VRAI Announces Their HEAT Simulation Data Product Now Available for VBS as well as for Unreal Engine.

VRAI have announced that their HEAT simulation data product is now available to integrate with both VBS and Unreal Engine. VRAI believes human performance data is currently the untapped resource in training and have built HEAT to transform this resource into actionable insights.

Having previously developed integrations for HEAT with both Prepared 3D and Unity, this latest announcement allows a larger community of simulation creators to leverage HEAT. It is a significant milestone in enabling world leaders in the simulation market to capture, store, analyse and visualise simulation data.

Speaking on the announcement, Niall Campion, VRAI Managing Director for Customer & Product said:

“Our mission is to redefine exceptional human performance through better training, unlocking every individual’s full potential in order to save lives. In order to harness the power of data & AI in simulation training, the first step is to start capturing data. We have developed HEAT to enable our customers to start their data journey, unlocking actionable insights that can deliver higher performance in training, increase training speed, reduce training cost or a combination of all three.”

VRAI believes that simulation, as well as being a great technology for delivering training at scale, can also be used to capture, store, analyse and understand human performance to enable this transformation of how people are trained.

“We see the data generated within VBS as a rich source of potential insight into human performance, that can enable objective assessment, enhanced evaluation and, when combined with AI, provide the ability to predict performance,” said Oliver Arup, Chief Product Officer at Bohemia Interactive Simulations. “VRAI are a leading SME working in the area of simulation data exploitation and we are excited that their cutting-edge HEAT product is now available to integrate with VBS, to give more options to our customers, and provide even deeper insight into their personnel’s performance”.

VRAI’s vision is to transform the way people are trained by leading the widespread adoption of data driven insights in simulation training in order to improve human performance.

They envision a future where simulation is within arms reach of anyone who needs it, and individuals can access personalised, adaptive learning environments, so that they can develop the skills they need to flourish.

HEAT has been developed to make this vision a reality. HEAT is designed and capable of ingesting very large datasets (capable of capturing up to 25 million data points per hour of training) that includes data relating to the individual trainee drawn from their user profiles, data from the sim such as those mentioned above, and data relating to the physiological impact the training scenario is having on the individual and crews. These sessions can be performed across the Live, Virtual and Constructive (LVC) training spectrum. Any training device capable of generating data can provide an input to HEAT.

HEAT can be integrated into a project in a number of ways, but primarily through a HEAT API which allows 3rd party developers to integrate it into their simulators.

HEAT can be broken into four (4) core functions:

• Capture - Individual user profiles allow personalised data to be captured across any training environment
• Store - Approx 25 million data points per hour are stored securely in a standardised format.
• Analyse - Machine learning/data analysis is deployed across data sets to generate insights.
• Visualise - Actionable insights are displayed on the dashboard at individual, instructor, and management levels.

For the user, it provides seamless integration into their existing training pipeline. An interface allows users to log into their training sessions. This facilitates individualised data capture. Data is stored securely in a relational database, is always owned by the end users and in a way that is GDPR compliant. When sessions are complete, users or instructors can log into their dashboard via any web browser - desktop or mobile - and view their results.

Metrics can be customised to organisational training requirements.

HEAT also has the ability to be utilised to develop and deploy high value machine learning algorithms. Building machine learning models utilising HEAT is undertaken in four phases:

• Phase 1 - Determine requirements. The VRAI Data Science team works with the customer to identify what predictions would add the greatest value and what data is required to be captured to make those predictions.
• Phase 2 - Capture data. The necessary data is captured from the training simulations. This dataset must be sufficiently large enough to ensure a high performing model can be generated.
• Phase 3 - Build model. New models are developed using modern data science techniques. The models are then performance tested and optimised.
• Phase 4 - Deploy model. The models are deployed to HEAT allowing for greater insights to be generated by training Instructors with the ability to now predict future performance of their students. As these are machine learning models, as the database grows, the performance of the models also improves.

For more information on the HEAT product, VRAI will be exhibiting at the upcoming I/ITSEC Conference in Orlando Florida from 27 November to 1 December at booth #772, or contact them via their website https://vraisimulation.com/.