ME 115(b): Solution to Homework #1



Similar documents
Simulation of Trajectories and Comparison of Joint Variables for Robotic Manipulator Using Multibody Dynamics (MBD)

INTRODUCTION. Robotics is a relatively young field of modern technology that crosses traditional

ACTUATOR DESIGN FOR ARC WELDING ROBOT

Constraint satisfaction and global optimization in robotics

INSTRUCTOR WORKBOOK Quanser Robotics Package for Education for MATLAB /Simulink Users

INTRODUCTION. Chapter Robotics

Mechanics lecture 7 Moment of a force, torque, equilibrium of a body

Intelligent Submersible Manipulator-Robot, Design, Modeling, Simulation and Motion Optimization for Maritime Robotic Research

Section 1.1. Introduction to R n

Kinematics of Robots. Alba Perez Gracia

Robot Dynamics and Control

Exam 1 Sample Question SOLUTIONS. y = 2x

We can display an object on a monitor screen in three different computer-model forms: Wireframe model Surface Model Solid model

Lecture 14: Section 3.3

Practical Work DELMIA V5 R20 Lecture 1. D. Chablat / S. Caro Damien.Chablat@irccyn.ec-nantes.fr Stephane.Caro@irccyn.ec-nantes.fr

Metrics on SO(3) and Inverse Kinematics

RoboAnalyzer: 3D Model Based Robotic Learning Software

FURTHER VECTORS (MEI)

Synthesis of Constrained nr Planar Robots to Reach Five Task Positions

+ 4θ 4. We want to minimize this function, and we know that local minima occur when the derivative equals zero. Then consider

Force/position control of a robotic system for transcranial magnetic stimulation

Design of a six Degree-of-Freedom Articulated Robotic Arm for Manufacturing Electrochromic Nanofilms

discuss how to describe points, lines and planes in 3 space.

Parametric Curves. (Com S 477/577 Notes) Yan-Bin Jia. Oct 8, 2015

Mathematics Course 111: Algebra I Part IV: Vector Spaces

A PAIR OF MEASURES OF ROTATIONAL ERROR FOR AXISYMMETRIC ROBOT END-EFFECTORS

An Application of Robotic Optimization: Design for a Tire Changing Robot

Solving Simultaneous Equations and Matrices

GRAPHING IN POLAR COORDINATES SYMMETRY

Lewis, F.L.; et. al. Robotics Mechanical Engineering Handbook Ed. Frank Kreith Boca Raton: CRC Press LLC, 1999

FACTORING ANGLE EQUATIONS:

Least-Squares Intersection of Lines

Recall the basic property of the transpose (for any A): v A t Aw = v w, v, w R n.

Lecture 7. Matthew T. Mason. Mechanics of Manipulation. Lecture 7. Representing Rotation. Kinematic representation: goals, overview

1 VECTOR SPACES AND SUBSPACES

1 TRIGONOMETRY. 1.0 Introduction. 1.1 Sum and product formulae. Objectives

Matrix Differentiation

December 4, 2013 MATH 171 BASIC LINEAR ALGEBRA B. KITCHENS

x1 x 2 x 3 y 1 y 2 y 3 x 1 y 2 x 2 y 1 0.

Numerical Analysis Lecture Notes

Continuous Groups, Lie Groups, and Lie Algebras

CALIBRATION OF A ROBUST 2 DOF PATH MONITORING TOOL FOR INDUSTRIAL ROBOTS AND MACHINE TOOLS BASED ON PARALLEL KINEMATICS

Kinematics and Dynamics of Mechatronic Systems. Wojciech Lisowski. 1 An Introduction

A vector is a directed line segment used to represent a vector quantity.

Geometry of Vectors. 1 Cartesian Coordinates. Carlo Tomasi

Figure Cartesian coordinate robot

13 MATH FACTS a = The elements of a vector have a graphical interpretation, which is particularly easy to see in two or three dimensions.

