January 9th, 2010
To address the miniscule details in front: as a C++ programmer - we count from zero - the decade is 2000-2009 for me. But lets get to the point: as this decade has closed, analysis about what has defined the decade was widespread as usual. A broad consensus has found that this was a “lost decade” - the dot.com and financial market crashes, the war on terror and the anthropogenic climate change topics dominating a rather bleak impression. It was argued that the decade was “lost” mainly because leaders and societies could not find an effective manner to tackle these challenges.
Read the rest of this entry »
Posted in Society dynamics (near future) | No Comments »
November 6th, 2008
Nine months ago, I wrote some comment about the turn of the tide concerning the direction western civil societies are heading and the primaries of the presidential elections in the United States of America as a part of that. I wrote that “something is happening that indicates at least some rebound, it seems as if the people begin to wake up” and “Of course, politics is full of false promises, often plain out lies. But what these candidates dare to promise, where their focus lies, which political spectrum they address tells a lot at least about their voters”. This week we have seen that growing tide becoming a surge.
Read the rest of this entry »
Posted in Uncategorized | No Comments »
March 2nd, 2008
During the past months, a student under my supervision developed a really exciting visualization of POMDP policies, quite an improvement compared to the old one. The new visualization is tailored to show more than three dimensions of the policy at the same time. This method will improve insight into complex POMDP policies around important belief points in the high dimensional space. It also has a certain kind of beauty:
Read the rest of this entry »
Posted in Research | No Comments »
February 27th, 2008
Yesterday for the assessment of the large SFB588 “Special Research Area - Humanoid Robots” of the German Research Foundation at our institute, I presented Albert in a demonstration of related projects (COGNIRON) and with it the person body tracking and probabilistic decision making. With a new AVCHD camcorder, I made new videos of this demonstration today, with both the whole scenario (100MB) as well as a live projected visualization of Alberts human body tracking (20MB), developed by colleagues.
Posted in Research | No Comments »
February 10th, 2008
These days, I am usually quite caught in a tunnel vision, focussed on the various lines of work concerning my main job. Apart from that, I follow the most important financial markets on their weekly roller coaster - but that’s basically it. The rest of the news sweeps by without any further, deeper consideration. Thus, an interesting phenomenon which seems to have appeared quite recently has evaded my view so far.
Read the rest of this entry »
Posted in Society dynamics (near future) | 1 Comment »
November 9th, 2007
POMDP decision making has been ported onto the service robot Albert since late summer, however videos were still lacking. Recently, I made new videos and a short sequence can be watched on the net (2 min, 60MB):
SRSR and autonomous Albert
Albert is able to act robustly even when facing very noisy sensor measurements. One reassurance question can be seen in the video, which means that Albert is able to reason if it is sure enough to have seen and interpreted the human body posture and understood the human utterance correctly. Albert is able to perform waiter duties continuously and I have made one hour long experiments with a continuously, autonomously performing Albert. There is no external control of Albert of any kind. The live-visualization of the thought processes is sent unidirectionally via WLAN.
Posted in Research | No Comments »
November 9th, 2007
1918 - 1923 - 1938 - 1990. And now this day stroke another time, 2007. Sadly, this time the pendulum shows a reverted course again. Well, it is obviously the zeitgeist, reenforced by a long and infamous tradition of authoritarian systems and their executive arms: Prussian secret police, secret state police (Gestapo) and the ministry of state security (Stasi). But this unparalleled surveillance of the whole population as mandated by the Vorratsdatenspeicherung law heralds a new era of suppression and the end of the age of a quite open society. This new, undesirable epoch, may come more slowly, more subtle than at other times, altering behavior and thinking year by year, but its progress seems inevitable.
Read the rest of this entry »
Posted in Society dynamics (near future) | No Comments »
July 21st, 2007
Pride has a delicate meaning in our modern, achiement-oriented society. Framed somewhere between hybris and appreciation, arrogance and accomplishement, the aspects pride may represent can be manifold in the eye of the beholder. Thus, I use it deliberately with an emphasis on the positive aspects: I am proud to have been awarded - as one of 2 recipients choosen from 260 graduates - the Award of the friends of the Research Center for Information Technologies for an outstanding thesis.
Read the rest of this entry »
Posted in Academics | No Comments »
July 21st, 2007
After I posted only POMDP visualizations some time ago, the focus is turning back to the real robot. During the preparations to get POMDP decision making working on our semi-humanoid robot Albert, I made extensive tests with mid-level controlling software on which I worked as a student and which I have now integrated into a new controller framework.
Read the rest of this entry »
Posted in Research | No Comments »
July 21st, 2007
Recently, at the end of the summer term 2007, I held my first three lectures, representing Prof. Dillmann in “Robotics II - Programming robots”. While I could start with my favorite topic - probabilistic decision making - it was still quite a challenge to explain the subject comprehensible and yet thorough to about 50 students. Preparation took far more time than I anticipated and talking for one and a half hours is quite exhausting. While I have still to improve in using the chalkboard for improvised drawing, the lectures went mostly well. Generally, it is a great opportunity to train detailled presentation of complex topics in front of a large crowd.
Read the rest of this entry »
Posted in Academics | No Comments »