MGP19 MotoGP 19: A Closer Look at 'A.N.N.A'

Paul Jeffrey

Premium
A.N.N.A. What's that I hear you say? Well, it's the new Artificial Intelligence system set to join the new MotoGP racing game later this year.


It has to be said that Milestone have done a very commendable job of improving the MotoGP game franchise year on year, and with the 2019 instalment set to drop in the not too distant future, the Italian development team have been busy making some rather bold claims as to the AI improvements set to feature in the new title.

Codenamed A.N.N.A, the new AI system in MotoGP 19 has been promised to offer players a much more challenging and authentic offline racing experience than has ever been possible before in this long running franchise.

Speaking to the official PlayStation blog, Milestone explained that the AI in MotoGP 19 isn't actually programmed to respond to certain situations, but actually learns and teaches itself how to race...

The only way to give this amazing community the AI they deserved, was by stop trying to create an AI, but letting it build by itself. So, the A.N.N.A project came alive. An AI that learns by its own mistakes, by trial and error, by building a Neural Network similar in every way to part of our brain. Just like we did when we first learned how to use a bicycle.
Hello.. that sounds interesting doesn't it? Well Milestone go on to explain to the PlayStation Blog in more detail (as well as that cool video at the head of this article)…

A.N.N.A. stands for Artificial Neural Network Agent: an AI that does not rely on predefined commands, written by a designer, but on rewards. Each action she makes – yes, some of us likes to address our AI as a “she” – can have a positive or negative outcome, and depending on this, she may receive or lose points.

This triggers a learning process that allows A.N.N.A to remember which actions are good (for example when to brake at the right time) and which are bad (like, well, going against a wall, or against another rider)

we now have a complex, incredibly smart AI that is capable to challenge and beat even the most skilled of our devs acting as a real pro rider – but no worries, its difficulty is scalable, so all players of all skill level will enjoy the game!

In fact, we ended with a super-tough AI and an even tougher problem: how to make it enjoyable also for the average players, without losing all the awesome work we did on it? We knew the priority was to preserve the complex behaviour it had developed, and to do so we had to work on some interesting techniques.

Meaning that we had to slow it down without making it look clumsy or unbalanced. We achieved this by controlling where and how much it needs to slow down. It’s still the neural AI in control, consciously not cranking up to 100%. Everyone here in Milestone is so excited to see how our Community will react to this new level of realism, of speed and fairness toward other riders.
This could be a revelation in sim racing AI.. I can't wait for release on June 6th to find out for myself!

MotoGP 19 will release June 6th for PS4, Xbox One and PC / Steam. The Nintendo Switch version will arrive a few weeks later.

Check out the MotoGP 19 sub forum to keep in touch with all the happenings from Milestones new bike racing title, and to chat and get speculate about the game with your fellow MotoGP fans!

Like what we do at RaceDepartment? Follow us on Social Media !


 
 
i love the bike games ride 3 etc but it is hard to trust milestone when they leave a lot of unfinished and problematic issues unaddressed while reaching for the $
I totally agree, and the speed with which they drop games is deplorable. Ride 3 I consider still new, but it has zero attention now. They don't even update the website with which DLC is released.
Such a good game, but still marred by massive bugs eg non re-mappable controls for multi player - I can't even change gears.
So NOW we are to trust them?
 
I'm concerned about how fast and precise this AI learning process appears to be turning out. I want it to make mistake, to miss breaking points, to cause accidents, not to have those mistakes polished!
I suppose mistakes can be added as a sort of "noise" to the actions, just like we know what we want to achieve, but are imprecise with our inputs.
The lower the difficulty level, the higher this noise can be, and then the NN will have to reassess it's situation and take further actions, which is pretty much what we do.
Sounds like a very reasonable approach to more realistic AI, I'm curious as to how this will develop.
 
We need a MotoGP with real AI (maybe A.N.N.A i hope) + the speed sense of "Isle of Man" game (bad game but great graphics+sense of speed) + GPBikes simulator physics (if you don't know this one for some reason, check it out). That'll be total perfection for a MotoGP game.
And why not... like Dirt games, add an "Arcade" physics mode for Casual players to make their life easier xD
 
I always felt racing AI haven't seen any real improvement since its early days. I felt not enough R&D went into it and it fell kind of into a rut. I hope that will now begin to change starting with this and more devs will try to innovate in this area. Hopefully it's not a gimmic.
 
i love the bike games ride 3 etc but it is hard to trust milestone when they leave a lot of unfinished and problematic issues unaddressed while reaching for the $

In this they are not unique.... sadly.
But they seem to get more grief than some other devs...

I think ALL devs should be held accountable and scrutinized, not just some.
 
We need a MotoGP with real AI (maybe A.N.N.A i hope) + the speed sense of "Isle of Man" game (bad game but great graphics+sense of speed) + GPBikes simulator physics (if you don't know this one for some reason, check it out). That'll be total perfection for a MotoGP game.
And why not... like Dirt games, add an "Arcade" physics mode for Casual players to make their life easier xD
totally agree GPBikes Sim is amazing been working with it for a week now and by far brilliant , whether it becomes complete out of beta we will see but with the mods i really like :)
 
Back
Top