Analysis, Simulation, and Control of Blindfolded Walking
MetadataShow full item record
Publisher:The Ohio State University
Series/Report no.:The Ohio State University. Department of Mechanical and Aerospace Engineering Honors Theses; 2017
People can walk fairly straight with their vision because vision provides the sensory feedback needed to control our movement direction robustly. However, without vision, for instance, when people are blindfolded, they cannot in general walk straight even if they think they can. No quantitatively accurate theory has been able to explain why individuals perform blindfolded walking so differently and explain the heading angle deviation in the blindfolded walking and what causes people not to be able to walk with a solid heading angle without using their eyes. In this research, we use human subject experiments to explore how physical asymmetries in blindfolded walking (specifically, introducing a weight to only one side of the angle, adding a knee brace, sound frequency) affect being able to walk in a straight line and how much the resulting deviation is caused by a certain physical asymmetry. We found that the asymmetric knee brace and the asymmetric sound input affected the trajectories of blindfolded walking, causing them to turn in a systematic direction, although there was considerable trial-to-trial variability in turning behavior. In addition to these experiments, we obtain a probabilistic model to simulate blindfolded walking trials based on data from normal walking data (without blindfolds). This simple probabilistic model simulates curved trajectories as observed in blindfolded walking. Normal people will not do blindfolded walking in daily life, but such blindfolded walking simulates the experience of blind people or people with limited vision. The unpredictability of the blindfolded walking can lead to a severe safety hazard. It would be useful to develop a portable robotic or sensory augmentation device to guide those people to walk straight. Our goal is to build a device using GPS and vibration feedback to correct blindfolded walking trajectories, for which we present ongoing work on device development.
Undergraduate Research Fellowship
Academic Major: Mechanical Engineering