The Design of humanoid Robot Arm based on
Morphological and Neurological
Analysis of Human Arm
Yongseon Moon1, Nak Yong Ko2 and Youngchul Bae3
1 School of Information Communication Sunchon National Universit,
2 Department of Control and Instrumentation Engineering, Chosun University,
3 Division of Electrical·Electronic Communication·Computer Engineering, Chonnam
Yeosu,550-745, Korea, email@example.com
Abstract – In this paper, we analyzed and verified possibility and validity of overcoming
present limitations of humanoid robot by using morphological and neurological analysis
of human arm. Through design, implementation and performance evaluation of
humanoid robot arm, we will be verifying applicability and effectiveness of humanoid
robot arm system based on SERCOS network that fulfills the concept of opening,
networking and modularizations that are progressive direction of future robot.
Index Terms – humanoid robot, UML, ISO 15745, SERCOS Communication
At first appearance of the industrial robots during the 1960s, the concepts for the usage of
the robots were only as manipulator in which to perform pre-ordered commands. After 20
years later during the 1980s, together with the appearance of microprocessor, it marked the
beginning of the intense research of the field of robot. As the research progressed, robots
were recognized not only as simple action performer but as a machine that have diverse and
variety of purposes and usages.
As the technology and recognition of robots improved in a variety of ways, humans began
to relate themselves with robots; hence, these robots are called humanoid robot in which
they resemble appearance of human and imitate their behaviour. There are two
representative humanoid robots that have currently developed. They are Japanese Honda’s
ASIMO which is well known as superior to any other humanoid robots and KAIST’s HUBO.
However, these robots have some limiting factors. First, they have limited control
performance due to DC motor usage for joint actuator and difficulty in constant repairing
and replacement of joint that are caused by wearing down of brush and commutator. These
factors cause limited life time of robots, which in turn brings economic burden. Second,
232 Humanoid Robots
current robot control networks are RS-232, USB(Universal serial bus), CAN(Control
area network), Ethernet which cannot sufficiently implemented the performance of
humanoid robots. Third, current humanoid robot technology is focused on production and
implementation of skills without proper mechanism for opening methodology of
development for analysis, design, implementation, and integration for robot development.
In this paper, we analyzed and verified possibility and validity of overcoming present
limitations of humanoid robot by using morphological and neurological analysis of human
arm. Through design, implementation and performance evaluation of human robot arm, we
will be verifying applicability and effectiveness of humanoid robot arm system based on
SERCOS network that fulfills the concept of opening, networking and modularization that
are progressive direction of future robot.
2. The Development Methodology of Humanoid Robot Arm
A. Open human robot development methodology
In order to establish a generalized opening technology system for the analysis, design,
integration, and implementation of the humanoid robot, we proposed a new mechanism of
development using generalized method of robot arm that is core make up component of the
arm. Fig. 1 show open development methodology for humanoid robot arm based on
Opened humanoid robot development methodology begins from analysing morphological
and neurological structure of human arm and through yielding of current developed
humanoid robot arm model. Then, we develop a design integration model of humanoid
robot arm using opening ISO15745 standardization criterion and definition. In ISO15745
standard, regulation is made by using UML (Unified modelling language) which is standard
model of orientation in designing the humanoid robot arm. Humanoid robot arm designing
integration class that has been finally deducted by using UML through another
standardized transformation process for profile description is described as a
XML(Extensible Markup Language) schema. Final design profile is prepared through
instance of document of schema and process. These humanoid robot arm design profile that
has been going through these processes is used as standard manual for robot arm
Fig. 1. ISO 15745 based open development methodology for humanoid robot arm
The Design of humanoid Robot Arm based on Morphological and Neurological Analysis 233
of Human Arm
B. Morphological analysis of human arm
In this paper, we applied mapping concept of human morphological structure for
implementation of robot arm that is similar to structure and movement to that of human
arm. Fig.2 show human -robot morphological mapping.
