Antietam & AI

MATE AI selected Objectives for Blue, 3D Line of Sight (3DLOS) and Range of Influence (ROI) displayed for the Antietam: Dawn General Staff scenario. Screen shot from General Staff Sand Box. Click to enlarge.

The author walking across Burnside’s Bridge in 1966 (age 12).

I have been thinking about creating an artificial intelligence (AI) that could make good tactical decisions for the battle of Antietam (September 17, 1862, Sharpsburg, Maryland) for over fifty years. At the time there was little thought of computers playing wargames.1)However, it is important to note that Arthur Samuel had begun research in 1959 into a computer program that could play checkers. See. “Samuel, Arthur L. (1959). “Some Studies in Machine Learning Using the Game of Checkers”. IBM Journal of Research and Development.” What I was envisioning was a board wargame with some sort of look-up tables and coffee grinder slide rules that properly configured (not sure how, actually) would display what we now call a Course of Action (COA), or a set of tactical orders. I didn’t get too far on that project but I did create an Antietam board wargame when I was 13 though it was hardly capable of solitaire play.

The Antietam scenario from The War College (1992). This featured 128 pre-rendered 3D views generated from USGS Digital Elevation Model Maps.

In 1992 I created my first wargame with an Antietam scenario: The War College (above). It used a scripted AI that isn’t worth talking about. However, in 2003 when I began my doctoral research into tactical AI I had the firm goal in my mind of creating software that could ‘understand‘ the battle of Antietam.

TIGER Analysis of the battle of Antietam showing Range of Influence of both armies, battle lines and RED’s avenue of retreat. TIGER screen shot. Appears in doctoral thesis, “TIGER: A Machine Learning Tactical Inference Generator,” University of Iowa 2009

The TIGER program met that goal (the definition of ‘understand’ being: performing a tactical analysis that is statistically indistinguishable from a tactical analysis performed by 25 subject matter experts; e.g.. active duty command officers, professors of tactics at military institutes, etc.).

In the above screen shot we get a snapshot of how TIGER sees the battlefield. The darker the color the greater the firepower that one side or the other can train on that area. Also shown in the above screen shot is that RED has a very restricted Avenue of Retreat; the entire Confederate army would have to get across the Potomac using only one ford (that’s the red line tracing the road net to the Potomac).  Note how overlapping ROIs cancel each other out. In my research I discovered that ROIs are very important for determining how battles are described. For example, some terms to describe tactical positions include:

  • Restricted Avenue of Attack
  • Restricted Avenue of Retreat
  • Anchored Flanks
  • Unanchored Flanks
  • Interior Lines
  • No Interior Lines

A Predicate Statement list generated by MATE for the battle of Antietam.

Between the time that I received my doctorate in computer science for this research and the time I became a Principal Investigator for DARPA on this project the name changed from TIGER to MATE (Machine Analysis of Tactical Environments) because DARPA already had a project named TIGER. MATE expanded on the TIGER AI research and added the concept of Predicate Statements. Each statement is a fact ascertained by the AI about the tactical situation on that battlefield. The most important statements appear in bold.

The key facts about the tactical situation at Antietam that MATE recognized were:

  • REDFOR’s flanks are anchored. There’s no point in attempting to turn the Confederate flanks because it can’t be done.
  • REDFOR has interior lines. Interior lines are in important tactical advantage. It allows Red to quickly shift troops from one side of the battlefield to the other while the attacker, Blue, has a much greater distance to travel.
  • REDFOR’s avenue of retreat is severely restricted. If Blue can capture the area that Red must traverse in a retreat, the entire Red army could be captured if defeated. Lee certainly was aware of this during the battle.
  • BLUEFOR’s avenue of attack is not restricted. Even though the Blue forces had two bridges (Middle Bridge and Burnside’s Bridge) before them, MATE determined that Blue had the option of a wide maneuver to the north and then west to attack Red (see below screen shot):

MATE analysis shows that Blue units are not restricted to just the two bridge crossings to attack Red. MATE screen shot.

  • BLUEFOR has the superior force. The Union army was certainly larger in men and materiel at Antietam.
  • BLUEFOR is attacking across level ground. Blue is not looking at storming a ridge like at the battle of Fredericksburg.

MATE AI selects these objectives for Blue’s attack. General Staff Sand Box screen shot. Click to enlarge.

We now come to General Staff which uses the MATE AI. General Staff clearly has a much higher resolution than the original TIGER program (1155 x 805 terrain / elevation data points versus 102 x 66, or approximately 138 times the resolution / detail). In the above screen shot the AI has selected five Objectives for Blue. I’ve added the concept of a ‘battle group’ – units that share a contiguous battle line – which in this case works out as one or two corps. Each battle group has been assigned an objective. How each battle group achieves its objective is determined by research that I did earlier on offensive tactical maneuvers 2)See, “Implementing the Five Canonical Offensive Maneuvers in a CGF Environment.” link to paper.

