A Novel Design of Delta Robot: Abstract
A Novel Design of Delta Robot: Abstract
A Novel Design of Delta Robot: Abstract
365
International Journal of Multidisciplinary Research and Modern Education (IJMRME)
ISSN (Online): 2454 - 6119
(www.rdmodernresearch.com) Volume II, Issue II, 2016
model and the dynamic model of the delta robot using the conventional approach for
solving the dynamics of the parallel manipulators. Alain Codourey[3] has framed the
velocity and acceleration kinematic model for the delta robot. M Lopez, E Castillo, G
Garcia, and A Bashir[4] have formulated the inverse, direct, and intermediate Jacobians
for the Delta robot.
A.Codourey, E. Burdet [5] have framed the body-oriented method for finding a
linear form of the dynamic equation of fully parallel robots, which helps in arriving at a
closed form solution of the dynamics of the delta robot and narrowing down to the
required motor torque. Pradya Prempraneerach [6] has laid out the concept behind the
work space generation of the delta robot. Excel Ultralite Carbon Fiber Tubes Datasheet
[8] provides the material properties about the carbon fiber tubes used for the robot
arms. Fag Data book[9] was referred for the selection of the various mechanical
components in the system which involves the angular contact ball bearings, lock nuts
and lock washer. Misumi Data book [10] was referred for the selection of the spring that
holds the forearms together.
In order to compare and benchmark the delta robot being developed a number of
delta robots available in the market were studied. The manuals or datasheet of the
robots compared are listed below Fanuc M-1iA Series_171 [11] provides the
specifications and information about the kinematic parameters of the Fanuc M-1iA
series of Delta robots. Asyril_ROBOT-POC-01_Datasheet_E[12] provides the
specification of the Pocket Delta Robot, a miniature delta robot made by Asyril. It
provides wide range of information about how a miniature delta robot can be designed.
DeltaRobot-R6Y3+InstallManual[13] provides assembly guide for the Omron Delta
robot and also the lengths of arms of the robot which was very helpful in benchmarking
the delta robot as the ratio plays an important role.
2. Delta Robot:
Delta robot is a parallel 3 DOF robot designed by Professor Raymond Clavelfor
realizing very high speed pick and place operations. The basic advantages of the delta
robot conform to that of the parallel robot. The basic structure of the delta robot is
explained with the help of Figure 1.
366
International Journal of Multidisciplinary Research and Modern Education (IJMRME)
ISSN (Online): 2454 - 6119
(www.rdmodernresearch.com) Volume II, Issue II, 2016
orientation of the lower platform to be parallel to the working surface (table, conveyor
belt and so on). The motors (3) set the position of the arms (4) and, thereby, the
XYZposition of the end effectors, while the fourth motor (11) is used for rotation of the
end effectors.
3. Kinematic Modeling of Delta Robot:
Robot kinematics involves analyzing the movement of structures with multiple
degrees of freedom. Robot structures usually consist of rigid bodies called links and
joints which are usually revolute or translational. This basically involves the
relationship between the kinematic chains and the position, velocity, and acceleration of
the links. Kinematics involves only the study of motion of the robot system without
considering the forces that cause it. Robot Kinematics can be divided into two parts, the
forward kinematics part and the inverse kinematics part. Robot kinematics is used to
generate the work volume of the robot.
3.1 Forward Kinematics:
Forward kinematics involves the computational method to identify the
configuration of the robot in Cartesian space while specifying the joint space
parameters. Robot parameters can be divided into two types, one being the Joint space
parameters and the other, the Cartesian space parameters .The joint space parameters
consists of the angle of rotation or translational movement of the joints and the
Cartesian space parameters involve the positions of the links and joints in the three
dimensional space rendered by a fixed reference frame. Forward kinematics gives you
the end effectors position for a given set of joint parameters.
3.2 Inverse Kinematics:
Inverse kinematics involves the computational method to compute the joint
space parameters when specifying the Cartesian space parameters. Here, the
endeffectors position is known and the joint parameters are the unknown. The inverse
kinematics procedure will result in the values for joint parameters, which is usually the
angle of rotation or the translational movement required to achieve the specified
endeffectors position.
3.3 Kinematic Constraints for a Delta Robot:
The physical setup of a delta robot includes several ball joints and revolute
joints. The ball joints are usually limited within their maximum pivot angle range and
there volute joints are not allowed to rotate completely through the entire 360 degrees
range of motion because of the mechanical constraints that are present in the structure
of the robot.
3.3.1 Revolute Joint Constraint:
The revolute joints are limited in their negative direction of rotation and the
angle to which the rotation is limited varies from robot to robot depending on the
physical setup of the robots base. In this design, the revolute joints are restricted in the
negative direction to a maximum of sixty degrees in each arm of the delta robot.
