The state of the art of climbing and walking robots

Citation

Waldron, K. (2009), "The state of the art of climbing and walking robots", Industrial Robot, Vol. 36 No. 4. https://doi.org/10.1108/ir.2009.04936daa.002

The state of the art of climbing and walking robots

Article Type: Viewpoint From: Industrial Robot: An International Journal, Volume 36, Issue 4

Keywords: Robotics, Conferences

The 2008 CLAWAR conference held at Coimbra, Portugal was the second held under the auspices of the independent CALWAR association. To my mind, it marked a coming of age in the technologies of climbing and walking robots, and of mobile robots in general. Climbing and walking robots are moving out of the laboratory and into industrial practice. The papers from the conference presented in this special issue exemplify this.

Like all new technologies, it has taken a long time, and there have been both triumphs and missteps along the way. It is now forty years since the first computer-coordinated walking machine, the “Phony Pony” built by Andrew Frank and Robert McGhee first walked. That is a longer period than the interval between the Wright Brothers’ first flight and the coming of age of aviation in the World War II. Evidently, fewer compelling applications have appeared for climbing and walking robots, and consequently there has been less investment in development and production. Nevertheless, we are now seeing applications in which our technologies are clearly better than the alternatives.

In this issue we have papers on climbing robots for inspection of offshore wind turbines, for remote NDT of objects with unknown geometry, and a commercialized climbing robot system for cleaning windows on glass-walled buildings. These papers are selected from a much larger number presented in the conference that included many other examples of commercialization of multi-limbed robots. To be sure, there are a much larger number of articles describing research results on systems that are still confined to the laboratory, but it is in the nature of research and development that this should always be the case.

The history of humanoid robots also goes back nearly forty years to the first Wabot constructed by Ichiro Kato and his students at Waseda University. This area of climbing and walking robots has attracted a great deal of support from both industry and research laboratories. This issue includes a highly rated paper on a humanoid control problem.

One of the most original papers in this selection is on an innovative concept for a robot that can crawl through collapsed buildings and locate victims. It uses a flexible envelope that covers the entire machine and can be driven to propel the machine through narrow passageways.

As we move forward workers in this field are addressing increasingly difficult problems. One direction of active research is toward increasing use of dynamic locomotion systems. This type of system offers greater speed and rapidity of maneuvering, but at the cost of having to actively stabilize the system. Another direction is the use of redundant sensing systems with sensor fusion techniques. Often in mobile systems a single type of sensor cannot reliably generate data about the environment, and the machine’s situation within that environment. Use of multiple sensing modes provides superior data for control and other purposes. Yet another theme that I expect we will see more in the future is deployment of cooperative teams of relatively simple robots. This allows robust data collection over extended areas.

The excitement of robotics is in its nature as a synthesis of mechanical systems, electrical systems, information systems and sensors into machines that are computers that can sense, model and physically interact with the world. Climbing and walking robots represent this concept in a very pure form. That is what makes the work described in this set of papers rewarding, and also representative of the true state of the art.

KennethJ. WaldronDepartment of Mechanical Engineering, Stanford University, Stanford, California, USA