If you still haven’t already succumbed to skipping my ramblings—go watch the demo of levelHead by Julian Oliver! (embedded video below.)

levelHead Video

levelHead v1.0, 3 cube speed-run (spoiler!) from Julian Oliver on Vimeo.

Main points of fascination

• Physical object interaction; virtual worlds are dependent- and intertwined with physical objects (the cubes) in the environment. (Opposed to displaying virtual objects that have no connection to reality, which in my opinion is removing the “reality” out of “augmented“.)
• The cube, simple as it is, gives the impression of a gateway into an entirely different world.
• Simplicity. With the environment shaded and lit, the flat white character is simple and adds a mysterious touch to the experience.
• The cube is the controller as well as viewer; an intuitive solution for containing the game experience entirely within a simple paper cube.
• The game is easy to replicate if desired; all you need is the right program and a paper cube with printouts.
• Considering the Future: Remember Myst? I can easily envision an entire game in that style: purely contained within a paper cube, or even interchangeable shapes such as orbs or other simple ones for different environments (perhaps even a few in-game tools).

LevelHead information excerpt

Using tilt motions, the player moves a character through rooms that appear inside one of several cubes on a table. Each room is logically connected by a series of doors, though some doors lead nowhere (they are traps).

The player has 2 minutes to find the exit of each cube, leading the character into the entrance of the next.

Work is also being done to use invisible markers such that the cube itself appears entirely white to the naked eye.

Visit the project page of Julian Oliver’s levelHead

Thanks for an inspirational game concept, Julian!

Automatic recommendations:

1. Augmented reality games, but what’s reality doing there?
2. Useful augmented reality tool shipped by US Postal Service

[If you see no video then click here to visit the post]

Automatic recommendations:

1. Wikitude AR – Augmented reality on Google Android

To answer this bluntly: Game AI is very different from it’s non-game counterpart, and it’s not my field of study. I’ve only compared modern games through a window. However, Alex of AIGameDev has superb coverage of AI in games and the top AI games of 2007, by community vote. The top of the line are Half-Life-2.ep.2 and BioShock.

But regarding Game AI in general: modern games are horribly void of intelligence. It depends on where you set the bar, certainly. There’s tons of AI in modern games compared to 5 years ago. But the first thing to note is that Game AI is not the same as AI. It’s a subset of it. Just like discrete mathematics are a subset of mathematics. And moreover, Game AI is a very specialized subset—it has well defined goals, models for construction and limitations.

Games are governed by laws of commerce first, then innovation.

A game is governed by different laws than academic, general AI R&D. It’s a commercial product, and commercial products depend on older methods wherever possible—methods that have proven successful. Most (sane) business men do not put all their money on a new and untried idea because they don’t know if it will succeed. So, most of commercial products are bulked up with a lot of things that’ve been successful in the past and then leave a breadcrumb for innovation.

This is very different from academic AI research where the point is to do things that’ve not been done before.

Now aside from these drastically different goals of commercial ventures and academic ones, a game’s purpose is to entertain. As long as the player is entertained it doesn’t really matter what goes on under the hood. Because of this, there’s a certain witch hunt that takes place in the game industry:

A modern game developer is on a mission to slaughter innocent intelligent processes wherever possible.

A game AI developer tries as hard as he can (usually at the bidding of a project manager) to minimize intelligence. One reason is that intelligent processes are massive processing-power hogs. Thus, like an obese overeater the systems must forcibly give away every other meal to accommodate an average person’s desktop PC — and those machines don’t have much elbow-room to replicate the massive crimson jelly residing in the heads of animals. Human or other.

And then there are graphics, another obese overeater, who also need a place at the table. And because games are governed by the laws of commerce, Game AI must leave at least five chicken wings more than it ate itself for its obese, graphics rendering sibling. Beautiful games get a lot of coverage and attention, and developing graphics is a question of engineering. In a business plan it’s therefore rational to emphasize graphics. Both in terms of predicting the amount of effort required to implement it and the potential payoff.

Because of these severe limitations on how much processing power the intelligence is allowed, developers are forced to dumb-down the processing and make their AI appear intelligent instead.

To some it may not be clear what the difference is between making something appear intelligent and actually making it intelligent. After all, there has to be some amount of intelligence if something’s intended to keep its appearances. Right?

