Evaluation of a foot-mounted slip sensor

Open Access
Hartman, John
Area of Honors:
Mechanical Engineering
Bachelor of Science
Document Type:
Thesis Supervisors:
  • Henry Joseph Sommer III, Thesis Supervisor
  • Mary I Frecker, Honors Advisor
  • stick-slip friction
  • heel-strike
  • friction tester
Bipedal robots have become a reality in recent years but still suffer from numerous operational difficulties, many of which concern maintaining balance. As robot locomotion advances, a dynamically stable gait simulating human walking will be a major focus since it permits much quicker movement. Slip during footfall will be a primary concern. Since slip-related falling is a prominent cause of accidental injuries in the United States, the study and measurement of human gait patterns, especially during incidences of slip, have been conducted using varied instrumentation. A novel processing algorithm has been proposed that utilizes an unscented Kalman filter to identify slip events. Evaluation of the sensor required a static test to simulate conditions during stance phase and a dynamic test to simulate conditions during heel-strike. Initial data from the static test indicated the Kalman filter had potential, but the acceleration transients produced from stick-slip motion were not sharp enough to properly identify slip initiation or sticking. Since most falls occur during heel-strike, data from the dynamic test will be a better indicator of the viability of the proposed sensor.