Charles River Analytics developed a Compressed Parallel Architecture for AR/VR Training in Data Constrained Terrain (COMPACT) for the US Army. COMPACT is designed to support the Army’s Synthetic Training Environment (STE), which uses augmented and virtual reality (AR/VR) to train soldiers at their point of need anywhere in the world.
Training must be provided in remote locations, such as forward operating bases far from training headquarters with only satellite internet access. Training an individual in these potentially low-bandwidth environments is challenging, and the Army must scale to thousands of soldiers training simultaneously.
The compressed, parallel COMPACT architecture uses dynamic compression algorithms that ensure XR data flow during training meets bandwidth constraints. The architecture also applies data prioritization heuristics derived from human training needs, along with physiological constraints, to minimize the latency experienced by soldiers in point-of-need environments. This makes high-bandwidth training possible in low-bandwidth environments.
Charles River designed a benchmarking framework that demonstrates the performance of a COMPACT’s data management techniques. This tool uses Charles River’s proven MultiMAST modeling system, which can support over 15 million simulated entities, to evaluate COMPACT in reality-scale scenarios.
This material is based upon work supported by the U.S. Army Small Business Innovation Research Program Office and the U.S. Army Futures Command under Contract No. W911NF-20-P-0051. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the U.S. Army Small Business Innovation Research Program Office or the U.S. Army Futures Command.