Abstract
Purpose :
To develop a mixed reality (MR) headset with the customized obstacle detection software and voice guided navigation (CU glasses) for avoiding indoor obstacles using a spatial mapping technique and compare its performance with a white cane.
Methods :
Ten normal sighted people aged 29-65 years were recruited into this pilot, cross-sectional study. The experiment room was set up with ten soft non-harmful obstacles including 4 face-level, 3 trunk-level and 3 knee-level obstacles. All subjects were blindfolded and trained to use both the CU glasses (CUG) and a white cane (WC) before entering the experiment room. All subjects were randomly assigned to 2 groups (CUG or WC) and walked through the 10 obstacles from entrance (point A) to exit (point B) one by one. After that all subjects were switched the group and walk through the 10 obstacles once again from point B to point A. The number of bumping into obstacles and time from entrance to exit were recorded.
Results :
The CUG group had lower number of obstacles hit compared with the WC group
The mean (SD) number of obstacles hit of the CUG group, and the WC group was 0.7 (+/- 0.48) and 1.4 (+/-0.84) respectively. The mean difference (95% CI) of number of obstacles hit between 2 groups was 0.7 (0.12, 1.27), p=0.016. The total time (SD) from start to finish of the CUG group and the WC group was 5.18 (+/-3.42) and 3.21 (+/-2.78) minutes respectively. The mean difference (95% CI) was 1.97 (-4.39, 0.45), p=0.111. All subjects completed the test without any falls or injuries.
Conclusions :
The CU glasses can decrease the number of obstacles hit compared with white cane in our indoor experimental environment. Further studies in the blind subjects are needed to assess its full potential as an assistive device for improving orientation and mobility for the blind.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.