An intelligent system is expected to produce solutions to problems, uncertainty and often in complex situations. Appearances, however, are concerned with making an observer believe they are intelligent. To accomplish this in games the environment (the input to the AI) is kept controlled and limited. For a vivid example of this:

• NPCs in games don’t use computer vision to perceive where the player is, instead they get fed (X, Y, Z) coordinates, giving the appearances of eyesight and visual processing capabilities.

The game designers tailor the environment and its limits to make sure that the intelligent processes can handle them, and vice versa. It’s the lifelike gatekeeper who doesn’t need to know how to find his way home because he has no home. It’s the terrorist that can pull a trigger but couldn’t count his fingers if you took his shotgun and held it to his head (and you can’t).

For a different and real-life example of appearances versus actual intelligence, to show how diversified the game AI “trickery” can get:

• The developers of Halo 3 found a correlation between how smart the AI was and how tough it was. If they made the NPCs smarter the game became tougher. If they increased the health of the NPCs, making them tougher, the human players perceived them as more intelligent.

The first rule of AI in commerce is that you don’t talk about AI in commerce

Like I’ve stated in another post, the gaming industry is pretty much the only industry that dares market products using the term artificial intelligence. The reason for that is complicated and I won’t talk about it from all perspectives. But what I will tell you is that its partly due to people knowing what to expect from games.

The gaming industry builds from the sets of platform resources and devteam innovation. These are sets that the consumers know. It’s the set that the elite gamer knows because he understands the nature of programs and what the required Hz’s actually stand for. And it’s a set that the average gamer knows because he’s fought space aliens so many times that he’s learned what they’re capable of. With consumers that know what to expect, the term can be used without people boiling a can of hype.

In contrast, Academic AI builds with a set that appears to the consumer as one of infinite possibilities: Because people don’t know (exactly) how the mind works, they/we can’t evaluate how far we are from recreating it in machines. It’s unknown. And because the average consumer can’t accurately evaluate the unknown, it doesn’t matter if yesterday’s AI was primitive; most will still anticipate it advancing leap-years overnight. Just like kids in the backseat of a car asking if it’s: “human now? it must be human now? how about now?“.

Thus, when the term AI is used in areas where the limitations and previous products don’t foretell the nature of the next, people start imagining Terminators all over again. And then they get incredibly disappointed when they realize all the AI can do is answer questions about farm animals. (And only when you begin a sentence with “What is…”).

That’s obviously not a scenario a businessman would like to encounter, so it’s best just to focus on something else when marketing, like Apple does when presenting its operating systems to the public.

In summary

Game AI is just one of many different subfields of AI and is governed heavily by the laws of commerce, entertainment value and modern desktop computing resources. While it may seem it must intersect with other subfields, the truth is that its a somewhat isolated field with its own sets of tricks and tools. (Mostly tricks.)

At times, games may also feel like they are the only commercial products successfully employing AI. But this is largely because of too high consumer expectations to AI in other products, and the consequent fact that companies don’t like mentioning that their product uses, what is by definition, artificial intelligence.

And finally, a difference between apparent intelligence and actual intelligence is that the latter figures out solutions to problems, while the former doesn’t care about what happens under the hood as long as an observer thinks it’s intelligent. A lot of the times—that doesn’t involve intelligence at all.

Links & references

• Halo 3 AI “Trick” example from Teaming up with Halo’s AI

Automatic recommendations:

1. Leopard Sports Superb AI, Apple Doesn’t Mention It
2. My Robot is Your Congressman

It all started as a temporary venue for my creative outbursts, which regularly occur when I have an otherwise tough and monotonous schedule. But my interest and the project itself has grown steadily. Moreover, the diversity of the project is teaching me a heap of new and necessary things. And what’s most important is that it’s really fun.

The Basic Game Concept

The idea started simple: create a small 3D, first-person puzzle game. Two to three rooms where you have to click around to figure something out. I recently mentioned Myst is one of my favorite games, and that is the base concept I have in mind for gameplay … but of course with a bit of my own spices and homebrewed lager. A game like that is perfect for a short term project as it requires minimal stuff like animations and physics programming.

