Posts Tagged ‘Quadruped Walking Machine’

1984-93 – Nuclear Inspection Centaur Robot – ART Project (Japanese)

The ART Project’s Nuclear Inspection Centaur Robot

After the earthquake last year and the resulting damage to the Fukushima nuclear plant, observers criticized Japan’s lack of preparedness. In particular, many felt that the Japanese robotics sector’s focus on expensive humanoids had squandered time and resources better spent on more specialized robots.  However, this isn’t totally accurate.  The Japanese government, corporations, and universities have been working on robots for just this sort of problem for decades.  Back in the 1980s the Japanese government invested 20 billion JPY (still less than $100 million dollars at the time) into a massive eight-year program to build three types of advanced robots for hazardous environments.

The ART (Advanced Robotics Technology) Project had goals that were too big for any one institution to achieve, so a consortium called ARTRA (Advanced Robotics Technology Research Association) was formed. Financed and controlled by the Agency of Industrial Science and Technology, ARTRA brought two major government organizations, the Mechanical Engineering Laboratory (MEL; now known as AIST) and the Electrotechnical Laboratory (ETL), together with 18 corporations under the same banner, along with the support of academia.

The ART robots were designed for three major areas: nuclear plants, undersea oil rigs, and a third for disaster prevention in refineries.


The nuclear inspection robot would have a sensor head, four legs, two 7-DOF arms, and four-fingered hands with pressure sensitive finger tips (this configurarion led to it being known as the Centaur robot). It would be paired with a smaller, wall-climbing partner that used suction cups to adhere to wall surfaces. This would allow it to “climb up and down stairs, step over piping or other impediments, and relocate itself at high speed.”

It would have to work in 70 degrees (158 degrees Farenheit), 90% humidity, and 100 roentgens of radiation per hour. What started as a 1/3-scale model of the four-legged mechanism eventually became a robot measuring 188 cm (6’2″) tall, 127 cm (4’2″) long, and weighing 700 kg (1,543 lbs).


The undersea robot looked like one of the pods from 2001: A Space Odyssey, with multiple arms and manipulators. It would have to function 600 feet underwater, in tides moving at 2 knots, and in very poor visibility.  Finally, the disaster response robot would put out fires with a hose, move on six legs (each ending with a wheel) and had an arm for closing valves. It would have to work for thirty minutes despite temperatures in the range of 400 degrees (750 degrees Fahrenheit).

Susumu Tachi, director of the robotics department at MEL at the time, was in charge of developing the robots’ teleoperation system. Although the robots were intended to have some degree of autonomy, the majority of their work would be done under the direct control of a human operator. Basically, the human operator would wear 3D glasses or some other head-mounted display, and control certain robot functions using haptic controls and a pair of joysticks. Later Tachi established a lab at Keio University, where many unique “tele-existence” projects would take shape.


Teleoperating the Centaur nuclear plant robot

The first conceptual illustrations were unveiled in 1984, but the actual construction of the robots wouldn’t begin until 1987. There was a great deal of conflict between ARTRA members, not to mention over 40 research themes, so a huge amount of preparation work had to be completed. One researcher called the nuclear inspection robot’s configuration the result of “a lack of imagination”, and Shigeo Hirose (of TITECH’s Hirose-Fukushima Lab) had this to say about it:

    “It’s almost impossible to use a four-legged robot in a nuclear reactor. The passageways are too narrow, and it’s supposed to carry a huge load of equipment. I don’t think it’s practical, and I keep telling the people in charge that they should use a snake design, which no one else has used. We need a more functional design.

    I think the whole thing is a mess. They’re starting out without a long-range, completed vision, but with many corporations and a great deal of politics involved. The result, I fear, will be a very boring robot.”


Part of the reason the nuclear inspection robot has legs is because Hitachi won the contract to work on it, having worked with Waseda University on bipedal robots.  The Waseda-Hitachi Legs [WHL-11] appeared at the Tsukuba Expo ’85. Somewhat prophetically, Hitachi’s Hideo Maki had this to say about the Centaur robot:

    “I think two legs is the ideal, and I would like to develop such a robot, but technologically that’s ten to fifteen years in the future, so we’re settling for four. Professor Kato says that man used to walk on all fours, but that when he stood up and freed two of his limbs, a whole new world opened up for him. Our robot has four legs, and two arms, and we think it will yield the same results.”

Changes in specifications along the way also slowed progress. For example, Hitachi (responsible for building the Centaur robot’s legs) had planned for the robot to carry up to 100kg (220 lbs), but then the companies responsible for the sensors and other equipment kept making requests, forcing redesigns. The researchers studied the walking gaits of horses, elephants, hippos, and rhinos in an effort to better replicate four-legged walking, resulting in a speed of 2.5 kph (1.5 mph).


