The New BigDog Video

Loving the Machine

It’s so interesting how most people I’ve talked to (and me myself) feel an emotional response when BigDog is slipping or being kicked. I felt like giving it a hand to balance on the ice — although on second thought I’m sure he’d crush me with his frantic maneuvers. The version of BigDog in the video weighs 106Kg (235 lbs).

The speed and agility of the feet give the fellow such natural (”messy instincts”) appearance that it’s hard to keep in mind it’s a machine. Unbelievable.

The control system depends heavily on internal state, awareness of joint positions, acceleration et cetera — which enables intelligent, real-time control. I hope the technologies spread to consumer markets, rather than being exclusively used for military purposes (this is what the AIBO should’ve looked like).

BigDog has an on-board computer that controls locomotion, servos the legs and handles a wide variety of sensors. BigDog’s control system manages the dynamics of its behavior to keep it balanced, steer, navigate, and regulate energetics as conditions vary. Sensors for locomotion include joint position, joint force, ground contact, ground load, a laser gyroscope, and a stereo vision system. Other sensors focus on the internal state of BigDog, monitoring the hydraulic pressure, oil temperature, engine temperature, rpm, battery charge and others. [BostonDynamics]

Boston Dynamics are also the creators of RiSE — an insect-like hexapod robot that climbs; and whose video demo ranked 10th in the most popular robot videos. A hat tip to the Boston Dynamics team for their achievements.

Links & References

• Boston Dynamics’ official site
• Hat tip for infosharing to Think Artificial reader Alebotta.

The Absolut Machines

By visiting Absolut Machines you’ll eventually find yourself on a page with two live videofeeds, presented in an old-school, gray window system. One of the machines is placed in Stockholm, Sweden and the other in New York City. The machines at these locations are music-composing AI systems you can interact with to augment the music they generate.

The interaction sessions are recorded and you can get a compressed quicktime video of your visit sent via email or download it from the “Gallery” tab which lists all recent videos.

Think Artificial VIP Access
Dearly devoted Think Artificial readers have been allotted VIP codes that allow cutting to the front of the line to interact with the machines. Leave a comment on this entry and I’ll mail it to the address you enter in the comment form. Alternatively you can contact me directly. Note that there’s a limited supply of codes and they’ll be distributed on a first come first served basis.

Obligatory disclaimer: To participate in this giveaway you must be at least 21 years of age.

Absolut Choir

The Absolut Choir is a system composed of speech synthesizers implemented in the physical form of 10 robotic characters. Each of the machines, or choir members, has a unique voice ranging from women, to tenors and sopranos. A “mother character” virtually conducts by synchronizing and distributing sounds to the other members, each of which contains a Linux box for processing and a speaker.

As the Choir starts singing, the user may input words to the machine. As the machine receives the words, it immediately uses them to generate a musical composition and lyrics. The robotic choir follows the lead of its human partner, and with the help of generative algorithms, the machine engenders a melody, tempo, dynamics, timbre and lyrics inspired by the user-generated input. The composition is also infused with the machine’s current mood and from the most recently analyzed words input by previous users. A lot of short words with many consonants may result in a fast arpeggio-like song, while softer words may result in a slower composition. [Absolut Press Kit]

The sound feed was suffering from some technical difficulties when I tried the choir. But the video worked, and the choir was receiving my lyrics glorifying Think Artificial (I figured I’d attempt to create a themesong for us).

The video compilation I received afterwards was okay. But I discovered that the lyrics were (intentionally) rendered hieroglyphic by the choir, so it sadly doesn’t make the cut as our themesong.

Absolut Quartet

The Quartet is quite different from the Choir. The machines are three; the main one is a marimba which the system plays by shooting rubber balls into the air, aimed at the marimba keys it wants to hit — or multiple balls if the objective is to play a chord. It’s quite fun to look at.

The marimba rubber ball blaster implementation and design.

Overview of the Quartet under construction.

The second machine is a series of glasses which basically replicate the “finger on a wine glass trick”. The glasses are spun, each tuned to a various pitches, and small robotic fingers touch them to produce sounds. The third part of the installation is an automated percussion instrument.

