Studying Pilot Cognition in Ship-Based Helicopter Landing Maneuvers - - PowerPoint PPT Presentation

David Rancourt, PhD Thesis Defense David Rancourt, PhD Proposal Vertical Lift Research Center of Excellence 1982-2021: 40 Years of Rotorcraft Research Excellence

Studying Pilot Cognition in Ship-Based Helicopter Landing Maneuvers

Dev Minotra and Karen Feigh, Cognitive Engineering Center, School of Aerospace Engineering, Georgia Institute of Technology

1

2

Overview of the Project

• Focuses on the task of landing a helicopter on a moving ship deck.
• It is tied to a number of challenges not faced in ground-based landing

maneuvers.

• At Georgia Tech, we are conducting a study to develop intelligent guidance and

cueing to assist pilots in ship-board landing maneuvers.

Introduction

3

Introduction

Complexity elements - landing a helicopter on a ship

• Moving landing spot
• Limited visual cueing
• Spatial confinement in landing area
• Airflow hazards

Project Objectives

• Understand the cognitive task of landing rotorcraft on

ships. Challenges in Pilot Assist Function

• Matching the guidance algorithm output to how pilots

think.

• Choice of display modality and location of displays in

cockpit (i.e. heads-up vs. heads down).

4

Overview of this Study

• What are the cognitive issues in landing a helicopter on a moving ship?
• We conducted a Cognitive Task Analysis.
• This involved interviews with four navy instructor pilots with shipboard landing experience

(August 21, 22, 2017).

• Also interviewed ground-based pilots in an earlier study.
• Total recorded time was 6 hours 17 mins.
• Instructor pilots

― Number of shipboard landings are 200 - 810 ― DDG/FFG related experience 50 - 2000 hours ― Most experience with H60 Seahawk variants ― (MH-60R, MH-60S, SH-60B)

5

Methodology

• Applied Cognitive Task Analysis (Militello and Hutton, 1998).
• Adapted for the purposes of our study based on a previous preliminary

study with ground-based pilots.

• Focuses on expert-novice differences to reveal cognitive demands.
• Each interview consists of the following –

Surface Level Interviews Critical Incident Questions Knowledge Audit Questions

6

Methodology

• Surface-level interview

– Overview of task in 3 – 6 steps – Flow of basic steps involved

• Critical incident question

– “Take your time to think about an incident in which you had to bring in your helicopter piloting experience in a difficult shipboard landing operation that a novice couldn’t have carried out successfully. Can you please narrate such an incident?”

• Knowledge audit questions

– Big Picture – Past, Present, and Future – Noticing – Job Smarts – Self Monitoring – Anomalies – Equipment Difficulties

7

Overview of Ship-Based Approach and Landing

High cognitive load Lower levels of cognitive load

Line up

8

Cognitive Demands Identified

Transition in Visual Scans

– From instruments to external environment – This is a decision point – the timing depends on visibility or time-of-day – Switch is made once the ship is visible – At day time, once the helicopter has descended below clouds – Done before missed approach point (.5 NM) – Pilot guidance format – heads up display

Illustration of Possible HUD

9

Cognitive Demands Identified

Line-up with Ship’s Heading

– There are two types of line-up – straight and offset (a fixed angle). – Line-up starts after the ship is found and is continually done throughout the approach. – Novice pilots may get this step wrong. – Deviations are not noticed immediately. Leads to wave-offs.

Straight line-up Offset line-up

10

Cognitive Demands Identified

Crossing the Edge of the Deck

– In the past, it has led to deaths – landing gear stuck on nets – Interview revealed an instance of tail wheel tapping nets – Estimation of the height requires nuance judgement – Aircrewman plays a role in checking for separation – Design implication: auditory feedback, predictive feedback

1999 Accident at USS Pecos 6 marines and a Navy corpsman were killed.

11

Cognitive Demands Identified

Alignment of Probe Within RSD

– RAST system is still used – Not visible to pilot – Fine adjustments: Alignment of RAST probe and RSD requires aircrewman assistance – Design implication: auditory or tactile feedback

12

Cognitive Demands Identified

Decision to Touchdown

– Heavy sea state requires some hovering prior to touchdown – Novices spend more time hovering – Novices tend to chase the motion of the deck although without necessarily being aware of it. Type of spatial disorientation. – Time to hover ranges: <1 min to 10 mins – Sea state and visibility affect hovering time – Waiting in hover state can be frustrating

13

Cognitive Demands Identified

Decision to Touchdown

– Two ways to describe ‘when’ to initiate a touchdown

• When a quiescent period is identified
• When the deck is at level with the horizon bar

– Identifying the right time to touchdown is a nuance judgement – Intelligent guidance should reduce the overall hover time – Two approaches to achieve this -

• Path guidance leading to touchdown - cross the deck and directly touchdown
• Cross the deck, hover, and wait for the aid to indicate the touchdown period

Horz bar daytime Horz bar nighttime

14

Other Cognitive and Technical Issues

• Horizon bar - the size and position may not be ideal.
• There is no heads-up display in cockpit.
• Pitch and roll conditions are not electronically relayed into cockpit. Information

accuracy issues.

• Night vision goggles -
• Narrow FOV of 40 degrees.
• Visual horizon is not as visible through NVGs – issue .25 NM behind the ship.
• Equipment malfunctions – do not always manifest as electronic alerts. Differs from

expectations based on NATOPS manuals.

RAST trap and probe

NVG Vision

15

Discussion

• Summary of work

– Cognitive Task Analysis – Four pilots interviewed – Points of high cognitive load

• Transitions in visual scans
• Line up with ship’s heading
• Crossing edge of deck
• Alignment of probe with the RAST
• Hovering and touchdown decision
• Main implications

– Heads up display for pilot trajectory guidance – Finer adjustments involve auditory feedback – A number of system limitations were identified

• Confirmatory interviews scheduled

16

Acknowledgements

• Financial Support from the Vertical Lift Research

Center of Excellence (VLRCOE)

• Pilots at NAS Whiting
• Fellow VLRCOE Task 10 students & faculty
• Guidance from John Tritschler & James Prichard