I want it sci-fi themed as well, and I quickly realized that a sci-fi short story I drafted a while back would make a perfect setting for an intriguing and exciting puzzle game. I’m not comfortable with disclosing any details yet, all I’m saying is that it involves intriguing architectures and what I think are somewhat innovative concepts for puzzles.

3D Game Platform

What little experience I have in 3D, and other related cookie jars, was accumulated when I made my brain’s MRIs using jPCT (I was so happy to get my brain online that I designed and donated this logo to the engine). So it seemed an obvious choice for what was supposed to be a rapid-prototype project. However, now that my interest in the project continues to grow I’ve been considering whether I should explore other engines.

In-Game Screenshots

I’ve spent about 6-10 hours on the whole thing over the past four weekends and much of that time has gone into things you can’t actually see on-screen (object interaction, importing and interpreting models, etc). But I figure the latest in-game renderings are good enough to show. All models and textures visible in the screenshots were made from scratch, the textures in Photoshop and the models in Google SketchUp.

What to Do With Pet Projects That Want to Grow Up?

On whether I’ll start investing regular time in this brainchild of mine: Not yet. Although the idea has me captivated and I’d like to take the project to the next level, so do many other ideas and I only have so much time. A curse and a blessing is that I have a constant flow of ideas and am occupied with at least one long term project (usually more), and several smaller short term projects at any given time. The way I cope is write them down, that’s the rule of thumb: Write the ideas down and allow them to fester a bit. See what sticks. Other times, like this one, I feel a need for a creative venue and just start working with no care in the world of whether I actually finish it or not.

The ideas keep coming but some ideas I give more thought to and gradually build on. Others whither away. This process naturally fleshes out what’s important to me and guarantees each idea a good amount of boiling time, makin it nice and well done if and when the time comes to have it for dinner. This works especially well for big ideas — if I know it’s going to be time consuming I give it a longer period of consideration. If I’m still interested and enthused after a long enough period of time relative to the time required to implement it — then I’m more certain it’s a good idea.

This game project is still underdeveloped; the time I’ve spent thinking about a mature version of it is unproportional to the time required to implement it. Thus, I’ll continue doing what I’ve been doing. Gradually add one texture here and one wall there, until maybe I end up with a demo that looks so hot I’ll have to take the project to the next level. Time will tell. And we’ll have fun in the meantime.

Links and References

• jPCT 3D Engine (Java)
• My Brain Online in interactive 3D
• Think Artificial Post: Puzzling With Portals

No related posts.

Browsing through the 9rules Clips, I came across an article on a game I saw a video of some time ago, but had completely forgotten about: Portal.

Portal is a first-person perspective puzzle game created by Valve (creators of the famous Half-Life series) that debuted today in a gamepack called The Orange Box. In a nutshell the game consists of you — armed with a gun capable of creating portals: oval shaped holes that allow you to instantly travel from one point in 3D space to another. The gameplay revolves around using this weapon to get around various physical obstacles. For example, getting past a giant pit of fire by creating a portal on your side, and on the other side of the pit, then walking through. This is the first puzzle shown in the promo video they published about a year back (see below).

Of course, the puzzles get harder and this seemingly simple concept of a portal-gun allows for an incredible range of solutions and uses, especially because the physics of the game are superb — so you’ll be jumping down a hole on the floor which translates to thrust out of a corresponding portal on a wall. Another interesting use is a portal in the ceiling and floor which creates an infinite loop to fall through. The (brilliant) video promo below, although being almost a year old now, shows the full impact.

Excited yet? I can’t help but mention how awesome I think that ad is. From the haunting voice of the guide to the sneaky humor and animations; fantastic work of advertising right there. And it’s interesting that it doesn’t bother me at all that the storyline sounds flat, I just want to do those crazy puzzles. And yeah, there apparently is some kind of a storyline, here’s what Wikipedia says:

In Portal, players control Chell, a test participant named in the Combine Portal User Training Facility. [...] Guided by a female electronic voice-over named GLaDOS players use the “Aperture Science Handheld Portal Device” to perform a variety of tests, such as creating portals to knock over turrets and other objects or moving to a previously unreachable area. Players must either complete their set objectives or fail the test.