And then there’s the fourth part, us – the users. At the beginning of a session the human user plays notes on a miniature piano. The melody played dictates what kind of music the Quartet will produce, or in other words, your input is the machines inspiration for a following 3 minute song.

The machines are brainchilds of Dan Paluska and Jeff Lieberman. Both of which attend at MIT and have many cool projects in their backpack that combine aesthetics, artificial intelligence, kinetic sculpting and robotics.

Looking Closer at Robotics in the Media

This project is not an academic foray into the realms of creative AI, but rather a project intended to be looked at in terms of aesthetics. That being said: The artificial creativity of the machines is very primitive. If we take for example how the Quartet works; the software takes the melody played by a human user and compares it to a pre-existing collection of songs. Once a similar match has been found the machine mixes the two together producing the ultimate outcome. What interested me more than the software implementation of creativity was the overall aesthetic appeal of the project. In addition to Jeff and Dan’s artwork, the media related to this campaign was superb (partly handled by Noise Marketing, creators of the Appleseed website).

When exploring how we are creating our world; augmenting our environment — it’s intriguing to zoom out of AI context: How we (humans) advertise and perceive products is environmental augmentation. The ultimate sentiment is to be aware of the augmentations. To study them. Be aware of their effect and purpose; and to adapt and further develop whatever it is we want to achieve.

When I saw AI-colored advertisements from a major company (a company that essentially has nothing to do with machines) I immediately wondered whether it gave an indication of the public appeal of robots in Western societies. Certainly, machines in general play a larger role in everyone’s lives than ever before; and the same can be said about robotics even though we’re still in very early stages of that development. When we note that Puma has been sporting robotic-prosthetic cyberpunk campaigns as well, I think we can at least safely venture that robotics are on the rise in terms of public interest.

In my aforementioned geek-out craze I told my girlfriend about the device, called the Audeo, who immediately identified the problem of the device saying a thought you don’t actually want the other person to hear. You’re on the phone with your boss when you suddenly hear the device blurt out “Are you never going to shut up about those damn TPS Reports!?“.

Good point. But the creators say the device can differentiate between things that you’re thinking, and things that you actually want to say. You have to think about using your voice for the device to pick up on it.

I’m sure that this ability is a beneficial byproduct of making the device a “collar” around your neck monitoring the nerves that control muscles of the larynx.

Our Head’s Too Messy, Go for the Neck

The device is not a brain interface worn on the head, so it stands to reason that (a) they are monitoring neural activity to the muscles that control speech (larynx/voicebox), and (b) by doing so it’s easier to detect things that you actually want to say, as opposed to what you’re casually thinking.

 

 

The larynx is innervated by branches of the vagus nerve on each side. Sensory innervation to the glottis and supraglottis is by the internal branch of the superior laryngeal nerve. The external branch of the superior laryngeal nerve innervates the cricothyroid muscle. Motor innervation to all other muscles of the larynx and sensory innervation to the subglottis is by the recurrent laryngeal nerve.

However, I’m sure we’ve all been in situations where we are on the verge of saying something, perhaps in an emotionally colored debate, but think twice and eventually say something less aggressive. In such a situation I’m sure the device could accidentally be triggered. So the user must make sure to be perfectly balanced, one with himself and the universe before using it for important conversations. At least for now.

Writing this I get the idea that this problem could be overcome with AI; natural language processing could detect potentially insulting sentences or harsh language. The user could then be prompted to verify whether he meant to say a particular sentence (whether this would introduce too much lag is another question).

 

 

Voiceless Phonecalls

The device, currently able to recognize 150 words, is under development by Ambient Corporation, co-founded by Micahel Callahan who demonstrates the device in the following video at the TI Developer Conference’08 by placing a “voiceless phonecall”.

For the past few decades, humans have increasingly been extending their intellectual capacity with the use of machines. An example is using mobile devices to retrieve knowledge on the fly —

making each device-wielding human more intellectually capable than one 20 years ago. But this a matter of perspective, and many only see future invasive devices as “extensions of intelligence” (e.g. neural-interfaced memory storage device) and everything else as tools.

