• Register

Lord of Rigel is a turn based 4x (Explore, Expand, Exploit, and Exterminate) space strategy game. Our goal is to create an accessible, yet deep, turn based strategy game with real time tactical combat that will provide a complete space-opera experience. Lord of Rigel will feature a diverse universe with young and elder races that will struggle to shape the future of the galaxy. Players can make a choice between warring elder races or forging their own alliance with a coalition of alien races including minor species. Given all the resources of a galactic empire, players can choose how to navigate through complex diplomacy and survive grand menaces invading the galaxy using an array of game mechanics such as a galactic council, espionage, research, and military force. Players can expect Lord of Rigel to be completed in 2016 and available for Windows, OS X, and Linux.

Report RSS Mechanics Series #7- Tactical Combat AI

This entry of our game mechanics series covers the tactical combat AI in Lord of Rigel and how it works.

Posted by on

Artificial intelligence is a central but often not discussed part of games, especially strategy games. You need to have an opponent who can pose a challenge, but still be defeatable. Most importantly the AI needs to feel consistent. For Lord of Rigel, we wanted to take this a step further and we wanted each species to have its own set of behaviors in tactical combat to give a slightly different feel behind each one.

We use what’s called a state based AI. This basically means there’s a giant flowchart of “states” or conditions i.e. has a ship taken 50% damage, is there a hostile ship in range of its weakest shield facing, are there missiles targeting friendlies, etc. We then weight different possible responses ex. retreat, fire, move based on the current game states. The differences between species are weighted probabilities we put behind these. So in the case of a ship say taking 50% damage the Katraxi AI is weighted to attempt to position itself to self-destruct and damage as many ships as possible. The same game state for a Human AI will instead be weighted for the ship to attempt to retreat from combat.

The end result of doing these elaborate logic trees for each species is that we can create the feeling of unique behaviors for each race in the game. The Katraxi feel aggressive and more warlike than the Human AI since they’re weighted to self-destruct, close in on targets, and attempt to board ships. Humans instead use standoff distances, a heavy reliance on theoverwatch mode, and tend to retreat when damaged to cut their losses. The Tharrn will attempt to focus fire on what they view as the strongest threat. Overall this means that our AI won’t actually make optimal decisions in many cases (ex. always fire at weakest target, always use optimal firing range for beam weapons) but instead each species will feel like it has its own quirks, strengths, and weaknesses that a Human player can learn to play with. The goal with tactical combat isn’t a “smart” AI that will always beat the player, but instead an AI that feels like there’s a predictable “Katraxi” or “Selach” on the other side issuing orders. Since each species will be a little different in how the AI operates, it’ll also require some adjustments on the player’s part to adapt tactics to each species. The use of close range weapons like fusion cannons may work well against a Katraxi who will close in but won’t be as good against a long range fighting AI. The same philosophy behind our tactical AI we use for the strategic AI, which we will cover in a few weeks so stay tuned!

Join us next week as we discuss the colony population system!

Post a comment

Your comment will be anonymous unless you join the community. Or sign in with your social account: