Unlike war-themed video games, simulating a real-world battlefield means creating and manipulating huge quantities of data that represent complex, dynamic scenarios comprising hundreds of thousands of individual objects. Such objects—called entities—may include soldiers, sensor systems, weaponry, aircraft, munitions, vehicles, and equipment. Battlefield simulations must also account for scenario-environment conditions such as weather, civilian populations and their movements, air and sea traffic, and military assets outside the immediate battlefield.

Combining all of these requirements into realistic, high-resolution, graphics-based training tools is the mission of Livermore’s Conflict Simulation Laboratory. Led by Computing staff, the Conflict Simulation Laboratory has provided conflict-simulation software tools to military and civilian customers around the globe. “Our U.S. and international customers need realistic simulations in real time with real data so they can better understand and predict the ground truth,” explains Mark Piscotty, group leader for Simulation, Analysis, and Software—a team that supports and collaborates with the Lab's Global Security directorate and mission space.

One key tool developed for this purpose is the Joint Conflict and Tactical Simulation (JCATS) application, which accurately replicates entity-level operations scenarios for military tactical training and analysis. Originally designed for the U.S. Joint Chiefs of Staff J7 Division, JCATS is used to analyze civilian site security, emergency management scenarios, and terrorist attack threat vectors for Department of Homeland Security programs. It is one of the most widely used tactical simulations in the world and is installed in hundreds of U.S. military and civilian organizations, in NATO, and in more than 30 countries.

Feature-Rich Scenarios

In a JCATS simulation, human operators role-play friendly and enemy battlefield operations, controlling their entities via a graphical user interface on networked individual client workstations. The results are reported to higher level commanders and staff using actual battlefield command-and-control systems. From the bottom up, JCATS portrays both ground truth and perception that occur in response to tactical decisions.

JCATS owes much of its widespread adoption to its level of human-in-the-loop control. Rather than relying too heavily on automation, JCATS allows the operator to intervene, controlling automated behaviors to achieve the scenario’s training objectives. In addition, the application operates on low-cost, off-the-shelf hardware that allows for easy scalability. Piscotty, who serves as JCATS program lead, points out, “JCATS can run on a single workstation or a few networked laptops. It also supports large-scale operations, and we’ve tested it on as many as 300 workstations with a single server.”

JCATS closely portrays actual scenarios by coupling realistic algorithms and data for acquisition and targeting, along with feature-rich terrain such as vegetation and elevation contours. The application also lets users control, and interact with, a wide range of physical details. For example, buildings can be rendered as simple 3D shells or enhanced with interiors including multiple stories, walls, windows, stairs, and lockable doors. Users can explore floor plans or the building’s roof. Even ambient and artificial lighting can be simulated: overcast conditions, day or night, moon or no moon, lights inside a building, and light levels inside a tunnel.

Performance data for sensor systems, weapons, munitions, and vehicles (including ground, sea, and air systems) are user-defined, and can be customized to users’ chosen level of complexity to train or rehearse any tactical scenario—from a handful of entities up to hundreds of thousands of entities at the joint-task-force level. JCATS can also be used to simulate “what if” scenarios—for instance, “What if a tank could go faster or shoot farther?”—to see how they affect the simulated battle.

JCATS runs on v7.9 of the Red Hat Enterprise Linux operating system. In January 2022, the team released v17.0 with enhanced features for casualty and damage mappings as well as artillery and planning controls. The release also included airport runway data, which users can import into their terrains via a .csv file that contains latitude and longitude; runway length and width (in feet); and the heading (in degrees). “We regularly hold user conferences to gather feedback from our customers and learn more about their evolving use cases,” says Piscotty.

Realistic Training Environments

The JCATS Low Overhead Driver (JLOD) application adds another layer of realism to JCATS-based scenarios by simulating entities and activities that occur outside, but still affect, the JCATS playbox. A JCATS playbox is a high-resolution terrain up to 24x24 degrees of latitude and longitude, whereas a JLOD playbox incorporates the entire Earth.

Piscotty explains, “JLOD includes wrap-around data like global military assets and commercial air traffic and ships. For example, JLOD can simulate the realistic transit of B-2 bombers from a U.S. base to a JCATS scenario taking place across the globe.” JLOD also simulates population behavior and realistic patterns of life that may affect the JCATS exercises—from mundane activities like civilian traffic patterns to dramatic events such as mass refugee movement or disaster evacuation.

Moreover, JCATS and JLOD can connect to each other and to other simulation programs—such as tank simulators and 3D virtual systems—used by customers both domestic and abroad. The applications can operate in a joint training environment, interacting with users and systems while displaying locations and actions on JCATS and JLOD screens. All branches of the U.S. armed forces, NATO, and other allied nations leverage this federation capability. “The level of detail achieved by combining JCATS and JLOD is unprecedented, and our customers appreciate the improved realism that comes with using the applications together,” states Piscotty.

Learn more about JCATS and contact the team at csl.llnl.gov/jcats.