Printer Friendly

Mechanical design of a hip joint for an anthropomorphic leg.

Abstract: The design and development of legged robots has been one of the main topics in advanced robotic research. The research in this field is been oriented to the development of humanoid robots. In this paper it presents the work done on the original construction of a hip joint for an antropomorphic leg.

Key words: humanoid, robot, hip, joint, leg


Very little of the world is accessible by wheels robots. Biped robots can adapt easily to various types of grounds especially those unknown environment with a lot of constraint and obstacles. Biped robots can be used to explore inaccessible or hazardous locations, provide service in the places that are dangerous or not reachable for human beings. For the widely potential use, biped robot has been researched for dozens of years by a lot of research groups. There are still many issues to be studied and there is still a long distance between the humanoid robot and us human beings. Several humanoid robots have been developed in these years.

One of them is WABIAN constructed by Waseda University (Yamaguchi et al., 1999). WABIAN is succeed by WABOT-1 (WAseda roBOT-1), which is the world's first life-sized humanoid robot constructed in 1973 by late Prof. KATO's laboratory. It is no exaggeration to say that he was a pioneer in the development of humanoid robot. WABIN-RII has a completely humanoid figure with two legs, two arms, two hands, and two eyes, and is capable of walking and even dancing.

The most impressive humanoid robot should be HONDA humanoid robots. When P2, the second prototype HONDA humanoid robot, was revealed in 1996 after ten years secret research, the robotics world was stunned. P2 is the world's first cable-less humanoid robot, which can walk and can go up/down stairs (Hirai, 1997). In 2000, further downsizing P3, ASIMO that stands for Advanced Step in Innovative Mobility appeared with children-size (1200 [mm] height, 450 [mm] width, 43 [kg] weight including batteries, 6 D.O.F./Leg, 5 D.O.F./Arm, 1 D.O.F./Hand, 2 D.O.F./Head) and new walking technology (i-WALK). The introduction of i-WALK technology allowed ASIMO to walk continuously while changing directions, and gave the robot even greater stability in response to sudden movements. It is no exaggeration to say that the great success of HONDA humanoid robot makes the current research on the world's humanoid robot to become very active area.

The credit for a success of these humanoid robots nearly goes to the zero moment point (ZMP) theory invented by Prof. Vukobratovic (Vukobratovic & Juricic, 1969). He is also one of the inventors for anthropomorphic mechanisms and biped locomotion.

The more humanoid robots which can walk and can go up/down stairs are developed, the more humanoid robots are expected to perform several application tasks in an actual human living environment. However the application area of humanoid robots has still limited to the amusement and the entertainment.

There are many papers that detailed the design process of particular joints, but no one present the design process itself. This paper, will address the main consideration taken in designing a hip joint for an anthropomorphic leg.


There is various design consideration when designing this hip joint. Among the various factors being considered are (Choong et al., 2003):

* robot size selection

* degrees of freedom (DOF)

* actuator selection

* loads at joints

* sensor selection

* control hardware

2.1 Robot size selection

When we design a robot, we will eventually come to the questions: "What size should it be?" and "How large/tall should we design it?". To answer those questions above, we need to consider various factors that are affected by the size of the robot. The space available and the power requirement to move the robot will place constraints on the size of the robot. Anyway, we want the robot to be anthropomorphic. So, the size of an 8 year-old child was chosen. Its height up to hip is approximately 0.62m, and the weights of the two lower limbs are about 10kg.

2.2 Degrees of freedom (DOF)

Its want that the robot be able to move in 3 dimensions (3D). That means that the robot must be physically capable of changing walking directions, walking up stairs, and others similar tasks. To be able to do these tasks, the robot needs to have sufficient DOFs. Say, if want to only move the legs forwards and backwards (no sideway movements), would not be able to change the walking direction, and would find walking up stairs very difficult.

Therefore, the robot needs to have at least 12 DOF for it to be able to walk in 3D (hip--3 DOF, knee--1 DOF, ankle--2 DOF, total 6 DOF/ leg). Figure 1 shows the location and the arrangements of the joints on the robot.

The ranges of motion for the joints are modelled after the human joints. So, the hip joint has the following limits of motion, which are show in Table 1.


The three bending movements of the body are made possible in the hip region. The hip joint is equipped with a self-lock mechanism preventing it from collapsing in case of a power loss. Figure 2 shows one of the found solutions. The angular flexibility of the three flexion axes of the hip joint is show in Table 1. It can use Cardan joints for to construction the hip joint.

2.3 Actuator selection

For choice the actuator is needed some information. After simulation, it's obtained the data for the torque and speed. These dates are show in Table 1. The actuators preferably should have a high power: weight ratio and also lightweight.

