Inspired by the work that was done at the ACT, I continued working on optical flow landing at TU Delft. Today Bio & Bio published my article on a new theory that allows drones to see distances with a single camera. It shows that drones approaching an object with an insect-inspired vision strategy become unstable at a specific distance from the object. Turning this weakness into a strength, drones can actually use the timely detection of that instability to estimate distance. The new theory will enable further miniaturization of autonomous drones and provides a new hypothesis on flying insect behavior. Video: https://www.youtube.com/watch?v=Fm7SMJp8EA4&feature=youtu.be Article: http://iopscience.iop.org/article/10.1088/1748-3190/11/1/016004

Nice discussion about the singularity. Made me think of drinking coffee with Luis... It raises some issues such as the necessity of embodiment, etc.

Embodiment has some obvious advantages. For example, in the vision domain many hard problems become easy when you have a body with which you can take actions (like looking at an object you don't immediately recognize from a different angle) - a point already made by researchers such as Aloimonos.and Ballard in the end 80s / beginning 90s. However, embodiment goes further than gathering information and "mental" recognition. In this respect, the evolutionary robotics work by for example Beer is interesting, where an agent discriminates between diamonds and circles by avoiding one and catching the other, without there being a clear "moment" in which the recognition takes place. "Recognition" is a behavioral property there, for which embodiment is obviously important. With embodiment the effort for recognizing an object behaviorally can be divided between the brain and the body, resulting in less computation for the brain. Also the article "Behavioural Categorisation: Behaviour makes up for bad vision" is interesting in this respect. In the field of embodied cognitive science, some say that recognition is constituted by the activation of sensorimotor correlations. I wonder to which extent this is true, and if it is valid for extremely simple creatures to more advanced ones, but it is an interesting idea nonetheless. This being said, if "embodiment" implies having a physical body, then I would argue that it is not a necessary requirement for intelligence. "Situatedness", being able to take (virtual or real) "actions" that influence the "inputs", may be.

Nice application of swarm robotics for building various structures. See the video at: http://www.youtube.com/watch?v=LFwk303p0zY Their compilation process of a desired structure into local behavior rules seems of most interest.

TU Delft researchers have developed the DelFly Explorer, the world's first Micro Air Vehicle with flapping wings that can avoid obstacles by itself. The uniqueness of this achievement lies in the DelFly Explorer's very low weight (20 grams, i.e. a few sheets of paper), and this opens up new possible applications for both smaller and larger MAVs.

The most tiny autopilot in the world was introduced yesterday by the Technical University of Delft, The Netherlands. Named Lisa/s, she comes in at the scales with just a mere 1.9 grams, the 2*2 cm board has everything that a multirotor needs. Among the sensors are a 3 axle gyroscope, compass, barometer and a gps module.

I wanted to post this one, as plant-inspired robotic design and algorithms are a new kid on the block, and then noticed that some ACT-members may already be aware of this special issue...

The Robobee takes off without guide wires! It is still powered via a wire, and the control is done with the help of a VICON system and on an external computer, but this still is an amazing feat!

"Robohub is an online platform that brings together leading communicators in robotics research, start-ups, business, and education from around the world." What is nice about robohub is that they generally give a look behind the scenes of interesting robotic studies.

Today we released the Astro Drone app. People that have the Parrot AR drone can freely download the game. While they fly their drone in the real world, they are trying to dock to the ISS in the virtual world. But the app is more than a game. Players can choose to participate in a scientific crowd sourcing experiment that aims to improve autonomous capabilities of space probes, such as landing, obstacle avoidance, and docking. If participating, the app extracts visually salient features from the images made by the drone's camera. The features are then combined with the estimates of the drone's state and uploaded. The data is then used in a research aiming to improve robot navigation.

Recently, many competitions in the computer vision domain have been won by huge convolutional networks. In the image net competition, the convolutional network approach halves the error from ~30% to ~15%! Key changes that make this happen: weight-sharing to reduce the search space, and training with a massive GPU approach. (See also the work at IDSIA: http://www.idsia.ch/~juergen/vision.html) This should please Francisco :)

Nice article on how bumblebees optimize their path from flower to flower.

We discussed a few times about whether it is possible to determine the quality of a paper by extracting visual features from the paper and then learning a mapping to a measure of quality such as the number of citations etc. This paper circulated at CVPR 2010, and does exactly that, mapping visual features to estimate whether it has been accepted for the main conference or the workshops.

Autonomous sailing boat planned for setting a new record

"... scientists at the Technical University of Delft in the Netherlands yesterday published their findings that they have indeed found the Majorana particle. The announcement on the university website provides both a summary of the ..." Also see: http://tudelft.nl/en/current/latest-news/article/detail/nanowetenschappers-vinden-langgezocht-majorana-deeltje/

Much faster than the Fast Fourier Transform if the signal's frequency content is sparse, which is commonly the case for many real signals.

Amazing conversation...

Great introduction to the Bayesian view on the workings of the brain, which has been a successful view in explaining many psychological phenomena, visual illusions, etc. One of the possible criticisms on this view is that it neatly separates perception and action.

At the moment still in the phase of running in simulation, but who knows when the first robotic experiments will be.