Simulating LIDAR sensors

Prologue

Simulation of LIDAR sensors concerns the process of simulating the sensing acquisition process of laser-range-lidar sensors by a computer program.

The way in which these sensors operate is by "active" perception, in other words, by touching the surface at which they are directed. Touching is implicitly performed by casting rays of photos (or rays of sound) and detecting the collisions that occur along their direction.

Active sensors are used extensively in robotics as they can provide measurements that are invariant to lighting conditions as well as denser depth acquisitions compared to passive (camera) sensors.

Available resources

Generally, in order to simulate the behavior of an active sensor through software, the proper way to do it is by employing "ray tracing", as is very well known in Computer Graphics in order to realistically render 3D scenes.

In this post, i would like to share some resources that i found in the internet which provide this functionality as well as some relative papers that employ simulation of active sensing. I hope it would be constructive and useful for all those out there that wish to play with these concepts ;).

Software

• http://www.blensor.org/: Blender Sensor Simulation basically provides a blender add-on that can simulate various standard laser sensors and control various parameters. In my opinion, just superb!

• http://www.kitware.com/media/html/ASyntheticLIDARScannerForVTK.html: the author provides a well-explained description of the provided c++ classes that can be used in VTK.

Papers

• http://www.roboticsproceedings.org/rss05/p22.pdf: K. Lai, D. Fox, 3D Laser Scan Classification using Web Data and Domain Adaptation, Robotics: Science and Systems, 2009.

• http://gfx.cs.princeton.edu/pubs/Podolak_2006_APS/symmpaper.pdf: Podulak et al., A Planar Relfective Symmetry Transform for 3D Shapes, SIGGRAPH, 2006.

• https://diglib.eg.org/EG/DL/WS/3DOR/3DOR13/PDF/081-088.pdf: Sipiran et al., Large Scale Shape Retrieval using Simulated Range Images, Eurographics Workshop on 3D Object Retrieval: SHREC Contest, 2013.

Tutorials

• http://kos.informatik.uni-osnabrueck.de/icar2013/: Large Scale 3D Point Cloud Processing Tutorial, 2013. Organized by Andreas Nuchter and Thomas Wiemann.

I will try to keep this list updated as i come upon relevant material-resources. Any such proposal will be greatly appreciated :).

i-Diots

IDIOTS from BLR_VFX on Vimeo.

Article submission to EES (Elsevier Editorial System)

Having submitted articles to several different journals of elsevier through the EES site, it seems that submitting a latex-based manuscript can be quite burdensome.

In detail, i always had problems in compiling and building the entire manuscript including figures and bibliography. Finally, i arrived to successfully compile and build a prospective article hence i would like to share the recipe of success :). In detail:

* Make sure that you are compiling and building your manuscript locally in your computer without errors. Although you may be able to produce a readable pdf on your computer, EES will not produce a pdf unless there are no errors in the compilation phase.

* Following the previous rule, at least in my case, implied that i had to use .EPS figures.

* Uploading a separate bibliography file (.bib) is not sufficient for EES. You have to additionally upload the corresponding .bbl file that should be produced if you can build your article locally. Eventually, both the .bib file and .bbl file should be designated as "Manuscript" files when uploading.

* If you are only using the style files that come along with the elsarticle package, avoid uploading them again in your submission.

That should be all :).

Good luck.

Drone-assisted archeology (euronews)

... Archaeologist Olivier Feihl is making a 3D map of the dig using a drone equipped with a camera.

“It takes a photo every 1.5 or 2 metres to make sure the whole surface is covered, so that the photos will overlap, enabling us to measure the archaeological dig in 3D,” he tells euronews.

It’s a true revolution for archaeologists and helps them save a lot of time.

“Before we had this kind of technology, everything was done by hand,” says Sebastien Freudiger, who is also working on the project. “Each wall was drawn by hand, each layer was drawn by hand. Now, this new technology enables us to do all of that on a computer.”

It takes the drone just ten minutes to take all the snaps needed. They are then processed ...

(read the full article here)

A robot that flies like a bird

Plenty of robots can fly -- but none can fly like a real bird. That is, until Markus Fischer and his team at Festo built SmartBird, a large, lightweight robot, modeled on a seagull, that flies by flapping its wings. A soaring demo fresh from TEDGlobal 2011.

How Productivity and Impact Differ Across Computer Science Subareas

A recent article in "Communications of the ACM" that sketches the impact of different areas of Computer Science in community, with statistics over recent years.

Some computer science researchers believe different subareas within CS follow different publishing practices, so applying a single production criterion would be unfair to some areas. It is reasonable to believe the subarea of, say, theory follows different publishing practices from...

Full article

Prosthetic Arm by DARPA

A team of researchers at the Rehabilitation Institute of Chicago (RIC) demonstrated a type of peripheral interface called targeted muscle re-innervation (TMR). By rewiring nerves from amputated limbs, new interfaces allow for prosthetic control with existing muscles. Former Army Staff Sgt. Glen Lehman, injured in Iraq, recently demonstrated improved TMR technology. In this video, Lehman demonstrates simultaneous joint control of a prosthetic arm made possible by support from the RE-NET program.

Full story is available here.