Dear community,

We would like to inform you about our participation in the DARPA Subterranean Challenge as Team CERBERUS (University of Nevada, Reno, ETH Zurich, Sierra Nevada Corporation, University of California, Berkeley, Flyability). 

Please find out more for us through by watching our concept video:

And by vising our website: https://www.subt-cerberus.org/

We look forward to the novel research of this Challenge! 

Best,
Kostas

Dear community,

We would like to share the following result on aerial robotic navigation through obscurants. The test took place in a smoke-filled machine-shop environment.

Best,

Kostas

This is to share results on radiation source localization in GPS-denied environments using autonomous aerial robots that estimate the nuclear (Cs-137 in the experiments) source location and intensity, while simultaneously planning for informative measurements. Smart selection of measurement points is critical as the robot has to dwell for sufficient time to allow the Poisson statistics to derive a confident measurement. The robot is equipped with CsI(Tl) scintillator with silicon photomultiplier and miniaturized electronics for counting and spectroscopy. The first video is explenatory of our approach while the following is taking place in a more realistic environment: 

Additional details:

* Autopilot: PX4

* Position Controller: Linear Model Predictive Control running on an INTEL NUC (ROS)

* Localization: Visual-Inertial 

* Path Planning: Explicit function to optimize for distribution of measurements

In these two videos you will see some recent results on autonomous exploration of tunnels in visually-degraded (dark) conditions. Both localization and mapping, as well as path planning and control are handled completely autonomously. 

Best,

Kostas

Dear community members,

During this ICRA we presented our work on "Uncertainty-aware Receding Horizon Exploration and Mapping using Aerial Robots".

This work is now accompanied with an open sourced ROS toolbox available at: https://github.com/unr-arl/rhem_planner 

These videos present the planner in operation:

In case you find it interesting, we would be happy to support you in order to fully integrate it in your research.

Best,
Kostas

http://www.autonomousrobotslab.com

Dear community members,

This is very preliminary work on aerial robotic radiation detection in real-time and in a GPS-denied manner. The robot localizes based on visual-inertial data and integrates a deferentially installed 2-radiation detectors solution to exploit their polarity characteristics. Most of the research is ahead but we feel the topic is exciting and would like to communicate it to all of you. In this link you can get updates of our relevant work: http://www.autonomousrobotslab.com/robotics-for-nuclear-sites.html ;

As shown, it is not only referring to the radiation detection itself but also to the relevant challenges involved such as the autonomous navigation in visually-degraded environments:

Best,

Kostas

We would like to share two of our recent and ongoing results related to autonomous exploration and mapping in darkness. As opposed to approaches that rely on localization based on LiDAR systems, the videos below rely on visual-inertial localization with LEDs illuminating the scene appropriately. For the first case, visible-spectrum cameras are used, while for the second near-infrared cameras are employed. We believe that operation in darkness is an important feature for aerial robotics and we would like to see more in that direction. 

For more please visit: http://www.autonomousrobotslab.com/ 

Hi all, 

We recently released this video on autonomous exploration and inspection. Have a look!

Summary: 

The robot employs its visual-inertial navigation system in order to localize itself in the environment and simultaneously map it and 3D reconstruct it. Based on the newly proposed "Receding Horizon Next-Best-View Planner" the robot computes the next best step for efficient volumetric-exploration of unknown spaces. This is achieved by predicting a sequence of next-best-views via sampling-based methods and information gain approaches, executing only the first step and then repeating the whole process in a receding horizon fashion. Once full volumetric-exploration has been achieved, the robot starts focusing on the surface reconstruction of objects of interest in the environment.

Paper:

A. Bircher, M. Kamel, K. Alexis, H. Oleynikova, R. Siegwart, "Receding Horizon "Next-Best-View" Planner for 3D Exploration", IEEE International Conference on Robotics and Automation 2016 (ICRA 2016), Stockholm, Sweden 

The paper is accepted and will be presented at ICRA this year. 

Code: 

The code is to be open sourced and will be available once the paper is presented but also before. Send us an e-mail if you have interest already and we will update you once it is out. 

Previous relevant work (but at the problem of optimized inspection while known a geometrical model of the structure):

A. Bircher, K. Alexis, M. Burri, P. Oettershagen, S. Omari, T. Mantel, R. Siegwart, "Structural Inspection Path Planning via Iterative Viewpoint Resampling with Application to Aerial Robotics", IEEE International Conference on Robotics & Automation, May 26-30, 2015 (ICRA 2015), Seattle, Washington, USA .

Code:

https://github.com/ethz-asl/StructuralInspectionPlanner

Hope you like it!

Kostas Alexis

http://www.kostasalexis.com/

Dear members of the drones community,

For one more year... the time has come! 

The Autonomous Systems Lab at ETH Zurich just released the "Autonomous Christmas Lab 2014" video! Watch and share! Robots are indeed excellent Christmas companions! 

Happy holidays to all of you!

You can find product details at: http://www.01mech.com/supermodified

I copy the following from the site: