1
Phototextured Visual Databases for Helicopter and Fast-Jet
Flight Simulation
Nick Tatlock
Air Systems Sector, Defence Evaluation and Research Agency
Farnborough, Hampshire, United Kingdom
Introduction
This paper describes two separate simulation
trials that were sponsored by the Directorate of
Operational Requirements (Air) at the United
Kingdom Ministry of Defence. Both trials were
concerned with the development of phototextured
visual databases for helicopter and fast-jet flight
simulation and, the effect on pilot performance
and workload when the database resolution was
varied.
This research was undertaken by the Defence
Evaluation and Research Agency (DERA) within
an applied research programme which addresses
many issues related to simulation for air systems
training.
Until recently the role of flight simulators has
been limited to carrying out procedural type
training such as instrument flying coupled with
emergencies and malfunctions. This has been
due to several factors such as the design
limitation of the simulator, the fidelity of the
simulation, the quality of the display produced by
the image generators and the user requirements.
As the complexity and roles of aircraft change,
along with increased costs of operating and
maintaining them, more reliance will be placed
upon the role that flight simulation will play in
providing realistic cost-effective training for
aircrew. Flight simulators of the future will have
more demands placed on them to satisfy training
requirements and the maintenance of flying skills.
Their use will be extended to include many
aspects ranging from training of the ab-initio and
maintenance of currency/combat ready status to
complex full mission training and mission
rehearsal prior to deployment on operations. The
ability to network simulators will further enhance
their capability to support distributed mission
training with adversaries located in different world
locations carrying out collective training and
operations.
Flight training or exercises that rely totally on
external world references such as tactical flying or
battlefield operations will be critically dependant
on the effectiveness of the simulation of
appropriate out-of-the-window (OOW) scenes.
Attempts to increase the realism of simulation are
mostly beneficial to the performance of mission
tasks and the design aim of a simulator visual
system should be to display a realistic visual
scene of the outside world. The use of present
day technology will permit this to be achieved to a
high degree of realism but it will probably never
be completely possible to recreate the total
environment. Compromises will always have to
be made in the representation of the real world
and the important detail needed to be able to
perform the task.
In making these compromises, designers need
detailed information about the specific tasks to be
simulated and the features that are necessary for
the successful performance of these tasks. Often
visual databases that look quite good are found to
be disappointingly inadequate because the
customer and designer have not been able to
identify, specify and prioritise visual features that
are important to the tasks being simulated.
There are several fundamental problems in the
generation and display of a simulator visual
database which relate to display resolution not
being as good as a pilot's eyes, scene brightness
and contrast are generally much less than normal
daylight and two-dimensional displays lead to
compromise in presenting three-dimensional
effects. Most manufacturers and suppliers of
military simulation systems often propose
solutions that are constrained by the limits of
current technology though as technology
advances, improvements in display resolution will
no doubt overcome many of these limitations.
The basis of this research was to carry out two
separate simulation trials in which pilots were to
perform discrete flying tasks on phototextured
visual databases. Several discrete tasks were
selected that were typical of helicopter and fast-
jet flight over land. As the resolution of the visual
database was varied the effect on piloting
performance and workload was recorded.
Helicopter flying tasks consisted of navigation at
1500 and 500 feet, low flying at 150 feet, tactical