5.3 SOLVING TRIGONOMETRIC EQUATIONS. Copyright Cengage Learning. All rights reserved.

Equations Involving Lines and Planes Standard equations for lines in space

Vector Algebra II: Scalar and Vector Products

Section 9.5: Equations of Lines and Planes

Design-Simulation-Optimization Package for a Generic 6-DOF Manipulator with a Spherical Wrist

11.1. Objectives. Component Form of a Vector. Component Form of a Vector. Component Form of a Vector. Vectors and the Geometry of Space

Computer Animation. Lecture 2. Basics of Character Animation

Operational Space Control for A Scara Robot

Section 11.4: Equations of Lines and Planes

MOVIES, GAMBLING, SECRET CODES, JUST MATRIX MAGIC

Math 241 Lines and Planes (Solutions) x = 3 3t. z = 1 t. x = 5 + t. z = 7 + 3t

Chapter 11 Equilibrium

Introduction to Computer Graphics Marie-Paule Cani & Estelle Duveau

9 MATRICES AND TRANSFORMATIONS

Vectors 2. The METRIC Project, Imperial College. Imperial College of Science Technology and Medicine, 1996.

Gear Trains. Introduction:

Linear algebra and the geometry of quadratic equations. Similarity transformations and orthogonal matrices

Dot product and vector projections (Sect. 12.3) There are two main ways to introduce the dot product

Section 2.4: Equations of Lines and Planes

1.3. DOT PRODUCT If θ is the angle (between 0 and π) between two non-zero vectors u and v,

Typical Linear Equation Set and Corresponding Matrices

Adding vectors We can do arithmetic with vectors. We ll start with vector addition and related operations. Suppose you have two vectors

LINEAR ALGEBRA W W L CHEN

Figure 1.1 Vector A and Vector F

3. Let A and B be two n n orthogonal matrices. Then prove that AB and BA are both orthogonal matrices. Prove a similar result for unitary matrices.

Design Project 2. Sizing of a Bicycle Chain Ring Bolt Set. Statics and Mechanics of Materials I. ENGR 0135 Section 1040.

1.5 Equations of Lines and Planes in 3-D

A QUICK GUIDE TO THE FORMULAS OF MULTIVARIABLE CALCULUS

Kinematics and force control for a Gantry-Tau robot

10.5. Click here for answers. Click here for solutions. EQUATIONS OF LINES AND PLANES. 3x 4y 6z 9 4, 2, 5. x y z. z 2. x 2. y 1.

12.5 Equations of Lines and Planes

A Direct Numerical Method for Observability Analysis

Matrix Calculations: Applications of Eigenvalues and Eigenvectors; Inner Products

Linear Algebra Notes for Marsden and Tromba Vector Calculus

The elements used in commercial codes can be classified in two basic categories:

3D Tranformations. CS 4620 Lecture 6. Cornell CS4620 Fall 2013 Lecture Steve Marschner (with previous instructors James/Bala)

Essential Mathematics for Computer Graphics fast

The Matrix Elements of a 3 3 Orthogonal Matrix Revisited

Math 241, Exam 1 Information.

Kinematical Animation

5 Systems of Equations

Lecture 1: Systems of Linear Equations

University of Lille I PC first year list of exercises n 7. Review

Matrix Normalization for Optimal Robot Design

Motion Control of 3 Degree-of-Freedom Direct-Drive Robot. Rutchanee Gullayanon

1.2 GRAPHS OF EQUATIONS. Copyright Cengage Learning. All rights reserved.

521493S Computer Graphics. Exercise 2 & course schedule change

Math 113 HW #7 Solutions

( 1) = 9 = 3 We would like to make the length 6. The only vectors in the same direction as v are those

Human-like Arm Motion Generation for Humanoid Robots Using Motion Capture Database

Alabama Department of Postsecondary Education

Transcription:

