See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/316663146

Kinematic modelling of a robotic arm manipulator using MATLAB

Article  in  Journal of Engineering and Applied Sciences · April 2017

CITATIONS READS

48 4,108

5 authors, including:

Alejandro Leiva Toquica Luis Fernando Ortiz Martinez

Universidad Antonio Nariño Universidad de La Sabana
1 PUBLICATION   48 CITATIONS    1 PUBLICATION   48 CITATIONS   

SEE PROFILE SEE PROFILE

Ruthber Rodriguez Adrian Fernando Chavarro Chavarro

Corporación Universitaria del Huila Servicio Nacional de Aprendizaje SENA
85 PUBLICATIONS   602 CITATIONS    12 PUBLICATIONS   106 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Elaboración del diagnóstico de la situación actual del desempeño de la gestión estratégica, en los hoteles de la ciudad de Neiva, y la optimización mediante el diseño
de sistemas de gestión de la calidad para ellos View project

Improvement of the adaptive grasp functionality of objects with an active underactuated anthropomorphic prosthesis. View project

All content following this page was uploaded by Ruthber Rodriguez on 04 May 2017.

The user has requested enhancement of the downloaded file.

 VOL. 12, NO. 7, APRIL 2017 ISSN 1819-6608
ARPN Journal of Engineering and Applied Sciences
©2006-2017 Asian Research Publishing Network (ARPN). All rights reserved.

www.arpnjournals.com

KINEMATIC MODELLING OF A ROBOTIC ARM MANIPULATOR

USING MATLAB
Ruthber Rodríguez Serrezuela1, Adrián Fernando Chávarro Chavarro2, Miguel Ángel Tovar Cardozo1,
Alejandro Leiva Toquica1 and Luis Fernando Ortiz Martinez1
1
Faculty of Biomedical, Electronic and Mechatronics Engineering, Universidad Antonio Nariño (UAN), Bogota, Colombia
2
Department of Electronic and Telecommunications, Tecnoparque SENA, Bogota, Colombia
E-Mail: [email protected]

ABSTRACT
In this paper the design and implementation of a kinematic model for a manipulator robot arm type with four
degrees of freedom is developed, model robot performance can be checked mathematically using results from coordinate’s
frames, which set the proposed matrices by Denavit-Hartemberg method to determine the robot joins angle vector. This
procedure describes the direct and inverse kinematics. The goal is to determine the final robot´s position and orientation
according to the joint angles related to a coordinate system, the final effector position, where joint angles are located. The
results were implementedin a MATLAB application that performs fast calculations, it allows the verification of the theory
and at the same time becomes as a tool to simplify the analysis and learning for its friendly interface which displays
virtually the movements of the robotic arm AL5A.

Keywords: inverse and direct kinematics, robot arm.

1. INTRODUCTION 2. MATERIALS AND METHODS

In recent years, virtual reality has been delivering There are different mechanical aspects related to
great contributions in various fields of science and Robotic Arm, to understand the physics of this machine.
technology. Projects related to the design and
implementation of a kinematic model for a robotic arm Kinematics: Kinematics is the part of physics
four or more degrees of freedom have been achieved that is responsible for studying the movement of bodies
successfully, implementing analysis and kinematic model regardless of their cause, which essentially takes as a
of a robotic arm of 5 degrees of freedom giving as an reference trajectory in function of time.
objective modeling of direct and inverse kinematics of a
robotic arm from a theoretical and practical experience in Kinematic model: It presents the algorithm of
robotic systems, automation and control. [1], [2], [3]. Denavit-Hartenberg as a method by which it is possible to
Likewise, the development of software for the kinematic represent the robotic arm using homogeneous
analysis of a robotic arm called Lynx 6, which suggests transformation matrices.
more effective methods to reduce multiple inverse
kinematics solutions described. A visual software package Homogeneous transformation matrix: The
called MSG is also developed to test the characteristics of homogeneous transformation matrix established as a 4x4
arm movement [4], [5], [6]. In the same way, kinematic matrix allows to know the location, position and
analysis for a robot arm based on a prototype with three orientation of an axis system of coordinates related
degrees of freedom is presented. It uses an application that to the fixed coordinates .
allows the program run on the card, receive data and
operate allowing the clamp to be moved to a desired Direct Kinematics: For using the direct
position [7], [8], [9]. kinematics the vector and matrix algebra is used based on
the spatial location of an object in three dimensional space
fixed reference. Because the robot is regarded as a
kinematic chain consisting of links joined together by
means of joints, can create a reference system on the base
that describes the location of the links in reference to the
system described.

