Investigation of Human Factors in UAV Accidents Based on Analysis of Statistical Data

THIS ARTICLE IS BASED ON ANOTHER FLAGGED FOR PLAGIARISM BY IEEE. —Paged queued for deletion.

US armed forces UAV accidents

In 2011, Nasir and Shi-Yin produced research towards the impact of human factors in unmanned aerial vehicle using statistical data. It is said that human factors differ from and are greater likelyhood than those of manned flight. Due to the separation of the aircraft; there are issues involving optimum human performance, loss of sensory cues, delays in control and communication loops. Human factors categories relate to issues of alerts/alarms, display design, procedural error, skill-based error, or others. Human factors range and vary towards past accidents. There are many other causes to UAV accidents other than human factors. But the research article focuses on human factors only as the independent variable.

Illustration 1 : Types of UAV accidents used for the article
Types of UAV Human factors involved
Hunter UAV 32 Human factor involved (Hunter) 15 47%
Shadow UAV 24 Human factor involved (Shadow) 5 21%
Total US Army UAV: 56 Total UAV human factors involved 20 36.70%
Illustration 2 : Relation of human factors between models of UAVs
Comparison Means difference
Hunter vs. Shadow 0.26

Methods

Research approach

The research focuses on UAV and human factors relationships and understanding if there is a significance of the results. Human factors is the independent variable for the article. Human factors is further broken down into variables to seek where the majority of human factors is being caused of UAV accidents.

Sample

The sample of 56 UAV accidents was used in the research and analysis. The sample used is all the relevant UAV accidents and no further population can be gathered from this further.

Materials

Results where extracted from US Armed forces for the article. This makes the source trusted and reliable of results.

Variables

Two classification towards the accident data and the relationship between human factors and other conclusions of past accidents of the data. The fist category was the cause of the crash including nonrelated human factors (extraneous variables) and human factors. The second category was broken down from human factors into subcategories. Human factors was only taken into account if it involved the flight crew. “Human Factors” and “Aircraft” was separated if inadequate display to the crew causing it to be the Aircraft at fault. UAV accidents would also be included under human factors if it was an important factor towards the accident. Judgement on categories was based on personal judgement of the US Armed Forces.

Hunter

The hunter is a twin-engine tactical UAV aircraft.

Illustration 3 : Accidents cause breakdown: Hunter
Accident Issue Number of accidents Percentage
Maintenance 4 9%
Human Factors 15 47%
Aircraft 16 50%
Unknown 1 3%

The Hunter does not have a automated landing system (unlike the shadow) so the Hunter needs an external pilot in order to land and takeoff .

Illustration 4 : Human factors breakdown: Hunter
Human Factors Issue Number of accidents Percentage
pilot-in-command 1 7%
Alerts and Alarms 2 13%
Display Design / Stimulation Awareness 1 7%
External Pilot Landing Error 7 47%
External Pilot Takeoff Error 3 20%
Procedural Error 3 20%

Data analysis

The Hunter does not have an automated landing system (unlike the shadow) so the Hunter needs an external pilot in order to land and take off. This resulted in 47% of human factors relating to the landing phase and a further 20% caused due to take off by the external pilot form human factors. The external pilot needs to control the UAV by giving commands are opposite.
The display design causes absence of situation awareness towards all the essential information conveyed to the pilot in command in order to fly.
Another issue noteworthy is not following procedures properly. This came from multiple errors of not correctly following procedures and checklist.
This issues needs to be addressed to prevent future accidents.

Shadow

Compared to the Hunter, the shadow is a lighter, smaller, short range surveillance UAV. It has the operating density to fly at 14,000 ft at the cruising speed of 82 kts and an operational range of 68nm [Bibliography item 2 not found.]].

Illustration 5 : Accidents cause breakdown: Shadow
Accident Issue Number of accidents Percentage
Maintenance 2 8%
Human Factors 5 21%
Aircraft 10 42%
TALS 6 25%
Unknown 4 17%
Illustration 6 : Human factors breakdown: Shadow
Human Factors Issue Number of accidents Percentage
Pilot-in-command 2 25%
Alerts and Alarms 2 25%
Display Design / Stimulation Awareness 2 25%
Procedural Error 2 25%

Data analysis

Human factors with the Shadow UAV was largely due to the transfer of control during a flight and an accident also linked to the Hunter of the cause. The Shadow human factors relating to transfer of control was another instance of not following the checklist prior to the transfer. Another issue is not conveying the commands effectively to the pilot in command of the Shadow and not acknowledgment of commands. This would be fixed by a clear indication by the display design.

Conclusion

The sample size was not adequate to make a justifiable claim, so results will vary by increasing the sample size. The authors of the article say that there is a correlation does exist between accidents and human casual factors. Due to the 21% (Shadow) and the 47% (Hunter) an improvement is needed towards the human factors in UAV by identifying and further research. It is noted that electrical/mechanical failures are linked to contributing to UAV human factor accidents. Finding display issues and testing on pilot simulation to find improvements on these issues will decrease the percentage of accidents accruing in the future. The automation was said to solve the problem of human factors but has come apparent that automation can fail itself. This is proven by the 25% of the Shadow UAV automated take-off and landing systems. It is also noted that the Global Hawk has high impact of automation, which lowered the situation awareness and lowered ability to response of faults [3]. UAV hand off to another station needs to be aware of what commands have been issued by the other pilot before take over through situation awareness. Situation awareness is linked to be responsible for 40% of human factor accidents in the UAV. The results give further reason to link manufactures with developers followed by extensive testing with pilot testing simulation to correct any design errors and general errors.

References
1. NASIR Manzoor M. & SHI-YIN Qin (2011).Investigation of human factors in UAV accidents based on analysis of statistical data.// IEEE explore Digital Library 2011
2. SCHAEFER R (2003). Unmanned aerial vehicle reliability study. // Office of the Secretary of Defense Washington DC February 2003//
+++ **Footnotes +++
2. ###

1. NASIR Manzoor M.
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=6154281&tag=1

Want to know more?

Nasir and Shi-Yin's (2011) article
explanation

Contributors to this page

Authors / Editors

Matthew MULLER (2013).


Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License