I’ve just found an interesting article with a link to the farewell talk of the Ugobe CTO John Sosoka at Stanford university: http://www.gorobotics.net/the-news/latest-news/the-rise-and-fall-of-pleo-a-fairwell-lecture-by-john-sosoka-former-cto-of-ugobe
There have been several hints earlier this year, but now it is official. Ugobe, the maker of the robotic pet dinosour Pleo has just filed chapter 7 bankruptcy.
After Sony had stopped the production of the AIBO pet robot in 2006, I was quite excited when I heard about the release of Pleo, a new programmable robot, selling for less than a sixth of the price of an AIBO. Pleo can percieve its environment using a camera, two microphones, multiple touch sensors, tilt and shake sensors as well as an infrared sensor and it moves in a slow but very natural way. In terms of sensors and actuators, Pleo was not much worse equipped than its “canine” predecessor. Unfortunately, Pleo is lacking wireless connectivity, but there is a solution for this:
I really wonder, if there will be any decent and commercially profitable pet/entertainment robots anytime soon… AIBO was canceled out of profitablity issues, Pleo ended in a bankruptcy and many other
A software development kit (PDK) had been promised by Ugobe to the developer and research community, but unfortunately it has never exceeded prototype state. The only successful entertainment robots I have seen so far, are essentially children’s toys with very low end sensors and actuators selling at a very low price tag, such as the RoboSapien and its kin.
I am quite sure, that the reason is not that adults do not like robots - when I conduct experiments with AIBO, most participants are fascinated by the little robotic dog - but I doubt, this is enough to make people actually buy a robot which is essentially an expensive toy. For the adult mass-market [which (unfortunately )does not primarily consist of geeks and early adopters] a robot needs to be not only cute and fascinating but, above all, it needs to be useful. As useful functionality is still quite hard to implement in a small and affordable robot (How about crossbreeding AIBO and Roomba? ), I think we’ll still have to wait a couple of years for the first entertainment robot to become a commercial success.
Anyway, good bye, little Pleo!
I first saw Nissan’s Pivo 2 concept car at the Tokyo Motor Show last year. It has quite a number of interesting features, such as a cabin, that can rotate 360 degrees and tires, which can move independently from each other, which allow it to move sideways and facilitate parallel parking.
However, the most interesting thing from my point of view was their robotic co-driver. It monitors the driver’s mental state based on facial expressions and voice patterns, advises about free parking spaces and speed limits and helps you navigate. - and the best thing about it: It won’t tell you things like “Do you have to speed like an idiot???", “You should have turned left here!", “If you keep driving like a snail, we’ll never be in time"… so when can I preorder?
You can find more information on the Nissan website:
I just found the archive of webcasts from the plenary talks at the IEEE RO-MAN ‘07. I attended the conference this year and I particularly enjoyed the talk “How Robots May Evolve Their Own Language?” by Luc Steels. Here is the link, just in case someone would like to watch:
Not exactly robotics-related but anyway quite interesting: A video from the SIGGRAPH 2007 by Shai Avidan and Ariel Shamir on context sensitive image resizing, that is, automatically resizing an image in a way, that the “interesting” parts of its contents remain untouched instead of just enlarging or shrinking all parts of the image in the same way.
More details can be found on this website: http://www.faculty.idc.ac.il/arik/
….NAO, the humanoid made by Aldebaran Robotics!
I admit, I am quite surprised about the decision to make it a standard platform for the “not-any-longer-four-legged” league, although NAO was my personal favourite, because of it’s applicabiliy to HRI and a wide field of other tasks.
I guess, the decision to change platforms from four-legged AIBOs to humanoids will affect the teams in the standard platform league (which is the new name of the four-legged-league) a lot - judging from the games I’ve seen from the humanoid league, so far - but I am really looking forward to watching the first standard platform league games and seeing how the teams get along with the new robot.
I was browsing the RoboCup 4-legged-league mailinglist for the results of the selection of the AIBO successor when I found this video taken by the RoboCup team of Boðaziçi University.
I think, the kitten would make a good successor for AIBO. Its walking skills are exceptional and it is quite good at dribbling, however, it might need some software modifications to enhance its goal-keeping abilities.
The selection results of the real Aibo successor have not been anounced so far, but I guess, it is only a matter of days now.
From 1st to 8th of July, the Robocup world championship took place in Atlanta. One of the highlights, apart from the games themselves, was the presentation of potential successors for the discontinued Sony AIBO in the four-legged-league. The final decision will be announced on 31st of July.
This is what I could find out about the candidates, so far:
The following four robots have been presented:
Another robot, that was presented during the worldcup was the humanoid “Nao” by the french company Aldebaran Robotics. Some videos can be found on their homepage, too. This one looks very promising to me because it may have a broad range of applications also outside of RoboCup. The creators of the robot did not forget about expressivity and, moreover, the robot posesses stereo microphones and sound output for verbal interaction. While some of the presented robots seem to be 100% geared toward the robot soccer task, such as the robot dog proposed by the university of Darmstadt, which neither has a single LED in its “face” nor posesses microphones or a speaker but excels by its walking and video capturing capabilities, “Nao” might become the next real “AIBO Successor” being an allround platform for research and entertainment.
I just found this project called Julius, based at Kyoto University which offers a free speech recognition engine as well as acoustic models for Japanese and English which seem to be compatible with the HTK (Hidden Markov Model Toolkit). The English models must not be used for commercial purposes, though. Anyway, for those working in research, it might be helpful.
The project is explained in this paper, published in the proceedings of the conference EUROSPEECH 2001:
I have just started this blog to have a place for storing interesting HRI related information, I come across while researching literature for my thesis or just browsing the web and reading in my favourite online forums. I am particularly interested in multimodal interaction between a human and a robot so probably most of the stuff, you are going to find here, will be in the one or other way related to that field.
I am a Ph.D. student at the National Institute of Informatics in Tokyo doing research in the field of Human-Robot-Interaction. Sometimes I come across interesting information, papers or links, that I would like to share, so I started this blog.
If you are interested, please also have a look at my publications.
|<< <||> >>|