Inverse Kinematics: The inverse kinematics

seeks the values that should adopt the robot joint
coordinates 𝒒 = [𝒒 , 𝒒 , … . . 𝒒𝒏 ] to position the robot end
effector so that its end is oriented in a particular spatial
location. To obtain the equations is required a procedure
which is entirely dependent on the configuration of the
robot. Thus, as you should obtain the values of the joint
Figure-1. Prototype robotic arm manipulator with four
degrees of freedom.

2037
 VOL. 12, NO. 7, APRIL 2017 ISSN 1819-6608
ARPN Journal of Engineering and Applied Sciences
©2006-2017 Asian Research Publishing Network (ARPN). All rights reserved.

www.arpnjournals.com

variables so that the end effector has a certain position and

orientation.

Multiple solutions should be considered, as

minimize movements made by the arm to the current
position, the concept of allows the nearest solution, to
move the links of lower weight, and obstacles are
considered important to avoid collisions.
Some generic procedures that can be programmed
have been developed so that a computer can, from
knowledge of the kinematics of the robot (with its
parameters Denavit - Hartenberg) to obtain the n-tuple of
joint values that position and orient its end. The
disadvantage of these processes is that it is iterative Figure-2. Robot arm achievable space [11].
numerical methods, whose convergence speed and even
S_ convergence is not always guaranteed. When solving
the inverse kinematic problem is much more suitable to
find the solution closed. This is to find a mathematical
relationship explicit form. = , , =
(D.O.F.) [10].
To this solution, the kinematic problem reverse is
solved in real time, so has to follow a certain path. To a
solution in iterative time there is no guarantee that the
solution develops properly. Otherwise what will happen
with direct kinematic problem, quite often inverse
kinematics is not unique but there are different joints that
guide and position the end of the robot.

Position representation: The representation of

the position is determined by various systems that allow
locate a point in space. Thus systems Cartesian, cylindrical
and spherical coordinates are part of this important Figure-3. Representation of ABC system with
method. This article will use the Cartesian coordinate P point in space.
system.
Representation orientation: Point for a spatial
The reference system , , allows the location is described by a three-dimensional coordinates.
representation of a point in space from the position For a solid body it is necessary to know the orientation of
vector𝑷, so that the equation describes a column vector of it related to a reference system.
three elements;
It is important to unite the solid body coordinate
= , , (1) system , , that be described in space. Therefore, the
orientation will take a direction of the unit vectors of the
Likewise, the location of point P can be coordinate system attached to the body associated with the
represented by another coordinate system , , . The new axes of the source coordinate , , .
coordinate system, depends on the source location relative
to system taken as the main reference , , and Rotation matrix: This method solves the
orientation representing the coordinate axis. This figure problem of orientation, because it allows to use matrix
shows the algebra tools to establish relative rotation between main
Cartesian system A, B, C whose origin is at the axes of two systems sharing the source.
point system , , . The rotation matrix is defined as a
matrix which is used as an operator. The equation
enables a transformation vector on a position in three
dimensional space. Thus, the coordinates become
expressed in a system of coordinate axes rotated , , .

= , , (2)

The orientation is defined by rotating angles to

the main axis of fixed reference.