One of three Hitachi quadrupeds developed circa 1990

In 1993, the robot appeared in the Japanese pavilion at the Taejon Expo in South Korea. By that time the prototype had successfully completed a few trials, but didn’t make the jump into real use. And sadly, both photos and video of the robot experiments are extremely scarce, despite it being such a large project. Hitachi, for its part, says that the lessons learned from this and other robot projects are reflected in their recent robots, such as the EMIEW humanoids.

Many household names can be found in the following breakdown of responsibilities. Some of these corporations were selected because of their prior involvement with similar technologies. Hitachi, Mitsubishi, and Toshiba, for example, had developed robots for nuclear plants independent of the ART project. And Kawasaki, which builds Japan’s submarines, played a significant role in the development of the undersea robot.
1. Basic Technologies
A. Locomotion

    Fuji Electric R&D
    Yaskawa Electric

B. Control


C. Systems Support

    Int’l Robotics & Factory Automation Center

2. Nuclear Plant Robot
A. High Reliability

    Mitsubishi Electric
    Mitsubishi Heavy Industries
    Japan Power Engineering & Inspection

B. Radiation Resistance

    Japan Power Engineering & Inspection

C. Locomotion

    JGC Corp
    Japan Power Engineering & Inspection

D. Manipulation

    Mitsubishi Heavy Industries
    Japan Power Engineering & Inspection

E. Remote Control

    Mitsubishi Electric

3. Undersea Robot
A. Undersea Robot Positioning & Navigation:

    Mitsui Engineering & Shipbuilding
    Kawasaki Heavy Industries
    Sumitomo Electric Industries

B. Undersea Robot Vision

    Oki Electric Industries

C. Undersea Robot Manipulation

    Kawasaki Heavy Industries
    Komatsu Ltd.

D. Undersea Robot Supervisory Control

    Oki Electric Industries
    Mitsui Engineering & Shipbuilding

4. Disaster Prevention Robot
A. Durability

    Ishikawajima-Harima Heavy Industries

B. Sensors

    Matsushita Research Institute
    NEC Corp.

C. Environment Adaptation

    Ishikawajima-Harima Heavy Industries
    Kobe Steel
    Int’l Robotics & Factory Automation Center

Much of the information in this article was assembled from Frederik Schodt’s Inside The Robot Kingdom, which has a more detailed account of the project.

Source: Plasticpals

See other early Teleoperators and Industrial Robots here.

See other early Walking Machines here.

Tags: , , , , ,

1988 – “Rikky and Pete” Mechanical Horse – David Parker et al (Australian)

Plot Summary for Rikky and Pete (1988) 
Follow the lives of Rikky, a talanted geologist, and her brother Pete, an off-the-wall mechanical genius. To find peace of mind they travel to the outbacks of Australia and meet up with a desert mining town full of zany individualists.
In the movie, Pete builds a newspaper folding and launching machine accessory for his Mini-Moke car to make his 'paper round' easy to do.  He also exhibits a kinetic art piece at his sister Rikky's art show opening. It elevates and smashes eggs.
Later he heads off with Rikky to a mining town, and eventually they get a plot to mine. They need a rock drill so Pete builds the mechanical horse with multiple rock drilling heads at one end, and a hydraulic back hoe at the other. 
Produced by
Bryce Menzies …. executive producer
David Parker …. producer
Nadia Tass …. producer
Timothy White …. co-producer
Writing credits
David Parker   ….  writer
Cast includes
Stephen Kearney Pete Menzies
Nina Landis Rikky Menzies
Art Department
Aaron Beaucaire …. mechanical props assistant
Steve Mills …. mechanical props assistant

Back in November 2004 I contacted David Parker at his film production company in Port Melbourne, Victoria, Australia. He still had the "Mechanical Horse" from his "Rikky and Pete" movie (released in 1988).  In a small garage there were various props from various movies. Here's some pics of the walking machine as it was then.

The air-cooled engine above was a prop. The real motor was electric and is shown below.

Detail of knee joint.

Knee-joint close-up.

See other early Walking Machines here.

Tags: , , , , , , , , , ,

1987-88 – “First-Step” Quadruped Walking Machine – David Buckley (British)

First-Step by David Buckley  September 1987 
Four Legged Walking Robot, uses 3-D pantograph arrangement to produce a gravitationally decoupled leg mechanism similar to that used by Shigeo Hirose, A Study of Design and Control of a Quadruped Walking Vehicle, The International Journal of Robotics Research, Vol 3, No. 2, 1984.
Won a Silver medal at the 1988 Model Engineer Exhibition.

Photograph – Scale Models International, April 1988, p207.
Design and building started September 1987.
Size – body about 6" * 6" * 6", leggs extend to about 15" * 15".
Operational area – 4ft * 2ft plus host computer.

First-Step was Britain's first electrically powered advanced walking robot. (Britain's only other [circa 1987] 'walker' had six pneumatically powered legs.)

See David's page on First-Step here for the complete story.

