UAS and Manned Aircraft Autonomy

When discussing the interesting yet sensitive topic of automation in UAS, one must first understand that automation does not have to be an all or nothing approach. There are varying levels of automation that can be implemented in a system from minor to major with plenty of options in between. Diving into automation research, it can be devised that there are two categories of automation: static automation and adaptive automation.  

Static automation is considered “hard-wired into the system” by way of the designer making the decisions as to whom or how a task should be performed (Marshall et al., 2011). In static UAS automation all decisions are made within the design of the system allowing for overriding of automation. Static automation could be called traditional automation. Adaptive automation is considered to be summoned by an operator event such as workload or situational event such as takeoff speed (Marshall et al., 2011).

 

As previously mentioned about autonomy not having to be an all or nothing approach, there are several levels of autonomy to help classify all of the options. Low level of Autonomy (Low LoA) is considered for levels 1-3 which encompass UAS with limited internal situational awareness as well as low levels of automation for tasks (Marshall et al., 2011).

 

Mid LoA consists of levels 4-6 which include systems that interact with a human approximately 50% of the time while operating autonomously the remaining time (Marshall et al., 2011). With the Mid LoA the UAS has moved from a static environment which is considered low risk to a dynamic environment which is considered mid-risk.

 

High LoA consists of levels 7-9 which are levels where the UAS will have very little interaction with humans including execution of goals and will have real-time planning capability with high adaption and decision-making capability (Marshall et al., 2011).

                                                                    

Beyond level 10 the UAS would be considered to be performing at a human level where there is no longer human oversight but simply acquiring data (Marshall et al., 2011). This category for UAS can propose a strong artificial intelligence (AI) approach which means human cognitive behaviors and responses can be replicated by a machine (Marshall et al., 2011).

 

There are most definitely different considerations for manned verses UAS when it comes to autonomy. With manned systems, the pilot is in the cockpit and the aircraft at its core still needs to be flown by humans. The manned system has a pilot that must remain vigilant in the cockpit and capable of transition to manual flight at any given time. With UAS, the pilot is operating the system from the ground. Situational awareness is vastly different since the pilot is receiving all situational awareness inputs from the system’s screen itself and nothing from other senses as a manned pilot would. 

 

I think the aviation industry in its entirety, including UAS, is on the right track to attempt to implement autonomy where there is potential to increase safety. There are examples of manned flights where autonomy failed the system and placed the aircraft in danger such as 2008 OF72 flight from Singapore to Perth. However, other autonomous functions such as detect and avoid technology will greatly increase NAS safety with the integration of UAS and therefore needs to be continually researched and developed. 

 




References

 

Marshall, D., Barnhart, R., Hoffman, S., Shappee E., & Most, M. (2011). Introduction to unmanned aircraft systems. Chapter 7. CRC Press LLC, Boca Raton, FL. 

Comments

Post a Comment

Popular posts from this blog

Analysis of The Hub – UAS Design Application

This was my first time working with The Hub application created by ERAU for design and testing of unmanned aircraft systems (UAS). I found the system to be highly entertaining and relatively user friendly. What I found most challenging was the knowledge base on UAS designs themselves. I was making selections that others would probably have laughed at. However, I did come up with several modified designs, some that were successful and others that were not. During this trial I stuck with one model so I could make adjustments and compare those adjustments with previous designs in attempt to tweak it to perfection. Perfection was far from achieved but the system was still great to work with.    The most successful design was a Tern fixed-wing UAS and a man portable ground control station (GCS). When making these selections, the system was very user friendly and allowing the engineer to make selections and adjustments while providing an up to date calculation of the effect of those...
Read more

Physiological Issues in UAS

When considering taking medications in an aviation protocols in place, there are several determining factors that are involved to decide whether or not a mission should be flown. Manned aircraft pilots as well as UAS pilots experience similar limitations and cautions as it pertains to medications because the risks are great if not taken seriously. There are many physiological factors that affect pilot performance and those can include stress, fatigue, and drugs consumption. Anything that limits the pilot’s faculties and decision-making skills is considered detrimental to operations.    When examining UAS pilot’s levels of decision-making requirements, complexity in tasks, and skill set necessary to perform missions, one could argue that the over-the-counter OTC medications that pose the most significant risk to UAS pilots would be antihistamines and/or decongestants due to their potential to cause cognitive deficits. Now I will argue that decongestants are on the approved list...
Read more