The three most common types of actuators are the hydraulic actuator, pneumatic actuator and electric motors. Hydraulic actuators have good power: weight ratios, but they are definitely not lightweight. Pneumatic actuators are lightweight, but they do not have good power: weight ratios. Further to that, both require a very bulky pump or compressor, which is not feasible for walking robots.

So, each joint is actuated by Maxon's EC motors via distributed control with harmonic drive gear.

2.4 Load at joint

From previous studies of the human hip joints, it was found that the loads on the hip could reach up to 7 times the body weight respectively (Blaha, 1993). The estimated weight of the biped robot is approximately 10kg. For a load factor of 7 on the hip, the load would be 700N. This load needs to be taken into account when sizing the materials for the hip joint. Further to that, a safety factor is incorporated. The most critical part is at the hip which takes the highest amount of impact loading. The deflection of the shaft caused by the loading was also taken into consideration. However, it can reduce the impact load on the joint. This can be achieved by incorporating a shock absorbing material, say rubber, at the joints. This will reduce the impact on the joint during landing (Hirai et al., 1998). This has a positive effect to inhibit vibrations of the hip joint.


2.5 Sensor selection

One of the most important information required is the joint position and velocity. This information can be obtained via a potentiometer or an encoder mounted onto a joint. However, choosing between either is not as straightforward. Each has its own advantages and disadvantages. The signal from a potentiometer tends to be rather noisy. Some filtering might be required before it becomes useful. As for the quadrature encoder, there arises the task of reading the signals (pulses). This can be easily solved by using a quadrature counter. However, the cost of the entire setup (quadrature encoder and counter) could be prohibitive. Same goes for a good potentiometer.

2.6 Control hardware

Recently, PCI bus becomes the most popular bus in industrial field. However, employing PCI bus brings an issue. It is that PCI bus accepts only four or less PCI boards without a bus-bridge. This is an important issue for constructing a humanoid robot. Because, several kinds of function such as DA, AD, counter, and Digital Input / Output, and multichannels are necessary for control the humanoid robot.


The design steps outlined in this paper provides a systematic approach to design a hip joint for an anthropomorphic leg. It describes the considerations and the strategy towards achieving the specifications laid out in the beginning.

What can be done in the future? Well, it will be make this hip joint and design the knee joint and ankle joint for construction entire anthropomorphic leg.


This work was made at Vienna University of Technology, Institute of Handling Devices and Robotics in frame of CEEPUS RO 124-04/05. The authors would like to thanks to Univ. Prof. Dipl.- Ing. Dr. Dr. h. c. mult. Peter Kopacek.


Blaha D. (1993). Principles of joint prostheses. In Verna Wright and Eric Radin, editors, Mechanics of Human Joints: Physiology, Pathophysiology and Treatment, pages 373-392. Marcel Dekker Inc.

Choong E; Chee-Meng C.; Poo A. & Hong G.S. (2003). Mechanical design of an anthropomorphic bipedal robot, Proceedings International Conference on First Humanoid, Nanotechnology, Information Technology, Communication and Control Environment and Management (HNICEM), March 27-30, 2003, Manila, Philippines

Hirai K. (1997). Current and Future Perspective of Honda Humanoid Robot, Proceedings IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 500-508

Hirai K.; Hirose M.; Haikawa Y. & Takenaka T. (1998). The development of honda humanoid robot, Proceedings Of the 1998 IEEE International Conference. on Robotics and Automation, pages 1321-1326, May.

Vukobratovic M. & Juricic D. (1969). Contribution to the Synthesis of Biped Gait, IEEE Tran. On Bio-Medical Engineering, Vol. 16, No. 1, pp. 1-6

Yamaguchi, J.; Soga E.; Inoue S. & Takanishi A. (1999). Development of a Bipedal Humanoid Robot--Control Method of Whole Body Cooperative Dynamic Biped Walking, Proceedings IEEE International Conference on Robotics and Automation, pp. 368-374
Table 1. The limits of motion for hip joint

 Min. Max. Max. Max.
Joint angle angle torque speed

Hip roll -50 50 20 3
Hip pitch -30 100 50 0.3
Hip yaw -180 60 10 0.3
COPYRIGHT 2007 DAAAM International Vienna
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Vatau, Steliana; Cioi, Daniel; Radulescu, Corneliu
Publication:Annals of DAAAM & Proceedings
Geographic Code:1USA
Date:Jan 1, 2007
Previous Article:Experimental report on use of additive technologies in special vehicles development.
Next Article:Use of self-organizing neural network in modern manufacturing systems.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters