Interview in WIRED on the rise of AI

How real are the robots in Spielberg’s Extant? (Wired UK)

“Wired.co.uk speaks with the experts to see how close we are to the human-level artificial intelligences seen in the sci-fi series, and if we have anything to fear from their development.”

click the link below for the full interview with me and Murray Shanahan from Imperial College London:

http://www.wired.co.uk/news/archive/2014-07/18/how-real-are-extants-robots

Posted in Uncategorized

Why compliance?

Much of the world we interact with is solid and inflexible. We manage to navigate this environment without damage because our own bodies are tractable and ‘bendy’ – rigid bodies in a rigid environment is a recipe for damage. Common problems include vibration, backlash, unrecoverable hardware configurations leading to singularities or lockups.

Adding compliance to a mechanical system is one way to reduce or eliminate these problems – it also makes them safer for humans, and allows the system to adapt itself to external conditions. Compliant robots are more robust, safer, and human-friendly than their rigid counterparts.

Compliant design can be roughly broken down into two streams – passive vs active. Active compliance requires force-sensing elements, which then allow us to adjust the behaviour of the system depending on the external forces being sensed. This gives us a lot of control over the system dynamics, and allows for on-the-fly automatic tuning to adapt to different conditions. The major downside is that it works only when powered on, and electrical faults or current spiking can create serious problems with the potential to result in injury for humans, or damage for the hardware. Additionally, we must anticipate where force sensing is required, and depending on the joint or actuator type, it may not always be trivial to obtain accurate measurements or disambiguate between internal or driving forces, and unexpected external impedance.

Some examples of active impedance control in robotics and rehabilitative devices: The Barrett arm, Baxter, Lopes

Conversely, passive compliance involves building compliant elements into the actuation of the robot itself. This generally comprises the use of soft mechanical components, like springs and pneumatic elements, which do not require power to give them elasticity. We can also take advantage of the energy-storage capacity of compliant systems, which make them more efficient than purely motor-driven elements. Additionally, passive compliant mechanisms are much safer than active ones, however due to their lack of rigidity they are notoriously difficult to control, and cannot be tuned as quickly or readily as active approaches.

We can eliminate some of the issues with passive compliance while retaining their most useful features by using parallel methods of actuation. For example, a motor linked to a rigid body with springs (PEA/SEA), or using parallel springs with different force constants to provide a customized force curve for a particular joint or linkage. Parallel actuation is usually not considered when building robots, as it is seen as redundant and not cost-effective, however the advantages are several. Not only can we link several types of actuator to get the best elements of each, we can improve the sensing and force bandwidth by using one mechanism at eg. low speeds, high power, and another which responds quickly but with low force output. That said, getting multiple elements with different characteristics, response speeds and linearity to play nicely with each other is not necessarily trivial – at my current job we’ve been working on tuning one particular type of hybrid system for some months now. Nevertheless, hybrid actuation schemes have huge potential and should, in my opinion, be far more common than they currently are in robots, particularly those designed for HRI.

Additional reading: a nice talk about the compliant elements of COMAN from Nikos Tsagarakis of IIT

Posted in Uncategorized

Interview in the Guardian on Socibot

Our desktop robot, Socibot, is getting a lot more attention now we changed up the head design to make it more human-like. I was interviewed by Oliver Wainwright about the capabilities of the robot, and what it’s like to have in the office:

Capable of mimicking human expressions and emotions, the SociBot is designed to bring a human touch to teleconferencing – or to imitate your friends. But will it just creep you out?

SociBot: the ‘social robot’ that knows how you feel

Posted in Uncategorized

Interview on FoxNY on RoboThespian

The rapid progress of artificial intelligence

Meet the humanoid robot called Robothespian. He is designed to interact with people even through Skype. Created by England-based Engineered Arts, the Robothespian runs on algorithms and codes, or a form of artificial intelligence.

Posted in Uncategorized

Humanoids 2013

1395796_10202370535608499_1818672461_n
554966_677563018930004_745599747_n

1391584_675786832440956_1489487198_n
996633_677563292263310_1545672976_n

Posted in Uncategorized

ICRA 2013

262596_10201178344164458_1448226039_n

IMG_3648

IMG_3647

IMG_3644

IMG_3638

Posted in Uncategorized

news and changes

Haven’t updated this in a long while! Long story short, I applied for a job in the UK and after about five months of faffing around with bureaucracy, have finally been granted a work visa. So, what’s up? I’m now part of the R&D team for Engineered Arts, an engineering contracting company that specialises in the creation of humanoid robots for research and entertainment purposes.

image

Basically I’ll be helping to make these guys a little bit more human. One of the challenges with biomimetic humanoids is to create machines that are sufficiently lifelike to inspire natural responses, without making them creepy as all hell. It’s a narrow line to walk! Right now we’re hoping to collaborate on an FP7 grant which aims to parameterize the uncanny valley. Fingers crossed for the next judging round …

 

Posted in Uncategorized

quadrocopter-flying at the DLR

We put our polarisation sensors on board the DLR test-quadrocopter and took it for a spin …

quadro_main

quadro4

Posted in Uncategorized

The bee-eye sky

well, not really! only part of the bee eye sees the sky even remotely like this. But it’s still pretty cool.

Instead of the cumbersome mirror/camera set up we had prior, we started using four small cameras all aligned and pointed at the sky (at least three polariser orientations are required to successfully recreate the polarisation characteristics of a visual scene. Previously, we rotated the camera, which had a linear polariser inserted between the lens and sensor. Four cameras, each with their own polariser, allows us to generate the images much faster, reducing error). This system is much more lightweight; we can use it to reconstruct insect paths through canopies or under open skies, and get a much clearer idea of what the insect is seeing as it flies.

Posted in Uncategorized

Filming the sky

I spent large portions of today outside filming the sky with a dual-camera rig we set up to try and imitate the polarisation sensitive portion of the insect eye. To get the full sky hemisphere at low resolution, we used a normal (albeit small) camera pointed at a curved mirror – in fact, we used two cameras and two mirrors, because finding a sufficiently small camera that had good UV transmission as well as good visible spectral transmission was prohibitively expensive.

IMG_1921Some background on polarisation as it relates to insect vision: Many insects possess the ability to detect the directional component of light, what we term its polarisation properties. The dorsal rim area, in particular, is strongly sensitive to the direction of polarisation (the phase) of incoming light, and is thought to be used for navigational purposes.

The compound eye of an insect is made up of many ommatidia, which include a lens, cornea, and photoreceptor cells. Each ommatidium has its own ‘preferred’ orientation – this is the direction of polarisation which it responds most strongly to. By comparing signals coming from ommatidia with different preferred orientations, but which view the same area of the sky, the insect can create a ‘map’ of the polarisation properties of the incoming light.

When light originating from the sun hits our atmosphere, it results in a distinctive ‘pattern’ of different polarisation orientations and magnitudes. This pattern can be used as a sun-compass, giving us the ability to detect the position of the sun even on overcast days, or where the sun is invisible due to environmental features, eg. under a forest canopy. Other environment features, like water, or particular kinds of vegetation, can also be distinguished by their polarisation properties.

 

Posted in Uncategorized