Additionally they mention that Portal is set within the Half-Life universe, and that the protagonist in the game will potentially make a debut in Half-Life Episode 2, which was released in the Orange Box package with Portal (correct me if I’m wrong on any of these specifics, I unfortunately don’t have much time these days to play games and I don’t know that much about Half-Life).

It’s not often these days that I come across non-violent games I’d like to play even though I keep half an eye out for them (okay, this one is semi-violent, but the main goal isn’t killing). It’s not that violent games turn me off — No. Unfortunately those primal urges can work their black magic like their supposed to when I try those games. I don’t avoid them either to any significant extent (they’re just games, after all) — but when I get the opportunity, I’d take an engaging non-violent game over a violent one because those games are usually unique and innovative. It’s hard to create a game that doesn’t play on animal urges to get you excited; they require thought and care in implementation and conception. Portal is a fine example of innovation in gameplay that opens a new dimension, so to speak, for creating excitement and engaging the player by people using their head rather than losing it. Absolutely brilliant idea.

Links and References

• Portal on Wikipedia
• Valve’s main site
• The Orange Box on Wikipedia

No related posts.

A plasma is one of the phases of matter. The basic phases are solid, liquid, gas — and plasma which is typically, according to Wikipedia, an ionized gas that often takes the form of neutral gas-like clouds or charged ion beams, but may also include dust and grains called dusty plasmas. From what I can gather reading Tsytovich’s paper (see links below), it’s this dust that allows plasmas to form crystals — which are the foundation of the observations noted in the simulations. (Now, be warned: I’m not your resident physicist. I’m your intelligence guy — for physics experts you need to go past Google and turn left on Technorati Avenue. Please correct me if my physics explanations are off somewhere.)

Tsytovich and his colleagues demonstrated, using a computer model of molecular dynamics, that particles in a plasma can undergo self-organization as electronic charges become separated and the plasma becomes polarized. This effect results in microscopic strands of solid particles that twist into corkscrew shapes, or helical structures. These helical strands are themselves electronically charged and are attracted to each other. [Science Daily]

The paper details the conditions required for the appearance of spherical crystals, and these helix shaped crystals. It’s the helices that are of interest, as they bare resamblance to our notions of DNA. Specifically, it appears they can transfer information from one helix to another. There’s no explicit mention of how likely it is that this could happen in space as the paper mainly discusses how it can happen. But nonetheless, this is very exciting news.

To explain further, the particles could form microscopic helical structures capable of replicating themselves and interacting. Through these interactions and perpetuated replication — higher levels of complexity and organization can emerge. As Science Daily puts it, they could “[...] undergo changes that are normally associated with biological molecules, such as DNA and proteins, say the researchers. They can, for instance, divide, or bifurcate, to form two copies of the original structure. These new structures can also interact to induce changes in their neighbours and they can even evolve into yet more structures as less stable ones break down, leaving behind only the fittest structures in the plasma.”

Lem’s planet of Solaris springs to mind when I imagine what continued evolution of such life forms on a grander scale might entail. But I’m uncertain of how large and complex these structures can become; is there a known limit to their potential complexity? If somebody has an idea I’d appreciate it if you left an explanation in the comments.

If you’re not intrigued by now, I followed their example and saved a cookie ’til the end (from their paper):

Our analysis shows that if helical dust structures are formed in space, they can have bifurcations as memory marks and duplicate each other, and they would reveal a faster evolution rate by competing for ‘food’ (surrounding plasma fluxes). These structures can have all necessary features to form ‘inorganic life’. This should be taken into account for formulation of a new SETI-like program based not only on astrophysical observations but also on planned new laboratory experiments, including those on the ISS. In the case of the success of such a program one should be faced with the possibility of resolving the low rate of evolution of organic life by investigating the possibility that the inorganic life ‘invents’ the organic life.

Basically, there exist some uncertainties regarding the Theory of Evolution: We don’t know for certain how complex lifeforms such as ourselves could evolve in such a short period of time. What they’re hinting at in the quote above is that inorganic materials acted as a kind of bootstrapping-template for organic life (that’s you!).

Links & References

• From Plasma Crystals And Helical Structures Towards Inorganic Living Matter (PDF). Tsytovich’s paper in the New Journal of Physics.
• Quotes are from the Science Daily article
• Plasma on Wikipedia
• Carbon-based life on Wikipedia

No related posts.

