SubscribeATC simulators are used for all levels and types of controller training for both civil and military operations. Current progress is evolutionary using advances in PC products with innovative simulation software and services.
Successful users are those who can clearly identify the training requirements and the media performance necessary to meet them. They are also able to appreciate the physical limitations of the current technology that they can afford to employ.
Recent History
Fifteen years ago air traffic control (ATC) simulation was revolutionised by the introduction of PCs and raster scan displays. The new designs with their modular flexibility reduced the cost of entry. This reduction in price and the simplification of the technology made simulators available to a wider range of users while at the same time changing the nature of the market and the players.
We are now in a period of evolution. Hardware costs are falling while performance is increasing. Users are happy to accept PC technology with some even supplying and supporting their own computer platforms. The manufacturers of ATC simulators are differentiating themselves by price, software features and services.
The Status of ATC Simulators
ATC simulators are presented as two types. Radar simulators, covering area and approach control with additional procedural, or non-radar, capabilities. Tower simulators covering visual and ground control as well as radar operations. Tower simulators can have three dimensional (3D), out-of-the-window ‘photographic’ visuals or 2D stylised situation displays.
Modern simulators cover all levels and types of training using the same, high fidelity model of the ATC world. The required levels of training are met by the instructors tailoring the exercises and airspaces.
Radar Simulator: The student controller has a simulation of the radar situation display with the usual display management tools. The task environment is simulated complete with audio communications, flight progress strips and ancillary information displays.
3D Tower Simulator: A multi-channel 3D visual system is used to provide a realistic out-of-the-cab view for the trainee tower controllers, instructors and assistants. The scope of the simulation includes all of the ground environment, the visual scene and the performance of the aircraft in 6 degrees of freedom. The simulation of the working environment includes special tools such as lighting control panels, ground movement and short-range approach radar as well as the audio communications, information displays and flight progress strips.
Multi-Role & Part Task Trainers: In order to achieve training throughput and to reduce costs multi-role trainers, part task trainers and classroom systems are used as part of the training media progression. Classroom and part task trainers are typically PC workstations networked on desks with an instructor who can join and manage any of the simultaneous running exercises. The workstations can include: radar displays; audio communications; flight strip printing; voice recognition; information and 2D situation displays.
Multi-role trainers cover almost the whole task of radar and tower control but use 2D displays to represent the airborne and ground situation seen from the tower. Typically a large screen is used for the airborne situation to encourage the student to look up in order to acquire the information needed for carrying out visual control.
The ground situation is usually presented on a monitor in the console giving a controllable, stylised, plan view. These trainers are powerful and versatile. They allow the students to cover a large part of their training both generic and specific before progressing into the more expensive 3D tower and advanced radar simulators.
The concept is the same as that used for flight training where pilots progress from part task trainers through training devices into the full scope flight simulators.
Military Variants: Military air traffic controllers are trained using the same range of devices but with some additional variations. Special military aircraft performance characteristics are covered as well as unique military procedures and special airfield equipment such as Precision Approach Radar and arrestor barriers. The users can manage their own sensitive data.
Separate military specialisations such as fighter controller training are beginning to benefit from the developments made on the civil side. The new generation of fighter controller training simulators recently ordered by the UK MoD benefit from advances in PC technology, voice recognition and new levels of automation and user software tools.
The Application Of ATC Simulators
Selection: Some simple simulators are used for aptitude testing. They are combined with a battery of interviews and psychometric tests in an attempt to improve the overall success rate of training and to select for those elusive characteristics that make a successful air traffic controller.
Ab-initio/Rating training: Rating training combines theoretical teaching with practical instruction. Part-task and full-task simulators for both radar and tower training are widely used in the national and commercial training colleges where this level of training is provided. The simulations are generic and tend to use synthetic airspaces designed to cover all of the teaching points for the courses. Having achieved their rating in, for example, tower or radar approach control the students then move on for validation training at a particular ATC unit.
Validation Training: Most validation training is carried out on the job with all the inevitable problems and limitations. The restrictions are becoming worse as units get busier, operations become more intense and the airlines are less willing to suffer the consequences of reduced services. This is an area that would benefit from the greater use of specific simulation. The current situation is in contrast to pilot training where simulators are widely used for validation training.
Refresher Training: Air traffic controllers are checked every year for their ability to handle unusual situations and emergencies. They also receive refresher training. Again, this is an area that would benefit from more use of simulators. The current practice is in stark contrast to pilot training where simulators are the main medium for refresher training and regular checks.