Actuator + Joint
Fig. 2. Human-robot morphological mapping
C. Neurological analysis of structure of human arm
Through neurological analysis of structure of human , we define control network and
processor structures for implementation of humanoid robot arm. Humans perform
movement of muscles and acquire internal and external information through each organ
based on nerve activity. These nerves can be viewed as robot's network in functional sense.
Furthermore, according to human neurological classification, nervous system is classified in
brain, spinal cord, sensorimotor. These can be matched with functional module of processor
of robot control. Fig.3 shows a Human-robot neurological mapping.
3. Modeling of Humanoid Robot Arm
Based on requirement of robotic arm through neurological and morphological analysis, we
can develop designing model for humanoid robot arm.
Table 1 present basic requirement for implementation of humanoid robot arms currently
developed national and international wide through analysis of structure of humanoid robot
arm and human morphology.
234 Humanoid Robots
- Simplification of arm structure and wire problem
- Optimal behavior implementation based on human
Actuator AC servo motor - High precision control and everlasting use of joint
- High speed data processing
-Advance performance of motion control and multi
- Human nerve system, cable wire remove, and
Table 1. The requirements for implementation of humanoid robot arms
A. Humanoid robot arm resource integration model
ISO15745 standard represents an integration model called AIF(Application Integration
Model) which defines the opening development system. AIF integration model is
subdivided into a process integration model, a data exchange model, and a resource
integration model. Among these detailed models, we standardized them as resource
integration mode to define external structure of system l.
Through this paper, we defined external structure and configuration of humanoid robot
arm using resource integration model of ISO15745. (Fig. 4)
Fig. 3. Human-robot neurological mapping
The Design of humanoid Robot Arm based on Morphological and Neurological Analysis 235
of Human Arm
Fig. 4. The resource integration model for humanoid robot arm
B. Humanoid robot arm data exchange model
By using humanoid robot arm that behaves in ISO15745 standard that are based on
human neurological and morphological transaction structure, we presented behaviour of
internal system for movement of humanoid robot in the Fig. 5. 
Fig. 5. The activity model for humanoid robot arms motion
236 Humanoid Robots
C. Humanoid robot arm integration model
Through analysis of human arm movement and control component, the final designed
model of humanoid arm that are based on human structure like Fig. 5, can be created.
Fig 6 shows the integration class model of the humanoid robot arm structure that will be
Fig. 6. Integration class model for humanoid robot arms
D. Humanoid robot arm design profile
In order to write up XML standard profile for humanoid robot arm, we must define
DTD(Data type definition), schema that defines basic structure of XML documents. Fig. 7
represents standard profile of XML of humanoid robot arm that has been object instance
based on XML schema structure.
The Design of humanoid Robot Arm based on Morphological and Neurological Analysis 237
of Human Arm
Fig. 7. XML profile for humanoid robot arm
4. Sercos Based Robot Arm Design Analysis
A. Humanoid robot arm structure and system
The degree of freedom of humanoid robot arm joints consist of total 5 degree of
freedoms excluding 2 degrees of freedom from total 7 DOF of humans which are
radial/ulnar and flection/extension. Moreover, humanoid robot arm control network and
control system will consist of robotic arm control system that uses motion control SERCOS
communication and high precision AC servo motors which can overcome problems that
current humanoid robot system have.
Fig 8 represents concept of degrees of freedom and structure of humanoid robot arm that
will be implemented to humanoid robot.
In conceptual diagram of Fig 8, among total seven degrees of freedom that humans have,
except two degrees of freedom of wrist that have very little influence in motion control, the
238 Humanoid Robots
five degrees of freedom should be implemented including shoulder 3 degrees of freedom,
elbow 1 degrees of freedom, wrist 1 degrees of freedom. Control motions that are related
with the movement may deal the rest of 2 degrees of freedom.
