Hugh Herr, from the Massachusetts Institute of Technology Media Laboratory,designs state-of-the-art prosthetic limbs and mobility aids. But to do this,he has to understand how we move; `I study humans to build synthetic versions of humans' he says. However, none of the recent studies of human walking have considered the effects of angular momentum as we move, and only one study,based on measurements of a single step by H. Elftman in 1939, has attempted to measure angular momentum directly. As a comprehensive model of walking is the holy grail of biomechanists, prosthetics designers and robotic engineers, Herr and postdoc Marko Popovic decided to make the first accurate measurements of angular momentum during steady walking(). Recruiting ten fit young walkers, Popovic and Herr fixed markers to each individual's body before filming them as they walked steadily across a force plate, ready to calculate the angular moment of each subject's trunk, limbs and head to get a better understanding of the role of angular momentum in walking.
Which was when the hard work began. Having digitised each subject's limb,trunk and head positions, Popovic and Herr built a complex computer model to calculate each body segment's angular momentum while sauntering at steady speed. According to Herr, `the challenge is to get realistic mass distributions in the limbs... the shape has to be right.' He adds `it was an insane amount of work', but after months of painstaking computation, the team had calculated the angular momentum of each individual's 16 body segments, and were ready to see how angular momentum varied during steady walking.
Plotting each individual's whole-body angular momentum, the team could see that it fluctuated slightly, but was essentially zero throughout a walking cycle at steady speeds, despite the large angular momentums generated by the swinging limbs and other body segments. Herr explains that opposing limb movements cancel each other's momentum in three dimensions, and he suspects that whole-body angular momentum is minimised to reduce the metabolic cost during steady walking.
Herr also compared his results with current walking models. He explains that walking is often modelled as an inverted pendulum; the foot acts as the pivot and the body's entire mass is represented at a single point mass at the end of the pendulum. According to Herr's measurements, the inverted pendulum model works well; the body can be considered as a single point mass. However the inverted pendulum model fails when you assume that pressure exerted by the foot acts at a single fixed pivot point; it incorrectly predicts the forces acting on the body's centre of mass unless the pressure point is modelled as moving along the foot.
Herr also emphasises that walkers only minimise their angular momentum,with minor fluctuations around zero, while they're walking steadily. Ask them to do a turn from the Ministry of Funny Walks and it's a completely different matter. In those circumstances walkers have to modulate their angular momentum to counteract destabilising forces and maintain their balance. Herr admits that it isn't clear how walkers modulate the body's angular momentum to improve balance and manoeuvrability, but he says that `I hope this study will motivate additional studies in the field'.