Conversion Training: Conversion training to achieve validation at a new unit or to handle new equipment or procedures sometimes benefits from the use of simulators. The manufacturers of new ATC systems often provide CBT and embedded simulators for conversion training. But these tend to be focused on the use of the equipment and are not convenient for operational training. Conversion training would benefit from the wider use of simulators and is in contrast with pilot training where simulators are the main method used.
Design & Evaluation: Specialist fast time simulation tools are used to investigate new airports, new airspace structures and operating procedures. However, at some point these new concepts have to be tested in real time with controllers and pilots in the loop. Subsequently when the new design is fixed, the real time simulators can further be used for conversion training prior to the implementation of the new facilities. There have been some notable successes of this approach, for example, at the new airport at Kuala Lumpur. Here the combined tower and radar simulator was delivered ahead of the construction. It was used for teaching the ground vehicle drivers and testing the planned procedures as well as training the controllers. Design and evaluation is an area, which would benefit from the wider and more planned use of simulators.
Characteristics
The new generation of ATC simulators have a number of common characteristics. They are PC based and use commercial off-the-shelf hardware with Windows operating systems and networking. The user interfaces are Windows based and intuitive. The simulators are designed for technology insertion to take advantage of commercial developments in PC products. They are data driven, versatile and flexible. They support a wide range of displays for different training applications and have good facilities for importing and exporting data and for interacting with third party systems. They cover all levels and types of training in an integrated environment including audio communications and voice recognition.
Design
The simulators are based on a network of PC workstations. The software allows any station to play any role. But the hardware fit and installation layout will usually predicate the roles, such as controller, instructor, assistant, pseudo-pilot or data preparation.
The controller stations have multiple displays; some with touch input, and sound cards for the audio communications and lighting control panels. For the 3D tower simulators there is a gateway interface to the image generators, which in turn drive the visual displays.
The simulations are data driven. The user has comprehensive tools for customising easily their own: air spaces; exercises; weather conditions; maps; aircraft performance; voice recognition; audio communications and both 2D and 3D models. The tools have interactive, graphical interfaces with digitiser input for creating maps, ATM structures and airfield layouts.
Multiple exercises can be run simultaneously with individual recording and replay. The configuration of the multiple simulation circuits and the user displays can be selected at run time. As in the real world, the simulator allows the controllers to view the aircraft and the environment in different ways: as radar situations, visual scenes and various data presentations.
All the simulation data is held in a relational database that incorporates a flight plan library. Traffic samples and flight plan groups can be recalled into an exercise with a single entry. The simulation network communications use readable, ATC English text messages. This greatly simplifies: recording; scripting; bug fixing; customer help and interfacing with other systems and sources of data. Aircraft commands are passed around the network to synchronise the distributed workstations and the multiple exercises. The commands are recorded for replay, analysis and recovery purposes. They are also used as a source for the script editor so that scripts and command macros can be produced from the results of a development run.
The powerful scripting system supports wild card and conditional actions. It increases the realism of the simulations without cluttering up data entry and pilot commands. It allows adjacent sectors to be automated intelligently saving staff costs.
The simulation engine provides the links for the man-machine interfaces (MMIs). A wide range of different displays can be run at the same time. New MMIs can be created and added to the library using an open specification available to the user.
Real-world and simulation data can be imported into the database via tailored parsers. The user has access to the relational database so that he can make global changes, carry out queries and produce special reports.
All the exercise run is recorded. If the run is interrupted the recording enables the system to recover quickly and accurately.
Commercial and Operational Pressures
Unlike pilot training there is no undeniable commercial imperative to use simulators for air traffic control training. Training can be and often is carried out on the job. Air traffic control training and simulation budgets are small. The prices of air traffic control simulators are significantly less than their equivalent flight counterparts. An average airline spends more on simulators for its own pilot training than its national ATC provider does for all of its air traffic controllers. ATC simulator buyers are less experienced and less practised than their airline counterparts. However, air traffic controllers are responsible for many aircraft and their passengers. They continuously handle situations with high profile, high-risk consequences.
Stimulation
The use of simulators to drive, or stimulate, real ATC radar systems is growing. This approach gives maximum realism with full control over the environment and training progression. The student has on-console training with the ultimate fidelity of man machine interfaces, tools and workstation layout. Upgrades to the real system can be integrated into the simulator as they happen. The design is also good for the testing and evaluation of new upgrades and concepts. Spare workstations can be used and will still be available for emergency use. The interfaces between the simulator and the real equipment are typically customised and use ATC protocols rather than simulation standards such as HLA or DIS. The protocols cover radar data, flight data, general information and data links. The simulators can import data for the real maps, traffic samples and operational situations. They can export data for assessment and evaluation work.