3.3.2 Ball Joint Constraint:
Delta robots are usually equipped with twelve ball joints and each ball joint is
limited within its maximum pivot angle range, this in turn limits the maximum extent of
movement of the robot arms and also the maximum reach of the robot system as a
whole. Each ball joint is restricted to move freely only within its specified maximum
pivot angle range. Thus whenever the forearms are to swing sideways, the maximum
distance the end point of the forearm can reach is restricted which in turn restricts the
work volume of the robot. Here, the maximum pivot angle range is limited to (+/-
60)degrees.
367
International Journal of Multidisciplinary Research and Modern Education (IJMRME)
ISSN (Online): 2454 - 6119
(www.rdmodernresearch.com) Volume II, Issue II, 2016
3.4 Work Volume Analysis:
The work volume of the robot was generated using MATLAB coding by assigning
the kinematic parameters for the robot, on an iterative procedure the desired work
volume was generated and the final kinematic parameters were fixed. The major
kinematic parameters influencing the work volume were the lengths of the forearm and
bicep and the length of the sides of the base and end effectors triangle. The other factors
that affect the work volumes were the ball joint pivot angles and the mechanical
constraint implied on the revolute joints of the motor axis. Here, the work volume was
generated using the inverse kinematic procedure.
The desired work volume is usually described in terms of maximum reach
diameter of the delta robot which was decided to be 600 mm.Optimized Input kinematic
parameters for the desired work volume:
Bicep length = 175 mm
Forearm length = 400 mm
Base triangle side length = 500 mm
End effecter triangle side length = 100 mm
Ball joint pivot angle range = +/- 60 degrees
Motor axis revolute joint restriction = - 60 to +130 degrees
Work volume for the delta robot is shown in Figure2
369
International Journal of Multidisciplinary Research and Modern Education (IJMRME)
ISSN (Online): 2454 - 6119
(www.rdmodernresearch.com) Volume II, Issue II, 2016
−1
S1𝑇 𝑆1𝑇 𝑏1 0 0
J=[𝑆2 ] [ 0
𝑇 𝑇
𝑆2 𝑏2 0 ] (1)
𝑇 𝑇
𝑆3 0 0 𝑆3 𝑏3
Where Si is given by the following equation,
𝑥𝑛 𝑅 𝑙𝐴 𝑐𝑜𝑠(𝛩𝑖 )
(2)
Si =[𝑦𝑛 ] − 𝑅𝑖𝑅𝑧 [[ 0 ] + [ 0 ]]i = 1,2,3.
𝑧𝑛 0 𝑙𝐴 sin(Θi )
𝑥𝑛
Where [𝑦𝑛 ] - is the end effector position,
𝑧𝑛
𝑅
𝑖𝑅𝑧 - is the Rotation matrix for each bicep from the origin,
𝑅 - Displacement between the origin and a side of end effector plate,
𝑙𝐴 – Length of the bicep,
𝛩𝑖 - Angular motion of each bicep
And bi is given by
𝑙𝐴 𝑐𝑜𝑠(𝛩𝑖 )
𝑅 (3)
bi = 𝑖𝑅𝑧 [ 0 ]
𝑙𝐴 sin(Θi )
The Jacobian matrix maps the Cartesian and joint velocities by the equation
Ẋn = J ∗ Θ̇
The acceleration kinematics involves the mapping of the Cartesian acceleration (4)
to the joint acceleration by the relation given below,
Ẍ n = J̇Θ̇ + JΘ̈
Thus from the Cartesian velocity and acceleration, the corresponding joint velocity and
acceleration can be found by using this model. Hence these values can be used for the
motor torque calculation. (5)
Once the values of these parameters are known, the dynamic model can be
framed using the closed loop equation given by the following equation
𝜏 = (𝐼𝑏 + 𝑚𝑛𝑡 + 𝐽𝑇 𝐽)Θ̈ + (𝐽𝑇 𝑚𝑛𝑡 𝐽)̇ Θ̇ − (𝜏𝐺𝑛 + τGb ) (6)
Where
Gives the torque due to gravity acting on
𝜏𝐺𝑏 = 𝑟𝐺𝑏 𝐺𝑏 [cos Θ1 cos Θ2 cos Θ3 ]𝑇
the actuated joint
Torque due to gravity acting on the
𝜏𝐺𝑛 = 𝐽𝑇 𝑚𝑛𝑡 [0 0 −𝑔]𝑇
travelling plate
𝐼𝑏1 0 0
𝐼𝑏 = [ 0 𝐼𝑏2 0 ] Inertia matrix
0 0 𝐼𝑏3
Where 𝐼𝑏𝑖 = 𝐼𝑚 𝑘𝑟2 + 𝐼𝑏𝑟𝑎𝑘𝑒 𝑘𝑟2 + 𝐼𝑏𝑐
𝐼𝑚 – motor inertia Total inertia of the system where
𝑚𝑏 2
𝐼𝑏𝑟𝑎𝑘𝑒 – brake inertia 𝐼𝑏𝑐 = 𝑙 + 𝑙𝐴2 (𝑚𝑐 + 𝑟𝑚𝑓𝑏 )
3 𝐴
Net mass of one of the bicep to forearm
𝑚𝑛𝑡 = 𝑚𝑛 + 𝑚𝑝𝑎𝑦𝑙𝑜𝑎𝑑 + 3(1 − 𝑟)𝑚𝑓𝑏
assembly
Now with this closed form equation for torque, the various torque values can be
generated by iterating the end effectors position and the maximum torque can thus be
calculated.