Modern technology is starting to blur this line between intellectual extensions and tools. The “Smartest Person in the Room” project is one of these: Using the Audeo, a person thinks of a question —

the question is consequently sent to a web knowledge-application, the answer found and tunneled back out through the speakers. Question never audibly asked, yet answered. Quite brilliant.

 

 

 

 

Looking forward to monitoring the developments of this project, feeding my interest in machine interfaces right along Emotiv’s Epoc and Neurosky’s non-invasive neural interfaces.

Links & References

• New Scientist on the Audeo

Live Science reports on the project:

The use of the EEG data is a bit more complex [than the use of rapid eye movements]. Running it through a machine learning algorithm, we identified several patterns from a sample of the data set (both REM and non-REM events). We then associated preprogrammed robot behaviors to these patterns. Using the patterns like filters, we process the entire data set, letting the robot act out each behavior as each pattern surfaces in the signal. Periods of high activity (REM) where [sic] associated with dynamic behaviors (flying, scared, etc.) and low activity with more subtle ones (gesturing, looking around, etc.). The “behaviors” the robot demonstrates are some of the actions I might do (along with everyone else) in a dream.” [LiveScience]

And here’s a video of it, dancing away [Alt].
The project is the brainchild of Fernando Orellana and Brendan Burns, who used the equipment of The Albany Regional Sleep Disorder Center in New York to record the data.

A robot dancing your dreams. Can’t help but feel inspired by that quip.

I’ll start off by noting that there was not one, but two companies demoing brain-computer interfaces at GDC this year. Emotiv Systems & the EPOC neuroheadset, which we’ll be focusing on here, and NeuroSky. The latter is planning on selling their sensors and technologies to partners and will not be developing a specific headset on their own (they had a demo unit at GDC just to show the functionality of their systems).

What Are Brain-Computer Interfaces and the EPOC Neuroheadset?

If you’ve read Think Artificial before, you’re probably somewhat familiar with BCIs. Brain-computer interfaces. Devices that allow us to control machines using only our minds.

The key technology is called electroencephalography (EEG). A device monitors your brain’s electrical activity via sensors on your scalp. It’s been used for medical purposes for years — and the futuristic image on the side here depicts setup for a musical brainwave performance at the Deconism Gallery in 2003, for example. The audience of a concert hooked up to the EEG devices to affect music and lighting.

However, monitoring the waves is different from detecting their patterns and using them reliable “triggers”, like Emotiv Systems’ EPOC device and software does. For this to work, two things are essentially required: The user has to practice producing a repeating, recognizable pattern. But there’s always noise (because it takes practice to be able to visualize the same image, or sequence), so the second thing is that the software deciphering the electrical activity must learn to recognize trigger waves.

Naturally, EPOC is not an invasive kind of BCIs — you won’t need a drill and pliers to use it.

But you’ll still have to shave your head (just kidding).

The Emotiv EPOC Neuroheadset uses a set of sensors to tune into electric signals naturally produced by the brain to detect player thoughts, feelings and expression. It connects wirelessly with all game platforms from consoles to PCs. The Emotiv neuroheadset now makes it possible for games to be controlled and influenced by the player’s mind. [link]

Emotiv Systems have been working 4 years on R&D, and have come up with their commercially viable BCI — and at a remarkably low price considering its capabilities and that this is the first time such technology hits the market for general consumers. Which brings us to its features.

What Emotiv’s Epoc Neuroheadset Can Do

Let’s start off with an easy-digest list of features expected to be bundled in the first release of EPOC:

• Wireless headset – 12 hour battery-life (playing time)
• Demo Game – Makes use of- and demonstrates the headset’s features
• Emortal – Access to an online hub that allows users to interact with photos and music using Epoc

The EPOC system is comprised of three main software components, each of which detects different kinds of brainwave activity.

The Affectiv suite can reportedly measure the emotional states of the user. Anger, fear, frustration. Emotiv puts forth the example that this could be used to have games increase or decrease the difficulty level depending on the player’s state of mind. The Cognitiv Suite is the control mechanism that allows players to control objects, and the Expressiv suite which measures and interpretes facial expressions of the user. The descriptions and demos are vivid, for example: You smile and thus your avatar smiles.

