INTERNATIONAL DESIGN CONFERENCE - DESIGN 2018

https://doi.org/10.21278/idc.2018.0384

THE IMPLEMENTATION OF AN INDUSTRIAL

ROBOT DESIGN TEMPLATE FOR CUSTOMER
PARTICIPATION DESIGN

J. Li, Y. Nie, X. Zhang, K. Wang, S. Tong and B. Eynard

Abstract
Customer participation design plays an important role in achieving high degree of customization.
Complex product such as robot has complicated structure and strong professionalism, the design is hard
for users who are lack of professional knowledge. So an industrial robot design template is proposed.
Base on composition structure of robot, four elements of the template are analysed. According to
relationships and interactions among the four, the design template is established to guide users with a
simple design method. This study provides references for the design of other complex products.

Keywords: industrial robot, design template, participatory design, product design, design
methods

1. Introduction
Due to the complex work environment and high specialization of industrial robots, the traditional
method of configuration engineering by designers cannot completely meet the high degree of
customization requirements of different users. According to the characteristic of industrial robots, a
user-customized configuration design pattern is put forward to achieve rapid development of industrial
robotics (Li et al., 2015). But customers are not professional designers, hence cannot guarantee the
manufacturability and rational construction of the product. The vibration, deformation and other issues
of industrial robots during operation process are likely to lead to the failure of design. Therefore, it is
necessary to develop a design template for industrial robot to make it easy for customers to participate
in the design activities. The structure of industrial robot is relatively fixed which offers the possibility
for the establishment of customer participation design template.
Customer participation design is actually a design method for mass customization production. Because it
is very difficult for customers to design a completely new product, therefore more customers configure the
existing components to product. Randall et al. (2005) pointed out that the product customization needs to
meet the five principles, including providing different customized interfaces for different users, providing
customized starting point, providing modify space, providing design prototypes for customer design
reference. The most critical issue is to give customers a product structure template as a basis for the
customer participation design. The design template is a generic and abstract product structure composed
by components. User instantiates the structure via assigning to the components, then products model with
specific features can be obtained. Currently, there are many researches on the design template of complex
products in the field of configuration design. Zhang et al. (2010) set the definition of the product template
first, then built the ordered tree model of product configuration to support individualized product design. On
that basis, the product configuration process was analysed. Rapid product configuration design based on
product design template was studied in many literatures (Xu et al., 2011; Lu et al., 2014; Xu, 2016; Chen and

DESIGN METHODS 1209

 
  
Zhong, 2016). Different methods were used to build a product configuration template in these literatures. For
specific industrial robots product, Faiña et al. (2013) provided a heterogeneous modular structure, including
a slider module, scalable module, rotation module, hinge module as the basis or toolbox for the industrial
robot development. Pacheco et al. (2013) developed a heterogeneous toy robot platform. Customers with
non-professional knowledge can configure different toy robots using different modules based on their own
interest. In the literature (Ming et al., 2014), a configurable template base method was proposed to solve the
problem in multi-dimensional representation of product design knowledge, a software system based on this
method is applied in several enterprises in aerospace and shipbuilding industries. Lian (2013) put forward a
selection method of design templates, which implements rapid design through reusing the structure of design
template based on the layered ordinal relation analysis, and the feasibility of the method was tested by a case
of rapid design of metal forming machine. Based on three dimensional models, Zhang et al. (2012) proposed
a product modular configuration method. Designers created product configuration models that include
configuration template based on product family BOM and 3D information. Driven by customer’s individual
requirements, the configuration result that met the configuration constraints and requirements was obtained
and presented in the form of a BOM structure and a three-dimensional assembly model. Futhermore, a
construction method of configurable templates is elaborated, including product structure template, parameter
calibration template, BOM template and so on. Through parameter-driven template instantiation, the whole-
process product configuration design is realized (Wang and Wang, 2016).
These studies proposed a series of design templates for use in the process of product configuration
design. However, there are two problems with existing research about the design templates, on one hand,
the majority of templates are designed for designers, lack of support for customers who do not master
professional knowledge about industrial robots design. On the other hand, the templates are built from
the perspective of structure composition, such as BOM lists and structure trees, lack of the description
of connecting relations, constraints, interaction and other aspects of the components.
In view of above problems, this paper does the following research. Firstly, the industrial robot
composition structure is introduced. Secondly, the composition of industrial robot design template for
customer participation is analysed, including the elements, the external view and the parameters of the
template. Finally, an industrial robot design template for customer participation design is established.