2038
 VOL. 12, NO. 7, APRIL 2017 ISSN 1819-6608
ARPN Journal of Engineering and Applied Sciences
©2006-2017 Asian Research Publishing Network (ARPN). All rights reserved.

www.arpnjournals.com

Denavit-Hartenberg algorithm: 𝑖 axis

= = = (3) coincides with the hinge axis towards the displacement for
a prismatic joint and likewise the direction of the
rotational axis for the joint. For the Xi axis perpendicular
Rotation matrix operations: Product: The location takes a common reference Zi and Zi + 1.
coordinate system has different eventualities in a solid
body due to experience more than one rotation about Determining the Denavit-Hartenberg
different axes of the source system. To obtain such Parameters: The Denavit-Hartenberg parameters define
rotations it is necessary multiply the rotation matrices the dimensional relationships between consecutive links
representing individual rotations, achieving a composite and joint variables.
matrix rotation.
The multiplication of matrices is not Ɵ𝑖 is the variable angle joint for a joint rotation.
commutative, so these must be multiplied in the same 𝑖 is the distance of the variable joint for a prismatic
order in which they spend the rotations of the system in joint.
space. 𝛼𝑖− is the distance from the axis of joint 𝑖 − to 𝑖
joint axis measured along the perpendicular line
Location in three-dimensional space: The common to these axes. When the axes 𝛼𝑖− are cut
solids in three dimensional space can have movements of be zero.
rotation and translation. These allow the coordinate system 𝛼𝑖− is the angle between the axis of joint 𝑖 − and 𝑖
match a solid body translation vector to determine its joint axis; measured as a rotation axis 𝑖 +
position and a rotation matrix indicating the orientation around the common perpendicular until it
linked to a fixed reference system. The relationship of this coincides with the direction of the axis i.
concept is defined in a 4x4 matrix array called 𝑖 indicates the time of articulation. It is assigned
homogeneous transformation (H.T.M.). zero value for rotational joints and one for the
prismatic joints [13], [14], [15].
Homogeneous transformation matrices:
H.T.M.4x4 represents the position and orientation of a Meaning Denavit-Hartenberg representation:
moving three-dimensional system related to a fixed three- D-H parameters provide information about the number and
dimensional system. type of joints of a robot. It is thus that disclosed
dimensions of the links, the link between consecutive
A H.T.M. consists of four subarrays of different sizes: angular joints and the separation between them. Also,
R3x3 submatrix corresponding to a rotation matrix. determine the joint variables of the Robot.
P3x1 submatrix corresponding to a translation vector.
F1x3 represents submatrix perspective transformation. Transformation matrix D-H: Being determined
W1x1 submatrix corresponds to a global scale. the representation D-H it results in the successive
transformations corresponding to the 𝑖 ℎ system
Operation with homogeneous transformation coordinate with the coordinate system 𝑖 − ℎ. It
matrices: Product Multiplying several MTH have a performs basic operations MTH obtaining a composite
sequence of positions, which results in a trajectory 𝑖 − transformation matrix called D-H. The matrix
generation step. This result is called homogeneous presented below represents the function of each of the D-H
transformation matrix composite, which depends on the parameters in obtaining the matrix.
order in which the products are made.
𝑖− 𝑖
= , 𝑧,∅ ,𝛼 ,𝛼 (4)
Inverse transformation: The homogeneous
transformation matrix consists of four subarrays, of which cos∅i −cosαisen∅i senαisen∅i αicos∅i
both the perspective scaling should be constant, so cannot sen∅i cosαicos∅i −senαicos∅i αisen∅i|
apply the transpose. =|
senαi cosαi di

Denavit-Hartenberg representation: The

Denavit-Hartenberg representation is a method that applies Interpretation matrix transformation D-H:
the properties of MTH to represent the relations of The matrix 𝑖− indicates that A represents a MTH
translation and rotation between adjacent elements of a corresponding to the coordinate axis system (i) linked to
Robot. In 1955 Denavit and Hartenberg proposed an the joint 𝑖 related to the coordinate axis system (𝑖 − )
algorithm to establish systematically a coordinate system binding articulation (𝑖 − ) .The matrix obtained is
if linked to each link i of an articulated chain, this allows determined by a degree of freedom.
moving from one link to the next by 4 basic
transformations that rely exclusively on the geometrical Arm matrix: For a robot with n degrees of
characteristics of the link [11], [12]. freedom, the equation below shows the possible matrices
D n-H multiplying sequentially for . The end effector

2039

View publication stats