Pedestrian lateral foot placement and lateral dynamic instability of bridges

The most often purported mechanism causing the lateral dynamic instability of the London Millennium Footbridge is the synchronisation of footsteps to the lateral structural motion. However, evidence from full-scale measurements and treadmill tests has challenged this notion. Instead, an active control of foot placement is advocated to be the source of destabilising forces to the structure, occurring even without synchronisation. This is to say that, while walking on a laterally oscillating surface, pedestrians maintain their balance primarily by controlling the position of their feet, rather than adjusting the timing. Similar behaviour was previously observed in experimental tests measuring the response of pedestrians to an impulsive perturbation of gait. The analysis of the collected data suggested a simple linear foot placement control law, whereby the position of the foot at the instant of foot placement immediately following the perturbation depends on the instantaneous lateral velocity of the centre of mass and a constant offset. However, it is has been uncertain whether the same foot placement control law applies while walking on laterally oscillating structures. To test this proposition, an experimental campaign was conducted on a laterally oscillating treadmill with a test subject monitored with an optical motion capture system. The motion of the body centre of mass and the position of the feet were identified and analysed. It was found that a simple linear foot placement control law applies. Further tests were conducted to test the influence of the visual information on pedestrian stepping behaviour using virtual reality delivered via a head mounted display. It was found that the identified foot placement control law is very robust for different walking surface conditions and visual environments.