2. Industrial robot composition structure

Industrial robot is composed of mechanical system, drive system, control system and sensing system.
As shown in Figure 1, the contents of the dark box represent different types of the same layer, and the
contents of light-coloured box indicate different constituent parts of the upper layer.
From the figure, it shows that the mechanical body includes base, pillar, upper arm, forearm and end
effector. Depending on the task, the end effector is different, such as sucker, gripping fingers, welding
gun, spray gun, etc. Drive system includes power system and transmission system. Power systems
including drive, motor, and transmission system contains chain transmission, gear transmission, belt
transmission, turbo transmission and so on. Control system does not belong in the content of
configuration design which is studied in this paper, so it doesn’t be researched here. Sensing system
includes infrared sensor, force sensor, ultrasonic sensor, etc. The sensor can be selected according to the
needs of the robot to make robot have a certain awareness capability.
The existing of some common industrial robots are shown in Figure 2(a). Between the base and the pillar
is waist joint; between the pillar and the upper arm is shoulder joint; between the upper arm and forearm
is elbow joint; between the forearm and the end effector is wrist joint. Point P is the wrist reference point.
Driven by the drive system, the waist joint rotates around the z-axis, and .the shoulder and elbow joint
rotates around the y-axis. Waist joint, shoulder joint, and elbow joint form three positions degree of
freedom (DOF) of industrial robot, and wrist joint have 0 or more posture DOF. The more DOF the
industrial robot has, the more flexible of the industrial robot movement. But it also will lead to the
complexity of control. Therefore for the degree of freedom, more is not better. Usually 3 to 6 degrees of
freedom can meet most requirements. By pose transformation, industrial robots can move in space and
perform different tasks. In the process of moving, the set of point that P can reach constitutes the industrial
robots’ working space. Work space is closely related to the angular range that industrial robots’ joints can
reach. With ABB's IRB120 for example, the published work space is shown in Figure 2(b) below.

1210 DESIGN METHODS

 
  

Figure 1. Industrial robot composition

KUKA spot welding robot KR180-2 KUKA small scale robot KR5 ABB robot IRB120
(a) the examples of common industrial robots

(b) industrial robots’ working space

Figure 2. The examples of some common industrial robots and working space

DESIGN METHODS 1211

 
  
3. Compositions of the industrial robot template for customer participation
design

3.1. The elements of the template

The design template is a product model composed of components under certain constraints, which can
be expressed as follows:
 n m o q

T =  S i , Con , C ,  Ipar
j x  (1)
 i 1 i 1 x 1 v 1
y

n m

 S is the set of components that make up the template, n is the number of components; Con is
i 1
i
i 1
j

o
the relationships that connect the components, m is the number of associations; C is the set of
x 1
x

q
constraints, o is the number of constraints;  Ipar is a collection of template parameter interfaces, q
v 1
y

is the number of parameter interfaces. The parameter interface defines the types of input/output
parameters of the template and the interface type that the template connects with other models. The
specific contents of each component are as follows:
(1) The structure tree view of the template. The design template can be divided into product structure
trees by layers of components, as shown in Figure 3. Sp represents a product model composed of
components. The structure tree shows clearly the composition of the template, and the set of all
n
components that can’t be decomposed at the bottom is S .
i 1
i

(2) The external view of the template. The design template connects components to the external view
of the template, which is similar to the appearance of actual product. Different kinds of products have
different types of external views, which can be quickly understood and used by customers.
(3) The internal constraint set of the template. The components of the design template need to be in
accordance with certain constraints in the instantiation process. For example, when the two gears mesh,
the required transmission ratio determines the choice of gears’ teeth number; when the sensor and
controller connects, not only the interfaces must be the same, communication standards must be
consistent too. Internal constraints of template are the basic constrained which must be met between the
components. They guarantee the feasibility of the final design result.

Figure 3. The decomposition tree of design template

(4) The external parameter interface set of the template. The parameter interface of the template
mainly interacts with two parts: customers and technical models. The customer inputs the parameters
required by the template through the parameter interface, then obtains the calculation result parameters
from the template. The technical model provides parameter calculation and simulation analysis for the
template, generally established with simulation softwares such as Matlab and Pro/e. The template
transfers parameters required by the simulation analysis to the technical model through parameter

1212 DESIGN METHODS

 
  
interface. And the analysis result will be fed back to the template after the accomplishment of technical
model analysis through parameter interface too. Figure 4 shows the interaction process of the parameter
interface.

Figure 4. Interaction process of the parameter interface

The design template supports the customer participation design through the above two views and two
sets. The relationships and interactions between the four are shown in Figure 5. The components in the
external view and the structure tree view are constrained by the internal constraint set of the template.
The customer inputs demand parameters and obtains the final product performance parameters through
the external view, obtains the BOM list of the final product through the structure tree view as well.

Figure 5. The relationships and interactions between the design template elements

3.2. The external view of the template

The external view of the industrial robot needs to provide customers with an easy-to-understand view,
so it should be able to directly reflect the structure of the industrial robot. Zhang et al. (2000) introduced
a representation method with six types of modular joints to represent industrial robots. Although the
method is intuitive, the representation view is too complicated.

DESIGN METHODS 1213

 
  
From a mechanical point of view, industrial robot is an open chain structure composed by rod-shaped
parts which are connected by kinematic pair. In the customer participation design process, customers
only configure the whole structure and the movement pattern of kinematic pair, and not design the
specific structural of components. Therefore, the mechanical schematic which ignores the specific
structure of component is taken to establish the template. The mechanical schematic symbols used by
the template are shown in Tables 1 and 2. In Table 1, the symbols are taken from ISO3952/1-1981,
ISO3952/2-1981, ISO3952/3-1979. And the symbols in Table 2 are needed here but not taken from
ISO.

Table 1. Template symbols from ISO (a)

ISO3952 ISO3952
Components of Components of Components of ISO3952 standard
standard standard
industrial robot industrial robot industrial robot symbol
symbol symbol

Plane rotation Translation

Rotation pair
pair pair

Rack(fixed
Screw pair Shaft
rack)

Belt/chain Gear
Coupler
transmission transmission

prime motor
Electric motor Bearing
(universal)

Table 2. Template symbols (b)

Components of Components of Components of
Symbol Symbol Symbol
industrial robot industrial robot industrial robot

End effector Actuator Sensor

The industrial robot design template is shown in Figure 6. The template contains basic elements of
building an industrial robot. Each joint of the robot contains a prime motor, a reducer and an actuator
corresponding to the prime motor. Actuator is usually installed outside of the robot, and is connected
to the motor by control line. The contents in the dotted box represent undetermined elements, and
they will be determined by customers, then the dotted line will become solid line. For example, the
undetermined transmission way can be belt transmission, gear transmission, chain transmission and
so on. The undetermined DOF is wrist joint DOF. The sensor can be in installed in some part of the
robot as needed. L1~L4 represent the lengths of base, pillar, upper arm, forearm respectively. O0-X0-
Y0-Z0 is the absolute coordinate system. It takes the earth as a fixed reference, and it is regardless of
the position and orientation of the robot. O1-X1-Y1-Z1 is the base coordinate system. It takes the base
surface o of the robot as reference, and the default is generally the same as the absolute coordinate
system. Ot-Xt-Yt-Zt is the tool coordinate system. It takes the end effector as reference, and it provides
a reference for the end effector relative to the robot body. θ1~θ3 represent the angles of the three
position DOF. The base coordinate system is adopted to the three angles, and the positive rotation is
counter-clockwise.