ME 115(b): Solution to Homework #1 Solution to Problem #1: To construct the hybrid Jacobian for a manipulator, you could either construct the body Jacobian, JST b, and then use the body-to-hybrid velocity transformation: JST h RST 0 JST b 0 R ST or recall that the columns of the hybrid Jacobian take the form: θ N JST h (1) ω 1 ω 2 ω N where the forward kinematics equations g ST ( θ) take the form: g ST ( θ) RST ( θ) p ST ( θ) and ω j is: ω j This solution will use the second approach. ( RST RST T θ j ). Manipulator (ii): While this manipulator has a rather odd geometry, it is relatively straightforward to tackle this problem by suitable choices of geometry in the Denavit- Hartenburg approach. If β is the angle between the first and third joint axes when the manipulator lies in the configuration shown in the book, and if l 1 and l 2 are the two link lengths as shown in the book s figure, then the D-H parameters for this manipulator are: a 0 0 α 0 0 d 1 0 θ 1 variable a 1 0 α 1 π 2 d 2 0 θ 2 variable a 2 l 1 cos β α 2 π 2 d 3 (l 1 + l 2 ) sin β θ 3 variable a 3 l 2 cos β α 3 0 d 4 0 θ 4 0 where we have assumed that the tool frame z-axis is parallel to joint axis 3. Recalling the the relationship between link frames in terms of the D-H parameters: cos θ i+1 sin θ i+1 0 a i g i,i+1 sin θ i+1 cos α i cos θ i+1 cos α i sin α i d i+1 sin α i sin θ i+1 sin α i cos θ i+1 sin α i cos α i d i+1 cos α i 0 0 0 1 1

The forward kinematics of this mechanism can be found as: g ST g S,1 g 1,2 g 2,3 g 3,T c 1 s 1 0 0 c 2 s 2 0 0 c 3 s 3 0 a 2 1 0 0 a 3 s 1 c 1 0 0 0 0 1 0 0 0 1 d 3 0 1 0 0 0 0 1 0 s 2 c 2 0 0 s 3 c 3 0 0 0 0 1 0 (2) 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 (c 1 c 2 c 3 s 1 s 3 ) (c 1 c 2 s 3 + s 1 c 3 ) c 1 s 2 (a 3 c 1 c 2 c 3 + a 2 c 1 c 2 + d 3 c 1 s 2 a 3 s 1 s 3 ) (s 1 c 2 c 3 + c 1 s 3 ) ( s 1 c 2 c 3 + c 1 c 3 ) s 1 s 2 (a 3 s 1 c 2 c 3 + a 2 s 1 c 2 + d 3 s 1 s 2 + a 3 c 1 s 3 ) s 2 c 3 s 2 s 3 c 2 ( a 3 s 2 c 3 a 2 s 2 + d 3 c 2 ) Following Equation (1), the hybrid Jacobian is: J h ST ( v1 v 2 v 3 ω 1 ω 2 ω 3 θ 3 RST T ) (3) where: ( a 3 s 1 c 2 c 3 a 2 s 1 c 2 d 3 s 1 s 2 a 3 c 1 s 3 ) v 1 (a 3 c 1 c 2 c 3 + a 2 c 1 c 2 + d 3 c 1 s 2 a 3 s 1 s 3 ) 0 ( a 3 c 1 s 2 c 3 a 2 c 1 s 2 + d 3 c 1 c 2 ) v 2 ( a 3 s 1 s 2 c 3 a 2 s 1 s 2 + d 3 s 1 c 2 ) ( a 3 c 1 c 2 s 3 a 3 s 1 c 3 ) v 3 ( a 2 s 1 c 2 s 3 + a 3 c 1 c 3 ) ( a 3 c 2 c 3 a 2 c 2 d 3 s 2 ) (a 3 s 2 s 3 ) 0 s 1 c 1 s 2 ω1 ω 2 ω 3 0 c 1 s 1 s 2 1 0 c 2 Manipulator (iv): This is the Stanford Manipulator. Recall from a previous homework solution that the Denavit-Hartenberg parameters and the forward kinematics are: a 0 0 α 0 0 d 1 0 θ 1 variable a 1 0 α 1 π 2 d 2 0 θ 2 variable a 2 0 α 2 π 2 d 3 variable θ 3 0 (constant) 2

