BrainyBeard | Artificial Intelligence in Games News – Week 13 - BrainyBeard

Artificial Intelligence in Games News – Week 13

Posted by  on 
24th April 2017

Screenshot from Ultimate Battle Simulator

Weekly Roundup!

Welcome to the weekly roundup, where we draw your attention to everything newsworthy in artificial intelligence in games! If you’ve found or shared something that you think belongs here, contact us on Twitter! Last week, we dug deeper into Street Fighter II AI and the advances being made in RTS AI. You can find that here.

 

Ultimate Epic Battle Simulator

Source

Ever wanted stage ginormous battles with thousands of AI characters? Our highlight of the week is a podcast with the creator of Ultimate Epic Battle Simulator – Rob Weaver. In the interview, Rob talks about how he designed a VR title that transformed into the early access game available on Steam. Built in the Unity engine, the game allows players to create matches between hundreds of thousands of characters on screen at once. Obviously, there is the potential for a performance catastrophe here, but Rob has included all of the tricks in the book to save frames. This includes ditching the bones in traditional bone and mesh systems for characters. Well worth a look!

 

Grand Theft Auto Driverless Cars

Source

Once again Grand Theft Auto appears in the game AI weekly news. This time we’re taking another look at its AI drivers, traffic and driving systems in general. A while back researchers at Darmstadt University successfully developed a technique to acquire visual information from GTA V. This is significant because GTA offers a rich realistic game world perfect for testing simulations within, meaning speeding up research into self-driving cars at a low cost. It is possible to recreate almost any situation in a game, cars driving in the wrong lanes, pedestrians walking into the road etc. The best part is that these occurrences and be sped up, slowed down and repeated over and over again. The article mentions a technique Toyota uses called ‘The Quick Brown Fox Test’, which involves cars driving in the worst possible conditions.

Grand Theft Auto 5 screenshot

Screenshot from GTA V

 

My Love for Bots in Games

Source

Before the days of online gaming, instead of facing humans, we faced bots. Games such as Perfect Dark on the Nintendo 64 provided an entertaining challenge for players as they took on a team of AI. However, with the dawn of fast internet speeds and online multiplayer, the – as the author puts it – ‘competition shifted from schoolyard to world-class’ and almost killed off the bot. Playing humans at games is stressful especially those that don’t have responsibilities and therefore have all day to burn hours playing their favourite game. Bots offer more predictability and a relaxing experience, especially since they don’t scream abuse through the player’s headset. The author analyses the presence of bots in modern games such as Rocket League, For Honor and Overwatch and asks the question; Are there enough gamers becoming tired of online multiplayer games to see a bot resurgence?

 

Grids AI

Source

Every now and then a title comes around and revolutionises a genre. In the driving genre that was Grid back in 2008. As hardcore drivers waited for Gran Turismo 5, Grid offered gamers a twist by allowing them to rewind gameplay. In this article, the author describes two components of Grid that made it an awesome game, rewinding time and aggressive AI opponents. Rather humorously the author describes the player as surviving the AI not racing them! Whether it be cutting the player up, shunting them into walls or just being a general nuisance, the AI felt great to play against. Whilst this isn’t modern news, it’s great to read articles like this to help better understand what has worked and what hasn’t worked in game AI over the years.

Screenshot from Grid

 

Need More News?

BrainyBeard keeps you up to date with all of the latest artificial intelligence news in games. Sign up to our newsletter below to ensure that you don’t miss our first newsletter going live later this month!