As always, I appreciate comments and questions. Please feel free to email me directly with either.

References

References
1 However, it is important to note that Arthur Samuel had begun research in 1959 into a computer program that could play checkers. See. “Samuel, Arthur L. (1959). “Some Studies in Machine Learning Using the Game of Checkers”. IBM Journal of Research and Development.”
2 See, “Implementing the Five Canonical Offensive Maneuvers in a CGF Environment.” link to paper.

Thank You Wargaming Community!

Last week I posted an appeal to the wargaming community and backers of The General Staff Wargaming System that I needed more maps and Orders of Battle (OOBs). The General Staff Wargaming System is designed to handle any conflict in the Black Powder era and the machine learning AI needs as much input as possible.

The five layers that make up a General Staff map.

The General Staff Army Editor makes it pretty easy to create four of the five layers of a map file (see above). The problem is the beautiful background image that the user sees on screen (the computer AI couldn’t care less about the visual map). I’ve been able to locate a lot of great maps; especially from the American Civil War and the US Library of Congress but we still need more.

Waterloo from Glenn Drover (Forbidden Games) and Jared Blando. Click to enlarge.

A couple of days ago I received an email from the famous game designer, Glenn Drover (Forbidden Games), who offered us the use of three maps that he had researched and were drawn by artist Jared Blando. Here’s a link to Forbidden Games’ site. Please check out their fantastic board games!

Ligny from Glenn Drover (Forbidden Games) and Jared Blando. Click to enlarge.

The three battlefield maps were Waterloo, Ligny and Quatre Bras.

Quatre Bras from Glenn Drover (Forbidden Games) and Jared Blando. Click to enlarge.

What is especially amazing is how well these three maps fit the style that I’ve wanted to create for General Staff.

In addition to these three great maps, which we will definitely be using for the battles of Waterloo, Ligny and Quatre Bras, I’ve received emails from a number of other wargamers who have offered to research OOBs; especially some in another language.

I am completely blown away (I know it’s a cliche, but I don’t have any other words) by the kindness and generosity shown me by the wargaming community. Thank you very much!

AI Routines Added But We Need More Testing Data (Maps & Armies)

The AI routines for calculating battle lines and range of influence have been ported over from the original C++ code to C#:

Antietam displaying Range of Influence and Battle Lines. Click to enlarge.

I’ve written a number of blog posts about these AI routines which you may find interesting:

Battle Lines, Commanders & Computers

What’s Wrong With This Picture?

Wargame AI Continued: Range of Influence

That’s the good news. The bad news is that I’m also installing the Machine Learning AI that was the basis of my doctoral research and it needs more battles to learn from. A lot more. Currently there are 15 armies (click here) and 5 maps (click here). Ideally I would like about 50 armies and 30 maps used to create 30+ battle scenarios.

Are you a cartographer or a researcher?  If you are, and you’re interested, I could use your help if you would like to volunteer. All the maps and armies were created using the tools that you, as a backer, have already been provided: The General Staff Army Editor and The General Staff Map Editor. A little bit of PhotoShop or another paint program was used to clean up the old maps and a free program, Inkscape, was used to create the paths for roads and rivers. The most difficult task is the research. Finding Order of Battle Tables (OOBs) are pretty easy but General Staff requires knowing the actual troop strength of every unit. Sometimes, that is very hard to find. For the maps, adding elevation is usually the most difficult bit, but there are a number of built-in tools to make this easier.

If you’re interested in helping add to the data files please contact me directly: Ezra@RiverviewAI.com.

Wargame AI Continued: Range of Influence

In two previous blogs I wrote about how Artificial Intelligence (AI) for wargames perceive battle lines and terrain and elevation. Today the topic is how computer AI has changed ‘Range of Influence’  (ROI) or ‘Zone of Control’ (ZOC) analysis. Range of Influence  and Zone of Control are terms that can be used interchangeably. Basically, what they mean is, “how far can this unit project its power.”

One of the first appearances of range as a wargame variable was in Livermore’s 1882 American Kriegsspiel: A Game for Practicing the Art of War Upon a Topographical Map (superb article on American Kriegsspiel here).  Note that incorporated into the ‘range ruler’ (below) is also a linear ‘effectiveness scale’.

Detail of Plate IV, “The Firing Board,” from the American Kriegsspil showing a ruler for artillery range printed on the top. Note the accuracy declines (apparently linearly) proportional to the distance. Click to enlarge.