The simulation takes physics into account: each structure is evaluated depending on its joints and external (gravitational) forces. The propagated force of the entire structure is then used to determine whether the load is too much — whether the system needs to add or remove bricks somewhere to avoid collapsing (“dying”).

The Lego brick structures are represented as trees in the simulator, and it’s to this structure that genetic operators are applied: Mutations modify a brick’s size or position and recombinations interchange subtrees of root-bricks.

The system can thus be given a goal, such as building a bridge over a large gap; you can see an animated gif of this process — where they gave it the goal of bridging a gap between two tables in their lab. The end results are quite interesting to the eye, in two words I’d describe them as “naturally messy”.

The crane’s evolved design is noteworthy. To counteract the load posed on the crane-arm itself, the structure evolve a J-shaped crown that extended from its back, as evident on the picture above. Picture (c) shows the intermediate stages of the crane, picture (d) shows the crane in its final stages — where the counter-balancing structure evolved a solid connection with the base. The below diagram shows the design of the crane in three stages (top is first stage, bottom final stage).

It’s interesting how genetic algorithms result in, like I said before, naturally messy constructs. This is actually one of the reasons that genetic algorithms are popular within artificial creativity research. Evolution results in various “unanticipated” or “surprising” constructs which can be perceived as creative. The problem with this approach to creativity presents itself if you want a creative machine that creates less messy structures. You’d invevitably have to add additional constraints which would in turn minimize the surprising factor of the solution — voiding it of “creativity”. Quite an interesting dilemma.

Related Links & References

• EvoCAD Page
• EvoCAD crane (with video)
• Dr. Pablo Funes’ Website

[All photos courtesy of Dr. Pablo Funes]
[tags]artificial creativity, artificial intelligence, simulation, genetic algorithms, EvoCAD[/tags]

No related posts.

Montreal based company uMind made news yesterday with their AI eLearning software. Claude Frasson, founder of uMind, describes the system as an artificially intelligent tutor that actively evaluates the emotional distress, weaknesses and strengths of the student. Using this information as a basis, it custom-builds course material as the student advances. Taking a look at their webpage header, you’ll notice that education software is a looker as well. A virtual humanoid named Aimy. Via the Gazette:

“Intelligence means adapting to the learner and understanding the capability of the learner,” said Frasson, founder and president of uMind. “In any exam, you have a lot of learners who are able to succeed. We have found all the mechanisms (that help students succeed) and we have applied them to eLearning.”
Launched in March, uMind employs AI to create a virtual tutor that recognizes and adapts to the student’s limitations and emotional distress. The instructor knows when a student is stumped and activates extra teaching modules on the specific subject.

But have we reached a point where this kind of software actually does any good? Does it rival us meatmachines? Not really. But Frasson estimates that this approach, when compared to traditional eLearning, cuts the learning time in half and increases the student retention by 35%. The program also includes a virtual reality type mode, which has been used by the Montreal Transit Company to train employees in handling emergencies. Early estimates indicate training costs close to being reduced by 50%. (That’s according to this one article and man, I haven’t properly confirmed it.)

Hannes Vilhjalmsson over here at RU’s AI lab CADIA has worked on similar virtual environments for education in the past. His Tactical Language and Culture Training systems (TLCTS, developed at ISI) are aimed at training language skills and have, for example, been used by the US military to train soldiers in speaking Iraqi and Pashto.

From his webpage:

The objective of these series of projects was to develop tools to support individualized language learning, and apply them to the acquisition of the linguistic, gestural, and cultural knowledge and skills necessary to accomplish specific tasks in a foreign environment.

In short, the student plays a “serious game” where, in order to sucessfully accomplish his missions, he has to interact verbally with artificially intelligent agents. For those interested, these projects use the open source Unreal game engine. Really cool stuff, and you can imagine how much I wish I had this kind of system to train me german and danish in my younger years.

No related posts.

Neuroscientists’ quest to understand consciousness is a hot topic right now, yet you often pose things via psychology, which seems to be taken less seriously. Are you behind the curve?

