Kinect initial test

I've acquired a Kinect sensor, and have done some initial tests on it to see what the depth maps look like.



The quality looks very good, and it appears that there is relatively little noise in the range measurements. As I had expected, this sensor definitely doesn't work outdoors, but under all the indoor situations in which I've tried so far it seems to work well. Just like with stereo vision there is a minimum ranging distance, but this is quite short.

The sensor itself is quite bulky, and this is probably because it contains a motor used to tilt it up and down. If I use this as a replacement for the existing dual stereo camera head on the GROK2 robot then I'll need to design a fairly industrial mounting for it.

With the arrival of sensors like this the quest for good perception on mobile robots operating indoors and which don't require ranging distances more than a few metres is reaching its conclusion, and we are almost in a position where yet another of the necessary criteria required for a new robotics industry to begin can be ticked off of the list. No longer is it the case that to obtain good short distance range data you need to buy laser scanners costing thousands of dollars - like the nodding laser on the PR2 robot. It's also to be expected that other companies will try to produce similar sensors and that their cost will come down still further, as happens with all new consumer gadgets.

The next step is to try using this sensor for SLAM, and I think that Willow Garage are already in the process of doing that. The prospects for being able to build and navigate within 3D models of indoor environments over the next couple of years are better than they ever have been.

A reward for filth

The economist Gregory Clark gives an interesting talk proposing a particular idea about the relationship between population, technology, mortality and income (or wages).





The model has some counter-intuitive outcomes, but might help to explain a few things which have puzzled me for a long time, such as why within modern society so much vitriol is directed towards teenage mothers, whereas only a few generations earlier teenage motherhood was normal. This model would predict that teenage mothers shift the birth graph upwards and therefore the equilibrium point on the population/income graph shifts backwards, reducing living standards.

The model also challenges my preconceptions, in that improvements in technology might not always result in improvements in living standards (actually the reverse, if mortality curves remain unaffected). Typically though changes in technology may also have implications for mortality, so the relationship is a complicated one.

A theory of everything - in colour

The history of washing

A documentary about one of the greatest pieces of home automation: the washing machine. Washing machines are an early example of robotics in the home, and probably save more hours of tedious manual labour than any other home appliance. It's only really within the last few decades that the process of washing clothes has become mostly automated, and I vaguely remember that in the 1970s my grandmother had something resembling the mechanical washing contraption depicted in the video.

A point cloud model of the author

As observed by the forward stereo camera of the GROK2 robot. This is a composite "mind's eye" view generated from observations in several different pan and tilt positions. There are some depth errors around the computer monitor, but the depth estimates for most of the scene look reasonable.

The points were exported in X3D format, then viewed using the MeshLab utility.





Also here is a point cloud view from the Rodney robot situated on the desktop. This uses the Minoru webcam which has a shorter baseline distance of 6cm, so the depth resolution is not as good as from GROK2. The original intention with the Rodney robot was just to be able to identify objects within a short range on the desktop, so longer ranges aren't very important anyway. The large black object on the right is the giraffe-like "neck" of the GROK2 robot.





Undoubtedly the quality of the imager on the Quickcam 9000 is much superior to that on the Minoru, especially under low artificial lighting conditions.

Sentience project update

Over the last month or so using a new dense stereo correspondence algorithm called ELAS I've made some quite substantial improvements to the vision systems of the Rodney and GROK2 robots, meaning that they can reliably see the structure of the environment around them under most conditions, probably with enough accuracy to enable a "functional vision" capability.  Whilst the depth resolution may not be as good as for active sensing methods, such as laser scanners or the Kinect, for purely passive vision this is quite a pleasing result.

Here is an example depth map using the ELAS algorithm, where disparities are colour coded. The different colours make it easier to see structure in the distance than is otherwise the case with a monochrome representation of depth.



Converting the disparities to ranges, then projecting them in 3D gives a reasonable point cloud model. Here a few stereo images taken from different pan and tilt angles are combined into a unified 3D model.



In this model you can see the desk with the Rodney robot on the left, a cup, the keyboard and screen, and a small Surveyor robot in front of Rodney.



Whilst this isn't quite at the same quality as the CMU hallway runs produced by Moravec in 2000-2002, it's getting into the same ballpark and with some ray modelling and projection of rays into an occupancy grid this would certainly help to resolve the longer distance features.

My last attempt to do this, using an orange juice carton as a subject, was in 2004/5, and although the overall approach was the same the results were not as good using cruder algorithms and camera calibration.  Also it should be noted that this progress actually has little to do with raw processing power, and much more to do with improvements in algorithms.  The ELAS algorithm is pretty fast, especially on low resolution images, and would have been usable on a computer of five years ago had it been available back then.

Data war

I was going to wait a while longer to see what happens before making any comment on the whole Wikileaks debacle which has been dominating the news within the last week or so, but it appears that some of my contemporaries seem to have a sufficiently irrational view on this that I might as well make an opinion known.

My opinion on Wikileaks itself up to this time has been that I thought they were doing important work in the public interest, helping to highlight corruption and misdeeds such as violations of international laws, human rights abuses and financial fraud.  If we are to have any hope that the world of the future will be a more civilised one than the world of the past then information of this kind - unpleasant or embarrassing though it may be - does need to be brought to public attention and not remain hidden from view or be permitted to be erased from the history books.  In a properly functioning democracy, this is the function of whistle blowers and investigative journalists.

For a summary of Wikileaks and the motivations behind it, there's an instructive speech given by Jacob Applebaum at the HOPE conference.



However, since the takedown of Wikileaks what has been described as a "data war" has ensued against the various companies involved, such as Paypal and Visa.  Foolishly, even people who ought to know better are apparently supporting this kind of vigilante action, and in the BBC article even a Wikileaks spokesperson seems to be at least tacitly supporting it.

First of all, voluntarily downloading any sort of botnet program onto your computer is a really bad idea.  No really, I can hardly emphasise strongly enough how idiotic this is.  Are you sure that you know what that program contains?  Did you check the source code and compile it yourself, or is it just a binary?  Apart from attacking web sites - which is of course totally illegal - what other things might that program be doing on your computer?  Remember that malware developers are opportunists, and sometimes actively exploit news events.  Do you have any credit or debit cards, and if so is it really in your rational self-interest to attack companies which facilitate payments of this kind?  Is attacking web sites likely to help or hinder Wikileaks, or whistle blowers in general?

So to anyone who is tempted to join "operation payback", or encourage others to join, I would seriously advise you to think more carefully.  Boycott or civil disobedience campaigns are a better way of expressing displeasure than engaging in overt criminality.  I think it's a major error of judgement of anyone claiming to represent Wikileaks not to unambiguously condemn DDOS attacks on web sites at the earliest available opportunity.

Going beyond the current headlines, and regardless of what may or may not be the fate of Wikileaks, occasional and sometimes large scale leaks of information from companies or governments - sometimes by accident rather than by design - is likely to remain as a feature of the landscape.  This is really just a function of the increasing density of data storage and internet bandwidth.  Probably there will be some attempt to improve technical security and security procedures within governments and other large organisations, but technical measures alone won't prevent a determined insider from leaking information, although they may influence the volume of such disclosures.

One way in which organisations can respond to the threat of leaks is simply to become more transparent, with public disclosure as the default behavior whilst only guarding the most sensitive information.  An early example is data.gov.uk.  From the leaked diplomatic correspondences it's pretty clear that there exists a needless culture of secrecy, and that many of the documents have been gratuitously classified more out of habit than due to any good reason.

Also see 2600 Magazine condemns DOS attacks.