Fig. 8. The control system configuration of the humanoid robot arms
Unlike usual robot control method in microprocessor environment, SERCOS network
based on humanoid robot arm control system in Fig. 8 uses SoftPLC(Software
Programmable Logical Controller) technology which is the control method of constructing
false image of continuous control environment in PC and uses mater/slave control method
that is the method of continuous control through SERCOS communication.
The transfer of control data through SERCOS communication supports high speed
transmission that speeds up to 16Mbps, and it is possible to support precise motion of
humanoid robot arm through standardized motion function block and to realize precision
motion and synchronized control of separated humanoid robot joints.
Current International Motion Association provides standardized structure and function
blocks for PTP motion control. In this paper, we tried to control the motion of humanoid
robot arm by applying standardized PTP (point to point) control system and interpolation
Fig. 9 represents basic movement structure of state chart model that defines movement of
standardized PTP motion control method.
The Design of humanoid Robot Arm based on Morphological and Neurological Analysis 239
of Human Arm
Fig. 9. The configuration drawing for the standard control function block and the condition
Interpolation control is the method for path driving of mutual interpolation of variety of
axes in space that has been proposed to overcome disadvantage of PTP control method.
Interpolation control, instead of control over discontinuous section, was optimal for
Moreover, it is possible to make continuous section during discontinuous section using
interpolation control like in Fig. 10.
In order for interpolation control to work properly for humanoid robot arm in various
environments, detailed processes of motion interrupter establishment and control should be
Fig. 10. Motion trace from interpolation control and PTP control
240 Humanoid Robots
5. Humanoid Robot Arm Implementation and Evaluation
The humanoid robot arm in the right side of Fig. 11 was implemented with international
standard motion control SERCOS network and AC servo control system. Each arm links
machinery module structure that is possible to attached and remove.
The measurement and calculation for kinematics of matrix of humanoid robot arm motion
tracing has been implemented through visual C++. Fig. 11 also shows the experimental
diagram for humanoid robot arm motion control and calculation for tracing of rotational
angel with input condition by using kinematics model.
Fig. 11. The structure of calculating the information for the join rotation of humanoid robot
By using 3 dimensional space coordinating calculation of robot for input of kinematics
matrix for rotational angle of 5 joints of humanoid robot arm measured during motion trace
used as the method of measuring motion trace of humanoid robot arm, we used method of
yielding real motion trace value of humanoid robot arm.
In Fig 11, we performed performance evaluation for circle motion control of humanoid
robot arm using control performance, motion trace measurement, and calculation
structure. In this paper, we used double circle trace for the purpose of humanoid robot arm
motion control. Fig. 12 represents structure of double circle trace to be inputted for circle
motion control of humanoid robot arm.
In Fig. 13, rotational angle information that will be used as humanoid robot arm was
produced by input of position value of circle trace of desired formula for kinematics matrix.
The Design of humanoid Robot Arm based on Morphological and Neurological Analysis 241
of Human Arm
Since induced input joint rotational angle is dependent on kinematics matrix, in case of
induction of real kinematics matrix does not occur properly, we must use properly induced
kinematics matrix. Otherwise, it will cause totally difference output of desired motion due
to production of false input information.
Fig. 12. The outline of the circle motion control trace setting for the humanoid robot arms
Fig. 13. Circle control input degree trace
As we know input angle trace information from Fig. 14, we know it can be formed by
driving two joints that are circular traces. Furthermore, we created one circular trace
through the rotational angel trace of cosine type to determine direction of motion trace and
the rotational angel trace of sine type to determine position of motion trace.
Fig. 14 shows repeated motion trace and normal circular trace during the 5 repeating motion
trace as a PTP output trace for double circular motion with input rotation angel.
As we know through the circular control trace of PTP from Fig. 14, the repeated position
precision of trace is superior to position repeating precision trace itself. However, precision
242 Humanoid Robots
of position tracking for actual desired circular motion is fairly poor in the PTP control case
of humanoid robot arm.
Fig. 14. Result of the circle PTP control trace from plane (XZ, XY)
In this paper, in order to solve PTP control problem of low precision, we applied
interpolation control which is an optimal control for continuous section in the 3D space. Fig.