g ST g S,1 g 1,2 g 2,3 g 3,T (4) c 1 s 1 0 0 c 2 s 2 0 0 1 0 0 0 s 1 c 1 0 0 0 0 1 0 0 0 1 d 3 0 0 1 0 s 2 c 2 0 0 0 1 0 0 (5) 0 0 0 1 0 0 0 1 0 0 0 1 c 1 c 2 s 1 c 1 s 2 d 3 c 1 s 2 s 1 c 2 c 1 s 1 s 2 d 3 s 1 s 2 s 2 0 c 2 d 3 c 2 RST (θ 1, θ 2, d 3 ) p ST (θ 1, θ 2, d 3 ) 0 T (6) 1 0 0 0 1 Following Equation (1), the hybrid Jacobian is: J h ST ( RST T d 3 s 1 s 2 d 3 c 1 c 2 c 1 s 2 d 3 c 2 s 2 d 3 s 1 c 2 s 1 s 2 0 d 3 s 2 c 2 0 s 1 0 0 c 1 0 1 0 0 d 3 ) (7) (8) (9) Solution to Problem #2: To find the singularities of the regional part (just the first three joints) of the elbow manipulator, one can determine the conditions under which the Jacobian matrix of the manipulator loses rank. While one could use any Jacobian, for simplicity we will use the Hybrid Jacobian matrix. You could either recall from the class note (or derive) the forward kinematics of the Elbow manipulator: g ST ( θ) RST ( θ) p ST ( θ) c 1 c 23 s 1 c 1 s 23 c 1 (l 2 c 2 + l 3 c 23 ) s 1 c 23 c 1 s 1 s 23 s 1 (l 2 c 2 + l 3 c 23 ) s 23 0 c 23 (l 2 s 2 + l 3 s 23 ) (10) 0 0 0 1 where c j cos(θ j ), s j sin(θ j ), c ij cos(θ i + θ j ), s ij sin(θ i + θ j ), etc. Recall that the hybrid Jacobian for the regional part of a manipulator is defined as: JST h (11) 3 pst θ

and thus substituting Equation (10) into Equation (11) yields: s 1 (l 2 c 2 + l 3 c 23 ) c 1 (l 2 s 2 + l 3 s 23 ) l 3 c 1 s 23 JST h c 1 (l 2 c 2 + l 3 c 23 ) s 1 (l 2 s 2 + l 3 s 23 ) l 3 s 1 s 23. (12) 0 (l 2 c 2 + l 3 c 23 ) l 3 c 23 Singularities will occur when the determinant of J h ST is zero: The singularities occur when: det(j h ST ) l 2 l 3 l 2 cos(θ 2 ) + l 3 cos(θ 2 + θ 3 ) sin(θ 3 ) θ 3 0. In this case, the arm is fully streteched out, and thus this singular configuration corresponds to the manipulator s outer workspace limit. θ 3 ±π. In this case, the arm is folded back on itself, and this singular configuration corresponds to the manipulator s inner workspace boundary. l 2 c 2 + l 3 c 23 0. Note from the forward kinematics equations that in this case, x and y coordinates of the tool frame origin lie at x 0 and y 0. This occurs when the tool frame origin is placed anywhere along the first joint axis. Solution to Problem #3: Part (a): From Problem 1 above we know that regional forward kinematics that relates the joint variables (θ 1, θ 2, d 3 ) of the Stanford manipulator to the Cartesian location of the tool frame origin is: x d 3 c 1 s 2 p ST y d 3 s 1 s 2 (13) z d 3 c 2 T x where p ST y is the location of the tool frame origin. The inverse kinematic solution z can be derived by simple algebraic manipulation of these kinematic relationships. 1. Step #1: x 2 + y 2 + z 2 d 2 3 d 3 ± x 2 + y 2 + z 2. There are two solutions to this equation. In practice, one of these solutions will almost certainly be infeasible from a mechanical point of view. 2. Step #2: x 2 + y 2 d 2 3s 2 x 2 s 2 ± 2 +y 2. There are 4 possible solutions to x 2 +y 2 +z 2 this equation. For each different of the two possible d 3 solutions there are two different θ 2 solutions. 4

