Back

Explore Courses Blog Tutorials Interview Questions
0 votes
2 views
in AI and Deep Learning by (50.2k points)

There are many papers about ranged combat artificial intelligence, like Killzone's (see this paper), or Halo. But I've not been able to find much about a fighting IA except for this work, which uses neural networks to learn how to fight, which is not exactly what I'm looking for.

Occidental AI in games is heavily focused on FPS, it seems! Does anyone know which techniques are used to implement a decent fighting AI? Hierarchical Finite State Machines? Decision Trees? They could end up being pretty predictable.

1 Answer

0 votes
by (108k points)

The Fighting games are built on typical multiple tiers of mechanics i.e. from ranging basic movement of the model to the introduction of special attacks. The template set by Street Fighter II has since been adapted over the years by pretty much every successful 2D fighting game franchise, such as Mortal Kombat, King of Fighters, Samurai Shodown, the Capcom Vs series and of course, Killer Instinct. The subsequent strategies that emerge are often highly contextual, as we need to think not only about what our opponent is doing, but also what we want them to be doing in order to do maximum damage.

This introduces a number of high-level strategies such as:

  • Rushdowns: Close-range aggressive gameplay to try and force the player to perform the wrong move.

  • Footsies: Where players shuffle in their movement to try and get the opponent to expose themselves to attack.

  • Zoning: Where players use far-range projectile attacks to keep players at a distance.

  • Setups: Where you execute actions in advance of a combo in order to position both the player and opponent to create an opportune moment.

 Identify and develop an AI solution capable of recognizing long-term player strategies from a human player, and then mimicking these strategies. The solution would be deployed in a multiplayer fighting game, where two humans would play each other, with the AI agent learning the strategies of a given player. Once the AI agent has learned the strategies utilized by a human, it should be capable of using these strategies whilst playing against a human. You can refer to the following links for knowing how AI has contributed its part in building fighting games:
 

https://dspace.lboro.ac.uk/dspace-jspui/bitstream/2134/16380/3/Thesis-2014-Saini.pdf

https://towardsdatascience.com/killergroove-c4c606601a8a

If you want to learn more about Artificial Intelligence, then you can join the Artificial Intelligence Course.

Hope this helps!!!

Browse Categories

...