1214 DESIGN METHODS

 
  

Figure 6. The external view of the industrial robot design template

The internal parameter constraints of the industrial robot design template are derived from the
interaction with Simscape through the kinematics and dynamics analysis of the external view. The
specific calculation of the parameters has been given in the article (Li et al., 2017), which will not be
described in detail here.

DESIGN METHODS 1215

 
  
3.3. The external parameter interfaces of the template

3.3.1. Input-output parameter interfaces

According to the content of the dotted box in Figure 6, the parameters of template needed to be
determined by the users are as follows:
, , , , , , , (2)
is the number of DOF;
, ,…, (3)
is the kinematic pair type of the ith DOF, n is the number of DOF;
, ,…, (4)
represents the prime motor type which provides motive for . It is usually a motor except under
exceptional circumstances;
, ,…, (5)
represents the transmission way used by . It is usually a gear reducer except under exceptional
circumstances;
, ,…, (6)
represents the type of the ith sensor, m is the number of sensor used in the template;
is the end effector type, and it is determined by the robot task;
is the maximum load imposed on the end effector;
is the minimum length of the robot arm, and it determines the work space of the robot。
represents the static parameters needed to be determined in Figure 6. In addition to these, some
dynamic parameters need to be input to the template. The dynamic parameters P d are represented as
follows:
, , , (7)
represents the velocity of point P;
represents the accelerated velocity of point P;
represents the velocity of end effector;
represents the accelerated velocity of end effector.
After determining the template parameters, it will become an industrial robot template composed by
components. Through value assignment to the components, the template will become a specific product
composed by component objects.
The parameters that the external view outputs are mainly performance parameters of the final product
designed the customer, including the workspace, the vibration response, the stress-strain scope,
kinematic velocity, etc. obtained through the finite element analysis, and the workspace, kinematic
velocity analysed by the Simscape model, as well as a list of components composed of the final product
obtained from the product structure tree.

3.3.2. Technical model interaction parameter interfaces

The interactions between the external view and the technical model begin from the interaction with the
Simscape model. The external view inputs the length l, weight W, joint angular acceleration ω, etc. of
each rigid body L to the Simscape model and the Simscape model inputs the calculated Ti (The maximum
torque of the ith DOF) value to the external view, so as to provide parameter constraints for external
view. At the same time, the Simscape model can also input the product workspace, kinematic velocity
and other parameters to the external view for the customer to view.
In addition, since the external view needs to output the final product performance parameters to the
customer, it also needs to interact with the product performance simulation model to acquire the

1216 DESIGN METHODS

 
  
simulation results. This interaction with the product performance simulation model is performed after
all the parameters input by the customer have been determined and instantiated into a specific product.
The purpose is to extract the related product parameters after the design accomplishment and then carry
out performance analysis for the design result, thus providing a basis for the improvement of designing,
which is also an analysis of the design template.
Generally, we make the performance analysis of industrial robots with finite element analysis.
Finite Element Analysis (FEA) refers to decompose the physical structure into a finite number of
elements interacting on each other, set an approximate mathematical solution for each element, and
based on this, obtain the solution of the whole mechanical structure. FEA can achieve the simulation
of real mechanical structure under different loads, working conditions, constraints and other
conditions.
The industrial robot finite element analysis includes statics analysis and modal analysis. Statics analysis
researches the balance condition (including stationary and uniform linear motion) of the mechanical
structure under stress. The condition generally refers to the mechanical structure stress and strain in the
equilibrium state. Through applying load to mechanical structures, the mechanical structure stress and
strain analysis is taken. When the stress and strain of mechanical structure is too large, an appropriate
interpolation value can be provided in the control system, or improving the mechanical structure to
improve structural strength, thereby improving the motion accuracy of industrial robot. Modal is a basic
concept of structural dynamics. It refers to the natural vibration characteristics of the mechanical
structure (including frequencies and vibration mode). Mechanical structures have different modals, and
each modal has a corresponding frequency and vibration mode. The process of analysing and calculating
the modal of mechanical structure is modal analysis. The natural frequencies and vibration mode
belonging to each modal can be determined through modal analysis, so the actual vibration response of
mechanical structure in the role of a vibration source can be known. The industrial robot finite element
analysis is generally aimed at the main structural components including base, pillar, upper arm and
forearm (Zhan, 2016).
The finite element analysis software commonly used includes ANSYS, Pro/MECHANICA, ABAQUS,
etc. In this paper, the ANSYS is chosen to establish the technical model of industrial robot product, and
the product’s finite element analysis is completed by transferring the product parameters to the technical
model.

