Thesis
Investigation of the biometric effects and flight performance of drone pilots while inspecting bridges with auditory and visual proximity aids
University of Iowa
Master of Science (MS), University of Iowa
Spring 2024
DOI: 10.25820/etd.007388
Abstract
Bridges are an integral component of the infrastructure system in the United States, but almost 45,000 bridges were in “poor” condition in 2020. Current inspection practices are expensive to conduct, difficult to schedule, and can pose safety risks to inspectors. Furthermore, there are several concerns with the quality and reliability of bridge inspections. In the worst case, the lack of diligent inspections may lead to bridge collapses and jeopardize public safety. An increasingly common solution to many of these issues is using drones to assist in performing routine bridge inspections. Drones can access difficult-to-reach areas on the bridge, preserve inspectors’ safety, and inspect bridges without disrupting traffic flow. By taking photographs and videos, they can also be more efficient in gathering documentation of the bridge’s condition. To implement this new drone-based system, inspectors and pilots must receive adequate training and be equipped with the necessary tools to conduct safe inspections. Safety is critical, and crashes are quite expensive. A proximity detection system is one tool to alert pilots of impending danger. Many modern vehicles utilize these systems when moving in reverse. All three dimensions must be considered in aviation, especially when navigating a large bridge structure. Proximity detection systems often utilize aids to alert the pilot when the drone is too close to an obstacle. Two common modalities for communicating proximity are visual and audio.
In human analytics studies like this one, several methods exist to collect data on how humans interact with their environment and when presented with different stimuli. One standard method is using biometrics to assess physiological and psychological responses. An electroencephalogram (EEG) monitors human brain activity, an electrocardiogram (ECG) monitors cardiac activity, and eye-tracking glasses are used to observe visual behavior. The iv metrics collected in these studies have been extensively researched and may be used to assess psychological experiences such as cognitive workload, vigilance, visual attention, and stress.
This study utilized a drone-based bridge inspection simulation to investigate the effects of different capacities and modalities of proximity aids on pilots operating a drone to inspect a bridge. A baseline condition without any proximity aid was used, along with visual-only, auditory-only, and combined visual and auditory proximity aids. The four conditions were evaluated in two phases. In Phase 1, the baseline condition and the visual-only proximity aids were assessed. In Phase 2, the auditory-only and combined visual and auditory aids were assessed. An EEG, ECG, and eye-tracking glasses collected physiological and psychological data. The experimental task consisted of piloting a simulated drone through a series of checkpoints near a bridge, indicated by floating spherical orbs. At each orb, the participant took a picture of the bridge. The orbs varied in difficulty, and the participant tried to complete the task efficiently while maintaining a safe distance from the bridge.
The collected data were separated into a unit of analysis: each orb. A series of statistical analyses addressed the proposed research questions. Analysis methods, including t-tests, linear regressions, and structural equation models (SEMs), were used to identify data trends. Associations among the proximity aid conditions, physiological and psychological experiences, and performance measures were observed. The data indicated increased performance in the presence of any proximity aid. Moreover, performance improved with the auditory proximity aid, while decreased performance was observed with the visual-only and combined proximity aids. Future work may involve conducting a similar drone-based experiment in a true bridge inspection environment.
Details
- Title: Subtitle
- Investigation of the biometric effects and flight performance of drone pilots while inspecting bridges with auditory and visual proximity aids
- Creators
- Jacob M Konz
- Contributors
- David Nembhard (Advisor)Daniel McGehee (Committee Member)Thomas Schnell (Committee Member)
- Resource Type
- Thesis
- Degree Awarded
- Master of Science (MS), University of Iowa
- Degree in
- Industrial Engineering
- Date degree season
- Spring 2024
- Publisher
- University of Iowa
- DOI
- 10.25820/etd.007388
- Number of pages
- xi, 138 pages
- Copyright
- Copyright 2024 Jacob M Konz
- Language
- English
- Date submitted
- 04/22/2024
- Description illustrations
- illustrations, tables
- Description bibliographic
- Includes bibliographical references (pages 75-87).
- Public Abstract (ETD)
- Bridges are an integral component of infrastructure in the United States. Current inspection practices are expensive, lack quality, and pose safety risks to inspectors. A solution to many of these issues is using drones to conduct bridge inspections. Drones can access difficult to-reach areas on the bridge, preserve inspectors’ safety, and conduct inspections without disrupting traffic. To safely implement this system, drone pilots must have the appropriate tools to avoid crashing. A proximity aid that detects nearby objects can prevent crashes. These aids alert the pilot when objects are too close, often through sight and sound. This study used a drone-based bridge inspection simulation to investigate how pilots performed with different aids, including no aid, a visual aid, a sound aid, and a visual aid with sound. Brain activity, heart rate, and eye-tracking data were collected. The study objective was to measure task performance with different aids while the participant flew the simulated drone through a series of checkpoints near a bridge. At each checkpoint, the participants took a picture of the bridge. They were instructed to complete the task quickly without getting too close to the bridge. Several approaches measured how well the different aids performed. Relationships among the aid types, participants’ behaviors, and performance measures were observed. The data indicated that any of the aids were better than no aid. Additionally, participants performed better with the sound aid and worse with the visual-only and combined visual and sound aids.
- Academic Unit
- Industrial and Systems Engineering
- Record Identifier
- 9984647555902771
Metrics
8 Record Views