370
International Journal of Multidisciplinary Research and Modern Education (IJMRME)
ISSN (Online): 2454 - 6119
(www.rdmodernresearch.com) Volume II, Issue II, 2016
5.3.1 Speed Calculation:
The maximum speed required is calculated by taking arbitrary points within the
25-305-25 path that the robot has to travel and the inverse kinematic solutions are
found for the various points. This gives joint position of the three motors for the various
Cartesian positions of the end effectors. The path considered is plotted in fig 6.
YZ Plot
-670
-200 -100 0 100 200
-675
-680
-685
-690 YZ Plot
-695
-700
-705
60 Theta 1
40
20
Theta 1
0
-200 -100 0 100 200
Figure 7: Variation in Θ1
371
International Journal of Multidisciplinary Research and Modern Education (IJMRME)
ISSN (Online): 2454 - 6119
(www.rdmodernresearch.com) Volume II, Issue II, 2016
The theta variation can be best seen in the plots are shown in Figure 7, Figure 8 and
Figure 9.
Theta 2
50
40
30
20
Theta 2
10
0
-200 -100 0 100 200
Fig 8: Variation in Θ2
Theta 3
50
40
30
20 Theta 3
10
0
-200 -100 0 100 200
Figure 9: Variation in Θ3
The maximum variation in the angular position for the shortest period of time
gives the maximum speed that the motor has is shown in Table 2.
delta t1 delta t2 delta t3
0.257569 0.227843 0.227843
0.258041 0.224 0.224
0.272523 0.0604 0.0604
0.088069 0.098149 0.098149
0.368966 0.216248 0.216248
0.131133 0.0482 0.0482
0.36152 0.18076 0.18076
0.11944 0.30184 0.30184
0.1106 0.072867 0.072867
0.40571 0.171917 0.171917
0.150354 0.022766 0.022766
0.098123 0.311169 0.311169
0.462367 0.546531 0.546531
0.217098 0.248627 0.248627
max values
t1 0.462367
t2 0.546531
t3 0.546531
max rpm 111.5833
372
International Journal of Multidisciplinary Research and Modern Education (IJMRME)
ISSN (Online): 2454 - 6119
(www.rdmodernresearch.com) Volume II, Issue II, 2016
The maximum rpm thus calculated is 111.5 rpm. The speed can be used for the
selection and sizing of the servo motor. Thus the dynamic modeling helped in solving
for the maximum speed and torque that the motor needs to provide and the appropriate
motor was selected.
5. Mechanical Design of Robot:
The dynamic model is useful in solving for the various dynamic parameters of
the system. These parameters are then used for the various strength analyses in the
mechanical design of the delta robot. The mechanical design involves the design of the
various subassemblies and mechanisms within the system such as the base plate, the
bicep assembly, the forearm assembly and the traveling plate.
The basic structure of the delta robot consists of the following:
Base plate - holds the motor and gear box and acts as a support structure
Bicep Assembly – transfers the motion from the motor to bicep through ball
joints.
Forearm Assembly – holds the traveling plate and positions it at the desired
position.
Traveling plate – holds the gripper.
Figure 10 and 11 show the base plate used in the delta robot and the motor
mounting plate. The base plate should be a perfect equilateral triangle of side 500mm.
The plate holding the motor should be perfectly vertical so that it doesn’t affect the
kinematics of the system and hence the positioning accuracy. The provisions given in
the vertical plate for mounting the bicep assembly and the flange is shown in the above
image.
373
International Journal of Multidisciplinary Research and Modern Education (IJMRME)
ISSN (Online): 2454 - 6119
(www.rdmodernresearch.com) Volume II, Issue II, 2016
The calculated catalogue load for the selected bearing is well below the actual
catalogue load of the bearing, which is 6000N. Therefore the selected bearing can
operate without any problem for 30000 hours under standard operating conditions.