4. Industrial robot design template establishment

Through a series of processing to the parameters inputted, the customer participation design template
outputs the components list of industrial robot and performance parameters meeting the needs. Final
establishment of industrial robot template with the flow and analysis process of parameters is shown in
Figure 7.
According to the needs of the task, the customer inputs the required parameters to template, including
the DOF of wrist joint, the kinematic pair type of each DOF, the workload weight, the end effector type,
the minimum arm length, velocity and other parameters. Simscape model obtains parameters such as
load weight, arm length, joint angle from the design template, simultaneously query the weight estimate
table, calculate the parameters of the work space, joint torque and then output them to the template. In
the constraint of the Simscape calculation results and design template internal constraint, the component
objects meeting the constraints can be retrieved from the database to compose the design result. When
all components of the design template have become into component objects, the corresponding three-
dimensional models of objects and the calculated load data will be input to ANSYS. The ANSYS is
used in statics analysis and modal analysis of the design result. The analysis results include stress-strain
data and vibration frequency data. These data is output to template. Finally, the template will send the
design result and ANSYS analysis result to customer.

DESIGN METHODS 1217

 
  

t
t
t

Figure 7. The industrial robot design template

5. Conclusion
To achieve high degree of customization, customer participation design gains more and more attention
and has become the inevitable trend of product development. As a complex mechatronic product, it is
complex for industrial robots to take customer participation design because it is not possible for the
customer to understand the underlying relationships between components composed of the product. To
implement customer participation design, an industrial robot design template is given in this paper.
Contribution of the article mainly includes the following three parts: Firstly, the industrial robot

1218 DESIGN METHODS

 
  
composition structure is constructed, which is the basis for design template implementation. Secondly,
the customer participation design template elements for industrial robot are analysed, including the
structure tree view, the external view, the internal constraint set and the external parameter interface set.
At the same time the relationships and interactions between the four are made clear. Finally, the
industrial robot design template establishment is proposed through the specific flow and analysis process
of parameters. The design template provides non-professional customers with a simple design method
and design process, and ensures the feasibility of the design result. The study has certain reference value
to the implementation of customer participation design of other complex mechatronic products.

Acknowledgement
This work was supported by the National Natural Science Foundation of China (71572147, 71402140), Humanity
and Social Science Foundation of Ministry of Education of China (14YJCZH213, 12YJC630201, 17YJC630059),
the Natural Science Foundation of Shaanxi Province (2015JQ7277, 2015JM7378).

