Dexterous Motion of Hyper Redundant Robot- Optimum learning control of snake-like robot
with 10DOF to avoid obstackles
Dexterity is set as an objective function togenerate optimum trajectory.
Photograph of a prototype (12DOF-12Actuators)
Linear potentiometer
Linearactuators25W,305<θ<440 mm
compound linkmade of aluminum
Fixed joint(Magnet base)
(3)Module-based Control of Redundant Multi-loop Mechanism
Module : Minimum unit to synthesize and control of whole mechanism
6
(1)Straight line trajectory (2)Ellipsoidal trajectory
Experimental results of CP control
Large scale mechanism can be controlled by using ‘kinematic modules’.
(4)Control of Structural Flexibility and Its application
Output point
Motion control of a flexible link (coil spring)taking account of its reaction force
7
xz
y
Four legged walking machine with spring leg
(5)Estimation of Sound Radiation from Thin Platesubjected to Acoustic Excitation
Thin rectangular plate
Acoustic excitation
Transmitted sound power
Transmitted sound power should be estimated to designsilent structure.
8
Pseudo random noise Impulsive signal
Transmitted sound power can be estimatedwith an adequate accuracy.
Due to mode of vibration
Input sound pressure Input sound pressure
Transmitted sound powerTransmitted sound power
(6)Estimation of Sound Radiation from Frog-type Guiro
The calculated acceleration and sound pressureExpanded
9
Laser speckle interferometer with the quad aperture method
The interference image like a distorted lattice moves two-dimensionally as a target moves.
Piezostack
CCD area image sensor
D
a
b
He-Ne Laser
Target
x-y Stage
Half mirror
X
Y
x
y φd
Convex lens
Amp.
A.D.C
RAM
PC
4 Aperture plate Fiber sensor
Objective lens
2D light intensity can bemeasured and its movement should be detected.
(7)Intelligent Laser Speckle Interferometer
Phase detector by real-time Fourier transformcomposed of analogue electronic circuit
10
(a)Circular trajectory (b)8-figure trajectory
)sin,cos(),( tnBtmAYX
Two-dimensional displacement can be measuredwith the proposed laser speckle interferometer.
Objectives of lectureRedundant robots which have redundant degrees-of-freedomcan generate intelligent, flexible and complicated motions.These motions are important especially for human daily life.The design and motion control methods based on kinematics dynamics analyses will be explained with several examples.
Schedule of lecture3/18 Mon 12:00-13:00 Introduction,
Advantages and issues to be solved forredundant robots
16:30-18:30 Optimum motion control of serial redundant manipulator
3/19 Tue 10:00-12:00 Synthesis of redundant closed-loop mechanisms
15:30-16:30 Control of redundant closed-loop mechanisms
3/20 Wed 9:00-11:00 Design of overactuator mechanisms15:30-16:30 Control of overactuator mechanisms
with elastic elements3/21 Thu 10:00-12:00 Motion control of redundant robot based
on reflex motion of each linkExamination
12
3. Advantages and Issues to be solved for redundant robots
A robot in industry
A robot in human daily life
High performance robot which can generate intelligent andflexible motion should be required in the future.
Aiming to develop high performance robots which will work in human daily life, the methods to synthesize and control new robotic mechanisms with hyper redundancy(HR) should be established.
Microprocessor
There will be a possibility to build a robot which has hundreds ofactuators and can achieve intelligent motions with adequate flexibility.
Objectives :
Neural computing
Remarkable advances in computer techniques(Hardware & Software)
“The robot may not be required in industries but must be required in human daily life.”