3. Step #3: From the forward kinematics, we have that: cos θ 1 x d 3 sin θ 2 ; sin θ 1 from which we can get the unique solution: θ 1 There is a single θ 1 solution for each (θ 2, d 3 ) combination. y d 3 sin θ 2 ; (14) y x Atan2,. (15) d 3 sin θ 2 d 3 sin θ 2 Part (b): This problem is most readily solved by realizing that manipulators with wrists have a unique forward kinematic (and therefore inverse kinematic) solution structure. Note that a wrist is defined as a serial kinematic chain of three revolute joints whose joint axes all intersect at a single point, which we will term the wrist center. The wrist center will also be the origin of the references frames (as defined by the Denavit-Hartenberg convention) of those links that make up the wrist. Let the form of the forward kinematic relationship be denoted by: g ST RST (θ 1,..., θ 6 ) p ST (θ 1,..., θ 6 ) R D ST p D ST where RST D and pd ST are the desired orientation and position of the tool frame. Let g 6,T denote the location of the tool frame relative to the Denavit-Hartenberg reference frame of link 6 (whose origin will lie at the wrist point). g 6,T is a constant transformation that accounts for any difference in the location of the tool relative to the link frame. The desired position and orientation of link frame 6 will thus be: g D S,6 R D ST p D ST R6,T p 6,T 1 R D S,6 p D S,6 (16). (17) Note that the forward kinematics formula which describes the origin of Link frame 6 (the wrist center point) is only a function of (θ 1, θ 2, d 3 ), and is in fact the formula used in part (a) of this problem. Thus, (θ 1, θ 2, d 3 ) can be found from part (a), using p D S,6 is the desired position of the wrist center point. Note that the forward kinematics relationship that describes the orientation of link frame 6 is: R D S,6 R D S,T R 1 6,T R S,1(θ 1 ) R 1,2 (θ 2 ) R 2,3 R 3,4 (θ 4 ) R 4,5 (θ 5 ) R 5,6 (θ 6 ). (18) But (θ 1, θ 2, d 3 ) have already been determined by the placement of the wrist point at its desired location. Hence, we can rearrange Equation (18) to isolate the unknown joint variables (θ 4, θ 5, θ 6 ): R 3,4 (θ 4 ) R 4,5 (θ 5 ) R 5,6 (θ 6 ) (R S,1 R 1,2 R 2,3 ) 1 RS,T D R 1 6,T. (19) All of the terms on the right hand side of Equation (19) are known from the inverse kinematic problem state, the knowledge of g 6,T, and the inverse kinematic solution of the regional structure. 5

Note that the general form of the rotation matrix using the Denavit-Hartenberg convention is: cos θ i sin θ i 0 R i 1,i sin θ i cos α i 1 cos θ i cos α i 1 sin α i 1 (20) sin θ i sin α i 1 cos θ i sin α i 1 cos α i 1 For the Stanford manipulator, one choice 1 of the twist angles for the wrist substructure is: α 3 π/2, α 4 π/2, α 5 π/2. Hence: c 4 s 4 0 c 5 s 5 0 c 6 s 6 0 R 3,4 0 0 1 R 4,5 0 0 1 R 5,6 0 0 1 (21) s 4 c 4 0 s 5 c 5 0 s 6 c 6 0 and therefore: (c 4 c 5 c 6 s 4 s 6 ) (c 4 c 5 s 6 + s 4 c 6 ) c 4 s 5 R 3,4 R 4,5 R 5,6 s 5 c 6 s 5 s 6 c 5 (22) (s 4 c 5 c 6 + c 4 s 6 ) (s 4 c 5 s 6 c 4 c 6 ) s 4 s 5 Let the entries constant matrix R3,6 D (R S,1 R 1,2 R 2,3 ) 1 RS,T D R 1 6,T r 11 r 12 r 13 R3,6 D r 21 r 22 r 23. r 31 r 32 r 33 Equating terms, we can see that: cos θ 5 r 23 which has two solutions. Using these solutions, we see that: r 33 θ 4 Atan2, r 13 sin θ 5 sin θ 5 θ 6 Atan2 r 22 r 21, sin θ 5 sin θ 5 have the form: Hence, the wrist has two independent solutions, for each of the four different solutions to the regional structure inverse kinematics, yielding 8 different inverse kinematic solutions. angles 1 any other choice of twist angles that satisfy the D-H convention will differ only in the signs of these 6