References
Chen, W. and Zhong, L. (2016), “The development of boiler barrel parameterized design system based on design
template”, Software Guide, Vol. 15 No. 9, pp. 88-91. https://doi.org/10.11907/rjdk.161663
Faiña, A., Orjales, F., Bellas, F. and Duro, R. (2013), “First Steps towards a Heterogeneous Modular Robotic
Architecture for for Intelligent Industrial Operation”, 2011 IEEE/RSJ International Conference on Intelligent
Robots and Systems, San Francisco, CA, USA, September 25-30, 2011. https://doi.org/10.13140/2.1.2866.2409
Li, J., Daaboul, J., Tong, S. and Bosch-Mauchand, M. and Eynard, B. (2015), “A design pattern for industrial
robot: User-customized configuration engineering”, Robotics and Computer-Integrated Manufacturing, Vol.
31, pp. 30-39. https://doi.org/10.1016/j.rcim.2014.06.005
Li, J., Nie, Y., Bosch-Mauchand, M., Zhang, X. and Wang, K. (2017), “Configuration engineering of industrial
articulated robot based on object-oriented pattern”, The 3rd International Conference on Control, Automation
and Robotics, Nagoya, Japan, April 22-24, 2017, IEEE, New York, pp. 242-247.
https://doi.org/10.1109/ICCAR.2017.7942696
Lian, G. (2013), “Study on selection method of templates for rapid design based on layered ordinal relation
analysis”, Journal of Nanchang Institute of Technology, Vol. 32 No. 3, pp. 13-16.
Lu, C., Zheng, S. and Sheng, Y. (2014), “A product design approach using design templates to enhance customer
satisfaction”, Modern Manufacturing Engineering, No. 4, pp. 117-120. https://doi.org/10.16731/j.cnki.1671-
3133.2014.04.013
Ming, Z., Yan, Y. and Wang, G. (2014), “Multi-dimensional representation for product design knowledge based
on configurable knowledge template”, Beijing Ligong Daxue Xuebao/transaction of Beijing Institute of
Technology, Vol. 34 No. 9, pp. 917-922. https://doi.org/10.15918/j.tbit1001-0645.2014.09.029
Pacheco, M., Moghadam, M., Magnússon, A., Silverman, B., Lund, H.H. and Christensen, D.J. (2013), “Fable:
Design of a modular robotic playware platform”, 2013 IEEE International Conference on Robotics and
Automation, Karlsruhe, Germany, May 6-10, 2013, IEEE, New York, pp. 544-550.
https://doi.org/10.1109/ICRA.2013.6630627
Randall, T., Terwiesch, C. and Ulrich, K. (2005), “Principles for User Design of Customized Products”, California
Management Review, Vol. 47 No. 4, pp. 68-85. https://doi.org/10.2307/41166317
Wang, P. and Wang, G. (2016), “Research on product configuration design method based on configurable
template”, Mechanical Engineering & Automation, No. 4, pp. 214-216. https://doi.org/10.3969/j.issn.1672-
6413.2016.04.092
Xu, J., Qiu, L. and Zhang, S. (2011), “Progressive product configuration design technology based on knowledge
feedback”, Computer Integrated Manufacturing Systems, Vol. 17 No. 06, pp. 1135-1143.
https://doi.org/10.13196/j.cims.2011.06.17.xuj.004.
Xu, Z. (2016), Research on cross-platform knowledge integration technology for product design, PhD thesis,
Beijing Institute of Technology.
Zhan, G. (2016), Finite element analysis and structure optimization of arc welding robot, Master’s thesis, Harbin
Institute of Technology.
Zhang, M., Li, G., Gong, J. and Wu, B.Z. (2010), “Technology of product configuration design based on ordered-
tree”, Computer Integrated Manufacturing Systems, Vol. 16 No. 11, pp. 2333-2340. https://doi.org/2333-2340.
10.13196/j.cims.2010.11.15.zhangm.002
Zhang, M., Li, G., Gong, J., Wu, B.Z. and Deng, P. (2012), “Product modular configuration based on three
dimensional models”, Computer Integrated Manufacturing Systems, Vol. 18 No. 11, pp. 2370-2380.
https://doi.org/10.13196/j.cims.2012.11.18.zhangm.020

DESIGN METHODS 1219

 
  
Zhang, W., Liu, S. and Li, Q. (2000), “Data/knowledge representation of modular robot and its working
environment”, Robotics and Computer-Integrated Manufacturing, Vol. 16 No. 2-3, pp. 143-159.
https://doi.org/10.1016/S0736-5845(99)00055-1

Yafei Nie, Dr.

Northwestern Polytechnical University, School of management
127 West Youyi Road, Beilin District, Xi'an Shaanxi, 710072, P.R.China
Email: nyf1993@mail.nwpu.edu.cn

1220 DESIGN METHODS