I think the future of game will integrate with our human body that can make the feeling as the reality, I played many game console, and see that Nintendo Wii that they can do great job. Some day, it might only a small control in our hand that we can control for everything in game.

The popular set of AI primitives emerged based on experience meeting design needs and requirements. In my experience, design basically describes the behavior they would like to see AIs perform (or many cases, that they require for the AI to be considered acceptable) like an expert system.

‘If the AI outnumber the player 3:1, one of the AI should charge in.’
‘If the AI is takes damage, I want it to displace’
‘Can I get the AI to move to particular places based on where the player is? Don’t let him get too close though - hes uses a assault rifle which isn’t a good short range weapon’

People start looking beyond FSM permutations when the behavior design based on more complex design requirements, implementation complexity or maintenance issues. Pay more in performance/complexity when it is needed. See offline vs online data processing; doing it online means more opportunity to be dynamic, but offline may allow you to allocate those resources to other interesting AI behavior.

Moving forward, I think the next big application of AI/automation to be in the tools. Content creation has been the dominate dev cost (and therefore one of the biggest barriers to iteration/exploration of the design space) for a while. Surely there are plenty of ways we could apply AI techniques to increase production rate.

Cool game played it and can’t wait to start work on my own.

I hadn’t heard of Empire Earth using pathfinding as a pre-compute step like that, but I’ve done a similar thing with an FPS AI before.
It was a simple waypoint/node-based navigation system - I found that pathfinding from every node to every other node and counting how many times each node was ‘visited’ (and then normalizing the results) gave a pretty decent influence map for recognising choke-points.

I actually got the idea from a GDC 2001 paper called “Terrain reasoning for 3D action games” By William van der Sterren, which explains how to make influence maps for good sniper locations, specifically.

However, in line with this article, after finding ways to pre-compute all of these great influence maps, I didn’t get time to actually make much use of them! Some of them were used to supplement the A* heuristics (move from A to B, but stay in cover), but that’s about it…

wow, interesting to find a rather new article about exactly the topic i was looking for.

i want to implement a strips planner for a company for my bachelor thesis, the planner should create small scenes around the player to make the world look more “vital” and authentical.

so since i don’t have that much time left (as usual) i don’t want to do the planner implementation myself (since this is not what i should do anyway) but am looking for a free planner implementation in c or c++ - any suggestions?

thanks :)

Reading over my comment again I think it sounds a bit harsher than I intended…

Dynamic cover *is* a serious issue, and my work-around only holds up for this particular situation (the breakable wooden fence) - if the cover in question was made of metal boxes than can be pushed around, you’ll need yet another solution.

Also, while I’m at it, the wooden fence isn’t in fact cover at all - in military terms it’s only concealment, and a soldier should never try to use it for protection from bullets ;)

They’ll never use a wood fence as a cover, period. :-)

Even if the physics is dynamically broken you could use the volume of the bounding box as an estimator of how much cover was removed. So, at the start the fence is ‘100%’ cover. Then as pieces are removed it will decrease. AI tracks the cover ‘probability’ of the object it’s in cover with. As bounding volumes are very conservative you will reach 0% a lot faster than the actual visual representation suggests, but in practice that shouldn’t be a problem (it makes your AI appear smarter the earlier it leaves cover which is getting destroyed). At some point your behavior will just decide that cover is getting blown to bits and the AI should do something. This could also take distance to attacker into account (as well as angle of attack). The guy will likely die anyways within the next 2 or 3 seconds, but getting some reaction out of him will make him appear smarter :)

Looking forward to parts 2 and 3.

Nice post. I just feel that hiding behind a fence while they are shooting at you is a bad idea no matter how many “fence bits” are missing… :)

Techical note: I can’t see the embedded video using Opera. I had to dig through the html source to download it.

I love the video blog.

Although raycasting isn’t necessary the only solution as Hodgman pointed out. In both Alex’s example and the former, a character would not consider moving if only the small top piece of the geometry (clearly exposing his head). Of course a precise check like this also isn’t necessary, to ignore it would take away from the immersion and quickly summarize the AI (formations, cover, planning) into being “stupid”.