Tags: , , , , , ,

1976 – KUMO-I 4-Legged Walking Machine – Hirose & Umetani (Japanese)

Research on Quadruped Walking Machines    

KUMO-I (1976, see photo above), PV-II (1978-1979, see here). The method of locomotion called "walking" requires considerably more actuators than the wheel me thod of locomotion, the drive system is heavy; and it is not simple to control. However, walking machines, because they can move while separately selecting the point of leg contact with the ground by adapting to the shape of the terrain, are fully practical, depending on the use, because they have such characteristics as:

1) Can move stably over a rugged surface, and can pass over fragile objects on the ground surface without touching them.

2) Can make holonomic omnidirectional motion without slipping or damaging the ground surface.

3) Utilizing the degrees of freedom of the legs it can become a stable and active platform even on a rugged surface when stopped for some manipulation task.

We have been working on walking machine research since 1976. As for the number of legs, we selected four, which is the minimum number capable of executing statically stable walk. Photo. 1 (above) is the first generation model "KUMO-I", that was first manufactured on the model of a daddy-longlegs spider and that has a leg length of 1.5 meters and a weight of 14 kg. Photo. 2 (here) is a second generation model (PV-II). The leg length is 0.9 meters and the weight is 10 kg.

Fig.1 Power consumption generated by the hips and knee joints during horizontal locomotion (The negative power cannot be regenerated, and a large amount of energy is consumed.)

When a leg is designed as shown in Fig. 1, the actuators on hip and knee joints consume positive and negative power, but a normal actuator can not regenerate the negative power. For this reason, a large amount of energy is lost. We found out that this is the reason why the energy efficiency of conventional walking vehicle is so low. In order to improve this, GDA (gravitationally decoupled actuation; a drive system configuration method that separates drive in the gravitational direction and drive in horizontal direction) shown in Fig. 2 was introduced into the prototype model. Fig. 3 indicates the 3-D pentagraph mechanism for GDA that was utilized in the PV-II. This expands the prismatic motion of the three orthogonal axes provided on the torso part, lightens the legs, and simplifies their control. In 1979, the PV-II was the world's first success in sensor based stair climbing utilizing leg-end tactile sensors and posture sensors.

Fig.2 A leg mechanism that prevents the negative power loss of Fig.1

   1. Orthogonal coordinate mechanism (basic form)
   2. The mechanism used in KUMO-I
   3. The 2-D pentagraph mechanism

Fig.3 The newly developed three-demensional pentagraph mechanism and leg-end orientation control mechanism


   1. Shigeo Hirose, Yoji Umetani; Some Consideration on a Feasible Walking Mechanism as a Terrain Vehicle, Proc. 3rd RoManSy Symp., Udine, Italy,, , pp.357-375 (1978)
   2. Shigeo Hirose, Yoji Umetani; The Basic Motion Regulation System for a Quadruped Walking Vehicle, ASME Publication 80-DET-34,, , pp.1-6 (1980)
   3. Shigeo Hirose, Yoji Umetani; A Cartesian Coordinates Manipulator with Articulated Structure, Proc. 11th Int. Symp. on Industrial Robots, Tokyo,, , pp.603-609 (1981)

Tags: , , , , , , ,

1960c – “Golden Horse” Walking Machine – Maratori – (Italian)

An entirely different approach by Spartaco Maratori(8) produced his 'Golden Horse' which, in the final analysis, is somewhat similar to Shigley's approach. Maratori based his concept on an analysis of the locomotion of the horse. He studied the way horses walk, trot, and gallop and after carefully cataloging the various leg motions, attempted to duplicate a horse with mechanical linkages, His concept appears in Figure 3: the source of the design is patently obvious. Maratori presented many refinements in his description of his concept but in almost all cases these refinements were to more closely duplicate the form of an animal's leg. Where this was not the intent, the refinements were aimed at an increase in stability through an increase in the number of legs.
Maratori calculated that his machine could achieve a speed of about six miles per hour with a trotting motion or 4.5 miles per hour with a walking motion. These speeds were for a machine about the size of a horse and his calculations did not consider the inertia forces. Since his treatment in the quoted reference was more or less descriptive, he may well have considered the inertia forces and found them acceptable.
Maratori discussed the advantages and disadvantages of his concept over both conventional vehicles and animals. When compared to 'beasts of burden' he concluded that: 'Some advantages are at the quadruped's side as for instance:— To have a brain and some judgement which helps him in choosing his way in finding a firm foothold.' In effect, Maratori's machine has the same problem as Professor Shigley's: how to control the machine, since Maratori's allusion to the animal brain and judgement implies the control process.

8. Maratori. S., " Project Chin Ma (Golden Horse)", Brochure published by the author, Piazza Cavour 4-3, Chiavari, Italy.


From R.A. Liston – "Walking Machine Studies" – U.S. Army. 1966.

Tags: , , , , , , , ,