I don’t see neuroscience as serious. What they have are nutty little theories, and they do elaborate experiments to confirm them and don’t know what to do if they don’t work. This book presents a very elaborate theory of consciousness. Consciousness is a word that confuses possibly 16 different processes. Most neurologists think everything is either conscious or not. But even Freud had several grades of consciousness. When you talk to neuroscientists, they seem so unsophisticated; they major in biology and know about potassium and calcium channels, but they don’t have sophisticated psychological ideas. Neuroscientists should be asking: What phenomenon should I try to explain? Can I make a theory of it? Then, can I design an experiment to see if one of those theories is better than the others? If you don’t have two theories, then you can’t do an experiment. And they usually don’t even have one.

So as you see it, artificial intelligence is the lens through which to look at the mind and unlock the secrets of how it works?

Yes, through the lens of building a simulation. If a theory is very simple, you can use mathematics to predict what it’ll do. If it’s very complicated, you have to do a simulation. It seems to me that for anything as complicated as the mind or brain, the only way to test a theory is to simulate it and see what it does. One problem is that often researchers won’t tell us what a simulation didn’t do. Right now the most popular approach in artificial intelligence is making probabilistic models. The researchers say, “Oh, we got our machine to recognize handwritten characters with a reliability of 79 percent.” They don’t tell us what didn’t work.

No related posts.

No related posts.

Yeah, I applaude the release of free software for robotics — but I have my concerns. I commented on the Studio’s Digg submission that there was inevitably something in the studio which would handcuff you and your robot to the Microsoft empire, mentioning C# as one of those things. Immediately, an angry Microsoft user accused me of not knowing about Mono, which are additional tools produced by Novell to run .Net technologies on platforms other than Windows (OS X, FreeBSD, Linux, etc.).

Of course, the angry MS user was right. It was a bit unthoughtful of me. But overall I feel that Mono just underlines my concerns: Mono wasn’t initiated by Microsoft; it was built by the community to make amends for MS’ tendency towards maintaining cross-platform incompatibility.

I’m tired of arguing about Mac vs. Windows vs. Linux: I don’t hate Microsoft.

But Microsoft is a corporate giant. A person, basically — just a giant one. When companies grow this large, their bloated figure becomes too greasy for any single man grasp. Profits become an indispensible and controlling part of the hive mind; the overall result of which is that the users become the commodity: They are no longer selling good software to satisfy the user — they’re producing software which maximizes profits and incorporates the user as a cog in their corporation. The consumers become the consumed.

By making their software most compatible with their own system (or completely incompatible with others), they are trying to enforce that people buy more from their empire. I came across a robots.net post from 2005, which has info related to what I’m trying to convey:

Nelson Bridwell sends us Microsoft news from the International Conference on Advanced Robotics in Seattle where Stewart Tansley revealed details of what could be Microsoft’s latest effort to assimilate the field of robotics. Apparently this is now part of a larger strategy to create more University level students accustomed to using and developing proprietary software. “They have decided that the best way to increase enrollment is to work with universities to incorporate robotics and computer games into the computer science curriculum as class projects where students can exercise their technical skills.” The robotics and computer games would be developed using various Microsoft proprietary software tools instead of the currently preferred Open Source / Free software tools. Part of their plan is to develop robot platforms with hardware that runs Microsoft’s .net language natively and offering them at much lower prices than conventional robotics hardware…. read more

I buy Apple because with their ~3% markeshare I feel they’re still small enough to actually care about their customers, while still having the power of a company to produce big things for me. And I get that feeling every time they come up with a simple and elegant solution to my problems.

That’s not the feeling I get when I’m called by friends and family to come over and fix their Windows machines, going through dozens of helpless help files telling me to go this webpage when the problem is that the web connection isn’t working. I don’t want the same thing happening to the poor robots.

In summary; while I think the MSRobotics Studio might be good to increase interest in robotics, MS’ patents and copyrights might potentially limit the growth of your robot. I’d recommend looking into some of the other solutions that are already available before making any decisions.

• The Open Automaton Project (OAP)
• The Open Robot Control Software (OROCOS)
• The Rossum Project
• Communities & Other stuff
• Mindmakers.org — people building modular large scale A.I. systems
• DMOZ listing of robotics software

(Robot image credit: The Open Automaton Project)

No related posts.