Links to Flight Simulators
With the DFS and Lufthansa in Frankfurt we pioneered the linking of our FIRST/DESIM radar simulators with full flight simulators. The project gave very good results with benefits for both the pilot and the air traffic controllers. The pilot became exposed to the simulated ATC audio world with the delays, actions and restrictions caused by the controller and all the other traffic. The aircraft became part of the simulated ATC radar environment and the pilot’s problems, tasks and distractions added to the realism and workload of the controller. The extent and sophistication of the integration of the two systems can be extended. The aircraft appears in the simulated radar world. The pilot receives full audio communication with the ATC world on several channels. The aircraft’s TCAS can be stimulated with the other aircraft in the ATC world. The aircraft visual and the ATC displays, including tower visuals, can share the same correct view of their combined worlds. The aircraft and those in the ATC world can be made to react to each other’s actions.
The benefits of providing a full ATC audio environment for the pilot have already been recognised. Pilot radio communications are a significant part of his real task but a relatively small part of his simulator training. The pilot experiences greater realism and a higher workload by having to monitor and handle the radio traffic. Asking the instructor to carry out the ATC transactions increases his workload and has a negative impact on his prime tasks. There is a cost associated with adding the ATC environment and there will be a need for greater automation if it is to be affordable. The technology is achievable. The question is what price will the users pay for the different levels of added realism.
Synthetic Pilots
While pilots may be looking for a synthetic ATC environment, the creation of synthetic pilots is the holy grail of ATC training. ATC training is labour intensive with two or maybe more pseudo-pilots servicing every student. Removing or even reducing the use of pseudo-pilots will give the training organisations a continuing cost saving. Interest in voice recognition and frustrated trials have been going on for many years.Investments generated by other markets are leading to improvements. ATC has the advantages of a very limited vocabulary and rigid syntax. It has the disadvantages of needing a very high accuracy and real time operation with rapid transactions. Third party voice recognition engines are used. ATC simulators are being designed with standard (SAPI) interfaces for technology insertion as the investments generate improvements. Voice recognition is being used - particularly for basic training where the reinforcement of good pronunciation and correct phraseology is a goal. At higher levels of training where variable phraseology creeps in and more complex structures are required voice recognition has a role to play but has to be used more carefully and considerately. The systems being used tend to be software based, speaker independent and run on standard PC hardware. Special ATC user tools are provided for customising user accents, real names, phraseology and levels of expertise.
The ATC simulators provide all of the correct pilot radio prompts. In the synthetic pilot environment these are fed to the student via text-to-speech voice synthesis software. Multiple male and female voices with different accents are used. Each aircraft has a pilot with his own personality of voice and phraseology idiosyncrasies. There is a cost associated with improving the realism of the voices and providing multiple voices. In our experience the students, in the heat of training, soon ignore the computer quality of the lower cost text-to-speech outputs and welcome the realism of having individual identifiable voices for each aircraft.
As in real life the accuracy of voice recognition is not 100%. It is important that the simulator is able to resolve these situations with the student through correct, credible conversations. For example, the simulator can handle “negative” commands, cancelling the last instruction and inserting the next instruction provided.
The new levels of automation can be combined with the synthetic pilots and the audio communication system to simulate telephone calls and conversations with other agencies. Clearly, the scope of the conversation, having moved outside the bounds of R/T phraseology, is limited and needs careful design. But it adds to the realism and scale of training without the cost of human pseudo-pilots.
Standards: Attractions and Difficulties
There are no standards for the ATC simulators themselves. The training courses including the simulators are examined and approved to ICAO and local national standards. There is a high degree of subjective judgement in the acceptance of ATC simulators. The user has to make judgements when selecting a simulator in the expectation of getting subsequent approval for the course in which it will be used. In particular 3D visual displays are specified and selected subjectively. There have been attempts in the past and there are new moves to introduce standards. Objective standards will be attractive to the reputable manufacturers. But there are concerns that the imposition of standards will increase the costs for individual units. The specification and acceptance of 3D visual performance seems likely to remain a problem. The customary gap between user expectations and delivery is being addressed by demonstrating the choices up front and showing what you get for your money across the step boundaries in technology and costs.
Conclusions
ATC simulation is a niche market with relatively low budgets. However, ATC is a vital service with high profile, high cost risk potentials. Simulation technology has advanced in availability and performance and continues to progress. Key current areas include automation, voice recognition, visual systems and user tools. Integration with flight simulators has been demonstrated. The potential for wider interaction exists, the benefits are recognised and the technology is available. However, the commercial demand is in question. Given the commercial incentives the technology can be delivered and the benefits realised.