I don’t agree with your assumption that because the fence is being torn apart using a physics engine, then they only way for the AI to know this is to use ray-casts.

The fence seems to be composed of dozens of “fence bits”, which remain as static geometry until a “destroyed” event is triggered by the game (e.g. getting hit by ‘x’ number of bullets). When this happens, a “fence bit” stops acting as static geometry and is handed over to the dynamics system.

Seeing that this hand-over from static to dynamic is triggered by the game logic, it seems quite feasible that the game-logic could easily record which sections of fence are still intact, and which are not.
Meaning that the AI characters could quickly and efficiently query the number of “fence bits” in front of them - if this number is above ‘y’, they know their cover is ‘good’, if it’s between ‘x’ and ‘y’ the cover is ‘acceptable’, and if it’s below ‘x’ the cover is ‘bad’ not non-existent and it’s time to move!

Assuming that once a fence bit becomes damaged that it will never again act as cover, I think this particular examples has just been solved ;)

Great job Alex , i am waiting to see your AI skills in killzone2 next year :) .

Regards
Bader

I have enjoyed playing GTA4 thus far, well, most of the time.

As far as AI… theres really not much to talk about in GTA4. Its definitely not a selling point of the game, but its also not its biggest flaw either.

To digress from AI for a bit… GTA4 ( while a generally fun game ) has some super annoying controls from time to time.

As mentioned already - its a fight to actually get into cover or the right cover spot. And its hard to tell why sometimes it will let you blind-fire from cover and other times it won’t.

Sometimes you press y to get in a car and he starts running down the block to get in a different car… And then when you do he plays the same painfully slow animation even if someone happens to be 5 feet away shooting a shotgun at you.

And don’t get me started about trying to grab onto ladders or overhangs … especially grabbing ladders while in water ( ugh ).

Playing GTA4 ( and this might apply to 3rd person games in general ) I feel like I’m controlling the player through about 7 layers of interference. Actually getting him to do what I “really” want is a constant fight.

Actually, to bring this back to the AI realm … one possibility to improve the controls / usability of any game is to leverage AI. Theres plenty of room to go wrong here — Clippy says, “It looks like you are trying to start a new document” — but if done correctly a game with a mass of confusing buttons could be reduced to a few “smart” buttons ( for example ).

I responded with a blog post of my own: http://tinyurl.com/6adu4h

In short: I don’t have the amount of experience to know how much they can “fake” in games nowadays, but in general, it is the intractability of the problems we’re trying to solve that makes game AI still suck (or, you can’t separate it out from the excuse of limited processing power).

Good question, made me think.

I’m impressed with the level of dynamism they included in the first place anyway. Large improvement over the other GTAs.

This is just a stab in the dark but maybe the inclusion of Euphoria reduced the available resources that could be used on NPCs?

random can create the ilusion of npc thinking,an error introducing a little delay in their reactions can create the ilusion of npc thinking…and more important,deciding

AI,and what about HI(human inteligence)?,maybe what you think is inteligence is just a error from nature that makes one kind of animals(human) repeating with words what they are feeling

feel hungry,think you are hungry,open fridge and decide btw 2 kind of food,but are you really deciding or just translating into words what your body likes better?

Animation is neccesary to see how the AI works

I´m not programer,i´m fan of STALKER,they have create a really cool game,i would like to change a few things in npc behaiviour,but you can wacht AI working OK almost time,the problem is,thougt,npc in STALKER have their habilities cutted,100 npc working OK are RAM eaters
i like Amk mod…cause they have a primary premise to do,search for artefacts

if a npc moving with this objetive and then fighting and skiping dangers and colecting and trading is not good AI,god will tell

i´ve decide go back to school,i´m oldie but i love Human Behaviour,so i want make AI seem like human,i´t can be done,just needed power pc

thanks Dmitry