One of my earliest questions regarding EPOC was: can the system discern many patterns at the same time with any knowledge beforehand on what you’re trying to accomplish?

Most of what I’ve seen from their demos is task-and-turn based, where the player is moved between “phases”, each of which requires him to use one and only one specific action at a time. The Stonehenge Demo, for example, moves the player from stone to stone — but the player only applies one action to each stone (e.g. “rotate” or “lift”; not both).

Let’s elaborate. A user is inside Second Life and has created a plain box. My question is: Can the system handle rotating the box while the user is smiling/making the avatar smile? Or rotating the box, moving it a bit forward, then up — perhaps even rotating and lifting the box at the same time? Is all of this possible? Because if this were possible I’d be geeked out.

Keymap Your Brainwaves

I got mail yesterday. And I geeked out. The letter was from Emotiv reporting, amongst other things, more information on EmoKey — their software for mapping mental intention to keyboards (yes, meaning the Epoc headset will be connectable to virtually any application).

The descriptions almost sound surreal:

EmoKey Software – Use the Emotiv EPOC with your existing software

In our efforts to enabled our users, Emotiv has developed the EmoKey software application in conjunction with the Emotiv EPOC. EmoKey allows you to associate any of the Emotiv EPOC detections with keystrokes on your PC. EmoKey enables all of your existing PC software to be Emotiv EPOC compatible right out of the box! In practice, this means that you can link a “smile” detection to type the “smiley emoticon” in your chat application or link a thought, such as “rotate clockwise” to a series of keystrokes such as “a-w-d-s-a-w-d-s” to rotate your magic wand!

This appears to indicate that you can basically do any action, at any time, anywhere. Right? Well, almost. It’s not clear whether you can only “press” one button at a time (”a then w then d then s….”), or if you can press many buttons at the same time. It could even be a third case where you can press 3 buttons at a time — one from each detection suite (unlikely).

However, I can venture the guess that a feature of the EmoKey is to define a “virtual button” (if not, please spread the idea to Emotiv!). This could allow you to compose a series of virtual buttons. A specific thought could then be assigned to a series of them “ctrl+a, ctrl+w, …” instead of single physical buttons, thereby enabling you to press two buttons at the same time. Like enabling rotation of a something while smiling. This brings up the question of how many mappings there can be?

Now We Know How Santa Will Control His Robotic Reindeer Next Christmas

Regardless of questions and concerns, this is an incredible device that I would love to get my hands on. Granted, it is the first commercial edition, and we can expect that things may not run as smoothly as we hope. This video of someone trying the beta demo is actually the worst example I could find. I say actually because I don’t think that’s bad at all. The upload date & specific demo indicates this was at GDC’08, which means it’s probably a first time user.

Looking forward to seeing and hearing more about Epoc. In a recently featured article I talked about what great virtual reality gear could be; an important part being a non-invasive EEG device like Epoc for movement control.

Imagine making your avatar smile. Not via keystrokes. Simply by smiling yourself, walking towards that interesting monolith in the distance by seeing it happen in your mind.

Links & References

• REGEN3 – Image ©Regenerative Brainwave Music (DECONism Gallery)
• Emotiv EPOC Neuroheadset Images ©Emotiv Systems
• Another Epoc demo video at GDC’08
• A third company developing a BCI: OCZ Technology’s press release.

I’ve been waiting for this little wonder box too long that I had forgot about it… until yesterday… i went to apple’s website and it was there, ready to be ordered?! I just couldn’t resist, so I went to the italian web apple store to see if it was there too… it is!! I immediately called them and ordered one.

The bad news is it won’t ship till february… I hope not the end of february, but anyway, this gives me time to convert my existing divx library + all my original dvds to the iTunes format. And boy, if it takes time! I estimate ending the conversion in 1 to 2 months… just in time for the baby to knock at my door.

Livecams Online For Free – video powered by Metacafe

Matrix – For Real…joe Eigo – video powered by Metacafe