by Bryan Bergeron, Editor
Semi-autonomous robot toys and planetary rovers are exciting technologies, but the real promise of practical robotics is in alleviating human pain, suffering, and disability. Accidents, arthritis, and the normal decline in function with aging are a few reasons companies and universities in the US, Japan, and the EU are developing mechatronic prosthetics. You’ve probably heard about the various robotic prosthetic arm projects under development over the past several years. These arms have showcased advances in both myoelectric (control signals from muscles) and neurologic (control signals from the nervous system) control.
One of the latest examples of robotic or mechatronic prosthetic technology is an energy-recycling artificial foot developed by Art Kuo and Steve Collins at the University of Michigan. The foot (shown in the accompanying photos) uses the otherwise wasted energy of the normal heel strike to enhance the ankle push-off. The foot uses a microcontroller and two micro-motors to release the spring and reset the mechanism. An internal battery provides about 0.8W to power the assembly. It’s not clear from the literature how long the battery lasts between charges or replacement.
This mechatronic foot is significant for amputees because it promises to make the effort of walking less arduous. Conventional foot prosthetics require the wearer to expend about 24% more energy to walk — compared with walking with a biological foot. The new design drops the energy penalty by about 10%. The goal of most researchers in this area is to match and then exceed the mechanical efficiency of biological feet.
According to the inventors, the foot isn’t the first to recycle the energy of walking. All prosthetic feet based on a simple spring design store energy on impact and then release it. The problem with this generic spring design is that the energy is released as soon as the tension on the spring is lessened. However, this timing generally has no relationship with the energy required during push-off. Using onboard processing and a built-in battery, the foot determines exactly when to release the pent-up energy in the spring to maximally supplement push-off.
I liken the operation of the device to that of a mousetrap. The impact of the heel on the ground tenses the spring, and it latches when the spring is sufficiently tensed. Similarly, to set a mouse trap you have to tense the spring and then latch it to a trip mechanism. When a mouse touches the trip mechanism, the spring is suddenly released with devastating consequences.
It’s the same with the foot. The difference is that the burst of released energy is used to launch the person forward. In addition, the foot automatically resets itself when the heel hits the ground. The advantage of this design over previous designs is that the foot is self-contained. There are no cables to external batteries or controllers. I can envision a time when the efficiency of the design is improved to the point that the foot would generate enough energy to power the onboard electronics without the need for a battery.
The ancillary applications of this technology should be obvious – it’s a means of reducing the energy requirements of biped robots. I imagine the technique could be applied to any simple biped robot such as the Parallax Penguin (http://www.parallax.com) or the Lynxmotion Brat (http://www.lynxmotion.com), or a more complex robot such as a Kondo KHR humanoid robot (http://www.kondo-robot.com). In each case, the feet would need to be modified significantly because current production models don’t use spring feet. If you’re up for the challenge, you should read the original report that details the foot’s operation. “Recycling Energy to Restore Impaired Ankle Function during Human Walking” by Steven Collins and Arthur Kuo (Plos ONE, Feb. 2010, Vol. 5, Issue 3, e9307) is available at http://www.plosone.org. A PDF of the article is available for free download. SV