The dimension of the bicep tube is calculated after performing various stress analysis
calculations. The loads and torques acting on the bicep are due to the motor torque and
the inertial force of the moving mass at very high acceleration. The maximum
deformation and the maximum stress developed at the worst case condition is well
below the allowable deformation and ultimate tensile strength. Thus the carbon fibre
tube of OD 30mm and ID 27mm was selected.
The forearm assembly consists of the carbon fiber tube epoxies with the
aluminum heads on either side of the tube. The aluminum heads have provision for the
socket. The socket is made up of Torlon material which has very low coefficient of
friction and is self-lubricating. These two characteristics make Torlon suitable for this
ball joint application. The forearm assembly is given in Fig. 13
374
International Journal of Multidisciplinary Research and Modern Education (IJMRME)
ISSN (Online): 2454 - 6119
(www.rdmodernresearch.com) Volume II, Issue II, 2016
375
International Journal of Multidisciplinary Research and Modern Education (IJMRME)
ISSN (Online): 2454 - 6119
(www.rdmodernresearch.com) Volume II, Issue II, 2016
7. Conclusion and Future Work:
The requirement of this paper was to design a Delta robot capable of carrying
1kg payload, achieving a cycle rate of 120 cycles per minute covering a work volume of
400x300x200 mm3. The kinematics and the dynamics of the delta robot were studied
and the model was developed. The kinematic and dynamic analyses were done to meet
the work volume and cycle time requirements respectively. The conceptual design for
the delta robot was done to meet the necessary requirements. The various potential
failure modes were identified by carrying out Design Failure mode and Effect Analysis
(DFMEA). Once the failure modes were identified, the detailed design was done
ensuring that failure does not occur. The kinematic and dynamic design of the delta
robot has been completed for the desired work volume, speed and accuracy. The
drawings were released for manufacturing. The electrical part of the project involves
the design and programming of the control system. The control system has been
decided to be a soft PLC provided by the industry. The control algorithm has to be
implemented by programming the PLC.
The future work is to realize the robot in its physical form. In order to realize the
robot in its physical form, a lot of material procurement is to be done and then the work
will be to integrate the electrical and mechanical systems together and provide the end
user with an easy-to-use user interface. The mechanical parts are likely to be fabricated
within the industry and the electrical parts are to be optimally chosen and procured.
Most of the mechanical parts will be fabricated within the industry itself except for a
few parts like carbon fiber tubes for the arms which will be procured from other
manufacturers. The electrical part of the robot will include servo motors and drives and
the control system used to run them in an accurate and synchronized manner.
Once the system integration is complete, the robot system has to be to be put to
test runs in real time operating conditions and the system has to be tuned accordingly.
Apart from successfully integrating a working system, one has to provide the end user
with a proper and easy-to-use interface. The interface is usually a compact HMI
providing real time information about the robot or a teach pendant that is hand held by
the user or both.
8. References:
1. R. Clavel, “Delta, a fast robot with parallel geometry” 18th International
Symposiumon Industrial Robots 1988.
2. André Olsson, “Modeling and control of a Delta-3 robot” Department of
AutomaticControl, Lund University, February 2009.
3. A. Codourey, Dynamic modelling and mass matrix evaluation of the delta
parallelrobot for axes decoupling control. In International Conference on
Intelligent Robots andSystems, 1996.
4. M Lopez et al, “Delta robot: inverse, direct, and intermediate Jacobians”,
October2005.
5. A. Codourey, E. Burdet, “A Body-oriented Method for Finding a Linear Form
ofDynamic Equation of Fully Parallel Robots”.
6. Pradya Prempraneerach, “Workspace and Dynamic Trajectory Tracking of
DeltaParallel Robot”, 2014 International Computer Science and Engineering
Conference.
7. Christian Hombach, Felix Böttcher, “Delta robot having special arrangement of
theball joints” US Patent Number - US 20110259138 A1.
8. “Torlon 4301 Datasheet”, Solvay Plastics.
9. “Exel Ultralite TM tubes - technical data sheet”, Exel Composites.
376
International Journal of Multidisciplinary Research and Modern Education (IJMRME)
ISSN (Online): 2454 - 6119
(www.rdmodernresearch.com) Volume II, Issue II, 2016
10. “Fag Databook”, Fag, Germany.
11. “Misumi Databook”, Misumi, Japan.
12. “M-1iA Series_171 Manual”, Fanuc, Japan.Bibliography4
13. “Asyril_ROBOT-POC-01_Datasheet_E”, Asyril, Switzerland
14. “DeltaRobot-R6Y3+InstallManual”, Omron, Japan.
377