15 shows repeated motion trace and normal circular trace during the 5 repeats as an
interpolation control input trace for double circular motion with input rotational angel.
Fig. 15. Result of the circle interpolation control trace from plane (XZ, XY)
We can see the interpolation control which has smoother trace and high precision that is
superior to PTP control. There are a small error found between the actual trace and the
interpolation control trace. The comparison of PTP and interpolation control represented in
Fig. 14 and Fig. 15.
Fig. 16 shows measurement of trace for error between the interpolation control and actual
circular control. Since the errors that occur in each section of real control sampling are
different, we used mean error for each section and evaluated tracking error that occur in
peak location dot tracking for error section and repeating error that represents floating ratio
that occurs between sections. Calculated error value from measurement is represented in
The Design of humanoid Robot Arm based on Morphological and Neurological Analysis 243
of Human Arm
Fig. 16. Circle control error output (motion 5 times repetition error)
Table 2 represent the motion error measurement data of the humanoid robot arm. Through
the 5 controls, we found that mean position trace error is about 4.112mm and mean position
repetition error is about 0.135mm in the Table 2.
Mean trace error : 4.112mm
Mean poison repetition error : (0.1+0.2+0.12+0.12)/4 =
Table 2. The motion error measurement data of the humanoid robot arms
Times Sampling time(ms)Mean error(mm)
1 204mm 204/50 = 4.08mm
2 209mm 209/50 = 4.18mm
3 199mm 199/50 = 3.98mm
4 205mm 205/50 = 4.1mm
5 211mm 211/50 = 4.22mm
In this paper, we have presented the implementation and performance evaluation for
SERCOS based humanoid robot arm by using morphological and neurological analysis of
human arm. Moreover, we reviewed the possibility of application of these robot arms. First,
we proposed robot development methodology of open architecture based on ISO15745 for
“opening of humanoid robot.” Then, we verified the method of implementation of
humanoid robot arm and its application to the real world.
We have implemented robot arm using SERCOS communication and AC servo motor for
high precision motion control; in addition, we got a mean position trace error of 4.112mm
and mean position repetition error of 0.135mm as a control performance.
244 Humanoid Robots
Karl Williams, "Build Your Own humanoid Robots", Tab Bookks, 2004.
Christopher E. Strangio, "The RS232 Standard - A Tutorial with Signal Names and
Definitions" 1993~2006. "Universal Serial Bus Specification", Compaq, Intel,
Microsof, NEC, 998.
Robot Bosch Gmbh, "CAN Specification v2.0", Bosch, 1991.
John F. Shoch, "An Introduction to the Ethernet Specification", ACM SIGCOMM Computer
Communication Review volume 11, pp 17-19, New York, USA, 1981.
ISO TC 184/SC 5, "ISO 156745 - Industrial automation system and integration Part1, 1999.
David G. Amaral, “Anatomical organization of the central M. King, B. Zhu, and S. Tang,
“Optimal path planning,” Mobile Robots, vol. 8, no. 2, pp. 520-531, March 2001.
Rainer Bischoff, Volker Graefe, "HREMES-a Versatile Personal Robotic Assistant", IEEE-
Special Issue on Huamn Interactive Robots for Psychological Enrichment,
pp. 1759-1779, Bundeswehr University Munich, Germany.
In A. Zelinsky, "Design Concept and Realization of the humanoid Service Robot HERMES", Field
and Service Robotics, London, 1998
Rainer Bischoff, "HERMES-A humanoid Mobile Manipulator for Service Task", International
Conference on field and Service Robots, Canberra, December 1997.
Rainer Bischoff, "Advances in the Development of the humanoid Service Robot HERMES", Second
International Conference on field and Service robotics, 1999.
H. Netter MD, "Atlas of Human Anatomy, Professional edition", W.B Saunders, 2006. Function
blocks for motion control, "PLCopen-Technical Committee2", 2002.