Artificial Intelligence for making rapid human-level decisions.

  Recent Research   Previous Projects
Riverview AI demonstrates software capable of making optimal tactical decisions in less than 10 seconds.

Funded by a DARPA research grant (W911NF-11-200024) Riverview Artificial Intelligence, LLC, has successfully demonstrated that the MATE (Machine Analysis of Tactical Environments) software program is capable of analyzing actual real-world tactical situations and displaying an optimal solution in less than 10 seconds.

The demonstration incorporated the actual events of the Battle for Marjah, February 13, 2010 as shown in the HBO documentary of the same name. MATE was given the same tactical situation that was presented to the ground commander in the following clip.



Figure 1: Screen capture of MATE running the Marjah scenario. Blue and Red units were added by simply clicking on the screen. The background map was downloaded from Google Earth.

After clicking on the ‘Calculate AI’ icon, the ‘Analyze and Classify Current Situation’ button and the, ‘Generate HTML Output and Launch Browser’ button, MATE’s analysis of the tactical situation was displayed. Total elapsed time was less than 10 seconds.

MATE then automatically generated HTML pages of its recommendations including graphically displaying optimal paths for an envelopment maneuver that encircled enemy positions.

Figure 2: MATE’s suggested Course of Action (COA) was an Envelopment Maneuver that outflanked enemy positions.

To see the actual HTML output of MATE’s analysis, click here:

In addition to performing tactical analysis including terrain analysis and enemy position analysis MATE is also capable of performing the five canonical offensive maneuvers (Infiltration, Penetration, Envelopment, Turning and Frontal Assault) and displaying optimal COAs.

Research Project

MATE is a continuation of Dr. D. Ezra Sidran’s TIGER (Tactical Inference GenERator) which was funded by a DARPA seedling grant.

For more information about MATE contact

MOG93: A Case Study of a New Approach to a Web-based MOUT Simulation

We introduce here a new method for modeling a web-based MOUT simulation in which the Opposition Forces (OPFOR) are represented not by units, or agents, but by the city itself in which the operations are carried out. As more units are introduced into the city the hostility level of the city increases. As Blue Force units move through hostile neighborhoods in an attempt to secure and return with a High Value Target they are subject to casualties. Other factors, such as time of day in relation to khat consumption, also affect Blue Force casualties.

Simulation Project

Figure 1. Starting positions for MOG93. Screenshot from www.MOG93.com

1. Introduction

The need for MOUT (defined as either Military Operations on Urban Terrain or Military Operations on Urbanized Ter-rain) [1, 2] simulations has increased significantly over the past several years [3]. Furthermore, the public interest in MOUT has been piqued first by the publication of Black Hawk Down [4] and later by the release of the movie based on the book. We were approached by a commercial web site, Wargamer.com, just prior to the release of the movie to create a web-based simulation of the events in Moga-dishu October 3, 1993. We had an extremely short devel-opment cycle (only six weeks) and no budget to speak of. Because of these limitations, as well as a desire to approach the problem of MOUT simulations differently, we opted for a new model in which the entire city of Mogadishu was OPFOR and only Blue Forces (U.S.) were displayed and movable by the user. This differs from most MOUT simulations that rely on modified first person shooter (FPS) en-gines such as Quake, Unreal and Half-Life [5, 6].


Read / Download rest of paper here: MOG93.pdf


Copyright , Riverview AI – All Rights Reserved