Are there any shoot-em-ups in which the enemies actually evolve as a response to their relative success against the player?

By evolve, I mean that each enemy has some genetic information encoding their behaviour. Enemies which last longer or do more damage are allowed to mix their genes to create new enemies. This way the enemies would become better adapted to the particular player.

  • 1
    \$\begingroup\$ Never seen it (or not that I was aware of). It seems that it would take long to get a decent opponent modeled after the player though, and once the player mastered the evolved enemy the DNA is too specific to create a new challenge (as the weak genes have been evolved away). So you'd have to introduce mutation to reintroduce some old genes periodically I guess. Cool idea though \$\endgroup\$ – Kaj Aug 15 '10 at 1:29
  • \$\begingroup\$ You might be interested in Adaptive AI. While I'm not sure if it employs a genetic algorithm, several of the Crytek games feature Adaptive AI that adapts to player tactics. \$\endgroup\$ – DrDeth Aug 15 '10 at 21:26

There is a Japanese freeware PC shoot-em-up called "Warning Forever" that evolves the bosses based on how you approach them, in a way that is more than just switching between presets. It's also pretty fun for being free.

Another game I found is "Evolution Shooter" by Petri Purho, which uses an evolutionary algorithm to combine entire shooter game definitions, as opposed to individual enemies.


Not real-time, nor a shooter, but when developing a car combat game we did use an evolutionary process to build the AI. Enemy cars could see the edges of the track and locations of other vehicles.

However things like how hard to break, when to start, how hard to steer around opponents, when to target opponents vs. focus on driving were all setup as variables. Then the AI was setup to run laps all night long slowly tweaking the variables. Each morning we'd look at the previous nights data, find the ones that had the best lap times/kills and then mutate that again. The end result was an AI that could be dumped into any track and be able to compete.

This was all then baked into the game, so it didn't learn from playing against the player.

There are several sports games that track player tendencies and then work to anticipate those choices, not sure if anyone has shipped adaptive AI in a shooter though.


There is an RTS that is based around this concept. But instead of enemies that evolve, the player trains units which are able to breed, adapt, and evolve.


The technology behind this is a mix between Neural Networks and Genetic Algorithms, called NeuroEvolution.

Neuro-Evolving Robotic Operatives, or NERO for short, is a unique computer game that lets you play with adapting intelligent agents hands-on. Evolve your own robot army by tuning their artificial brains for challenging tasks, then pit them against your friends' teams in online competitions!


Galatic Arms Race

There is another game that used a similar technology but instead evolved custom weapons based the effectiveness and overall gameplay of the user.

A key objective of this game is to explore the potential for automated content generation technology to intelligently create content for games. In GAR, all player weapons are generated by the cgNEAT algorithm based on weapon usage statistics. However, cgNEAT does not simply respawn weapons that people like. Rather, it creates new weapons that elaborate on the ones that have been popular in the past.



Both of these games were developed as academic pursuits.


Not exactly evolution in the sense you mentioned but still interesting:

In both of the Left 4 Dead games by valve there is an AI director that adapts the level based on how the players behave.

"The Director places enemies and items in varying positions and quantities based upon each player's current situation, status, skill and location" - http://en.wikipedia.org/wiki/Left_4_Dead#AI_Director

  • \$\begingroup\$ To the best of my knowledge the AI Director is not evolutionary at all. It creates a dynamic environment, and is heuristically-informed, but is still based on static data sets. It's pretty hard to reliably feed whether you successfully "scared" or "tensed" someone back into a computer. \$\endgroup\$ – user744 Mar 21 '11 at 22:52

Rez changes the difficulty of the level boss based on how well you did in that level.

  • \$\begingroup\$ I'm mooking more for some kind of genetic evolution. \$\endgroup\$ – Rocketmagnet Aug 15 '10 at 1:19
  • \$\begingroup\$ Maybe something like Diablo 2's randomized enemies? \$\endgroup\$ – coderanger Aug 15 '10 at 2:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.