The Importance of Fidelity in Simulation and Training

Fidelity in Simulation

Fidelity in simulation has traditionally been defined as ‘the degree to which the simulator replicates reality’. Using this definition, simulators are labelled as either ‘low’ or ‘high’ fidelity depending on how closely they represent ‘real life’. Ideally one supposes, every simulation would be high fidelity and would replicate as precisely as possible the real world application, but to accept that premise is to accept that procurement budgets are unlimited and that is of course not realistic. The realistic view is that fidelity is almost always a trade-off between the degree of fidelity required and the cost of procurement and through life support.

So how do we determine what level of fidelity is sufficient? Let’s look at a few examples:

Let’s think for a second about a videogame for the commercial market. High fidelity in the videogame space is generally accepted as being defined by three key elements being the audio and visual arts (graphics), the gameplay and the story. The games that best combine these elements into a compelling and immersive experience for users are the ones that are the most highly successful products.

The importance of fidelity in videogames is well articulated by this statement: “For me, all the elements of a video game (gameplay, audio visual arts, story-telling) should come together to create something greater than any one of those single elements. For the perfect game, the whole is greater than the sum of its parts; and I think that point is reached when total immersion is achieved”. IGN Community Member, FrenchyV23, 2016.

I’m fairly confident that ‘FrenchyV23’ has a much better grip on the factors that make a high-fidelity offering in the videogame space than I do. It’s not hard however to imagine the consequences of a poorly designed and implemented videogame. It doesn’t sell and is ultimately unsuccessful. Poor sales will affect the reputation of the company and threaten future sales.

Now let’s consider an aircraft simulator. One can safely assume that if pilot training is going to take place in a simulator, then high fidelity is an essential requirement. The consequences of poor decisions in this environment can range from inconsequential to catastrophic in the real world and thus can’t be measured solely in monetary terms.

I’m certain that the many thousands of us who regularly take to the sky in commercial airliners would expect that any training done in simulation is training that represent real life situations as closely as possible. This is particularly true, I think, in terms of training emergency procedures. Thankfully, emergencies on commercial aircraft are few and far between (the reason that hardly anyone actually listens to a safety brief) and yet when they occur, pilots and crews must be able to react correctly and instinctively to correct the problem. Acting instinctively incorrectly is not at all desirable and yet if training for the event has been conducted in a low fidelity environment (one that does not replicate the real controls, the real outcomes to a high degree), this is a likely outcome.

Now let’s address the military simulation environment, which is, of course, my main interest and where Saab and other simulation solutions are used frequently by forces around the world to train for combat. Note: I’m not focused here on training the military staff that makes plans and command forces to execute operations. I’m instead concentrating on soldiers in the field conducting tactical combat training.

In the combat environment, training with live ammunition is necessarily one-sided and resource-intensive in terms of safety supervision, so substituting lasers and laser detectors for live ammunition enables realistic, high precision gunnery and force on force training. By tracking, monitoring and recording the actions of crews, the ideal conditions for effective training can be achieved. Undoubtedly, in this often high stress environment the fidelity of a simulator can have a significant effect on the end user’s understanding of the capabilities of the real world equipment being simulated, and will certainly affect performance in real life situations.

In order for a simulation to be effective in in combat training, it is absolutely fundamental that the simulation precisely replicates the ballistics of the ammunition being employed. This includes accurate replication of maximum effective ranges, ammunition path and time of flight, minimum arming distances for certain ammunitions, and end target effects. It is also critical that the simulation does not introduce any procedures (drills) that would not be relevant in the real world equipment. For combat training, low fidelity systems provide unrealistic results which many feel could have the effect of leading crews to believe they are well prepared for combat when in fact they are not. I tend to have more faith in the intelligence and real-world experience of our soldiers, so I believe instead that when they see unrealistic results they quickly lose focus and lose confidence in the value of their training.

As I mentioned earlier in the article, there is generally a trade-off between the degree of fidelity required and the cost of procurement. When discussing cost, one normally thinks of a monetary value; however cost can also come in the form of human life especially when the lack of fidelity in simulation adversely affects training standards and real world performance.

High fidelity simulation can be costly and customers can generally procure more low fidelity equipment for a comparable budget. There may well be an application for low fidelity solutions where the outcome of less than real world replication is not potentially catastrophic, but that application shouldn’t be in the domain of combat training.