The introduction of hexagon wargames (first at RAND and then later by Roberts at Avalon Hill; see here) created the now familiar 6 hexagon ‘ring’ for a Zone of Control:

Zone of Control explained in the Avalon Hill Waterloo (1962) manual. Author’s Collection.

I seem to remember an Avalon Hill game where artillery had a 2 hex range; but I may well be mistaken.

Ever since the first computer wargames that I wrote back in the ’80s I have earnestly tried to make the simulations as accurate as possible by including every reasonable variable. With the General Staff Wargaming System we’ve added two new variables to ROI: 3D Line of Sight and an Accuracy curve.

Order of Battle for Antietam showing Hamilton’s battery being edited. Screen shot from the General Staff Army Editor. Click to enlarge.

In the above image we are editing a Confederate battery in Longstreet’s corps. Every unit can have a unique unit range and accuracy. You can select an accuracy curve from the drop-down menu or you can create a custom accuracy curve by clicking on the pencil (Edit) icon.

Window for editing the artillery accuracy curve. There are 100 points and you can set each one individually. This also supports a digitizing pen and drawing tablet. Screen shot from General Staff Army Editor. Click to enlarge.

In the above screen shot from the General Staff Wargaming System Army Editor the accuracy curve for this particular battery is being edited. There are 100 points that can be edited. As you move across the curve the accuracy at the range is displayed in the upper right hand corner. Note: every unit in the General Staff Wargaming System can have a unique accuracy curve as well as range and every other variable.

Screen shot showing the Range of Influence fields for a scenario from the 1882 American Kriegsspiel book. Click to enlarge.

In the above screen shot from the General Staff Sand Box (which is used to test AI and combat) we see the ROI for a rear guard scenario from the original American Kriegsspiel 1882. Notice that the southern-most Red Horse Artillery unit has a mostly unobstructed field of vision and you can clearly see how accuracy diminishes as range increases. Also, notice how the ROI for the one Blue Horse Artillery unit is restricted by the woods which obstructs its line of sight.

Screen shot of Antietam (dawn) showing Red and Blue ROI and battle lines. Click to enlarge.

In the above screen shot we see the situation at Antietam at dawn. Blue and Red units are rushing on to the field and establishing battle lines. Again, notice how terrain and elevation effects ROI. In the above screen shot Blue artillery’s ROI is restricted by the North Woods.

The above ROI maps (screen shots) were created by the General Staff Sand Box program to visually ‘debug’ the ROI (confirm that it’s working properly). We probably won’t include this feature in the actual General Staff Wargame unless users would like to see it added.

This is a topic that is very near and dear to my heart. Please feel free to contact me directly if you have any questions or comments.

What’s Wrong With This Picture?

Computer AI representation of battle lines for Antietam, dawn September 17, 1862. The AI is locating the Schwerpunkt or place to attack. Click to enlarge.

I was looking at this screen shot I posted as an illustration in my blog post, Battle Lines, Commanders & Computers (link) and something didn’t look right. Specifically, it was the Flank markers for Stewart’s cavalry on the Confederate left flank. And, the more I looked, the more I noticed other problems: Stewart’s cavalry was captioned as being in two groups (Group 6 and Group 7). There was an extra Flank marker with the Confederate reinforcements entering the field at the bottom of the map, the Flank markers for Union Group 2 were clearly wrong, too.

This began a two week long bug hunt that took me places I didn’t expect to revisit. To make a long story short – and how often have you heard that phrase but this is actually one of those few occasions when it’s true – I had to go back and look at my original computer code that I wrote in grad school and it turns out that there was a ‘worst case’ bug that just happened to pop up with the Antietam scenario.

This is what the battle lines and Flank markers should really look like (note: the map layer is turned on here and you can’t see the elevation layer like the top screen shot):

Correct battle lines and Flank markers for the battle of Antietam. Screen shot from the General Staff Sand Box AI test program. Click to enlarge.

I also discovered a logic flaw – a mental bug, if you will – in my definition of Flank units. Previously, I defined them as as the ‘maximally separated units of a MST (battle line) group’. That seems correct but if you think about it there are rare circumstances when this is not correct (think of the horseshoe lines at Gettysburg for example). I changed the definition of Flank units to: the ‘maximally separated units of a MST group with only one neighbor (i.e. they are at the end of a battle line and, therefore, have only one neighbor).

I thought I would finish this blog post with something that is truly unique: the very first computer bug:

On September 9, 1947 Grace Murray Hopper records ‘the first computer bug’ in the Harvard Mark II computer’s log book. Apparently, the moth was caught in a relay switch. And, yes, this is where the term comes from. Click to enlarge.