Braden done!

Dr. Braden Stenning successfully defended his PhD thesis on Monday, November 19.  Prof. Alonzo Kelly, a renowned expert on path planning from Carnegie Mellon University's Robotics Institute, was the external examiner.  Braden will be staying on at UTIAS as a postdoc working on a cliff-descending robot concept for field geology applications.  Congrats and great work, Braden!

Robots Rule!

This weekend was the annual Robots Rule event at the Ontario Science Centre in Toronto.  We brought our Clearpath Husky A200 robot and demonstrated stereo visual teach and repeat throughout the event.  We also took the Husky down to visit the impressive Curiosity full-size mockup in the new space exploration exhibit, which is worth seeing and only on until January 1, 2013. 

VT&R Strikes Again!

Members of ASRL were at the Canadian Space Agency this week to help integrate our stereo visual teach and repeat (VT&R) algorithm with one of CSA's vehicles under the Robot Operating System (ROS).  This represents a big step forwards as CSA's new plan is provide ROS interfaces to all their new rovers in order to more easily share code and collaborate with external partners, including universities, in the future.  We had VT&R up and running in less than a day of integration, showing the strength of this approach.

Critical Mass

The Canadian Space Agency made a big announcement this week to let the world know about several planetary rover prototypes it has built over the last three years.  This came out of a $110M stimulus package announced by the Canadian Federal Government in January 2009.  The official CSA webpage can be found here.  A CBC article on the rover announcement can be found here.  Our lab was involved in the Mars Exploration Science Rover (shown at right) as well as the Lunar Exploration Light Rover; we provided algorithms for visual odometry, sun sensor attitude determination, and stereo visual teach and repeat.

Starry Night

We're back from the Koffler Scientific Reserve at Joker's Hill having spent the entire night gathering a lidar-plus-star tracker dataset.  We were worried towards the start of the night as some clouds drifted past but the stars were out after that and cooperating nicely.  The next step is to process the dataset with our visual odometry algorithm to demonstrate full dark navigation.

Working hard, or hardly working?

We are at the Koffler Scientific Reserve at Joker's Hill (part of the University of Toronto), which is 25 km North of UTIAS near Newmarket.  Tonight we'll be gathering our first full-dark-navigation dataset (star tracker plus scanning lidar for in-the-dark visual odometry) and this venue is far enough out of the city that we can see the stars at night.  Of course, the pond, dock, gazebo, and beautiful surroundings don't hurt either.  It's going to be a long night but should be worth it.

We've been blogged!

Here's an IEEE Spectrum article about our recent experiment at the CSA's Mars Emulation Terrain.  They didn't quite get it all right...but you know what they say about publicity.

Living on the Edge

We've had a good week of testing on the Canadian Space Agency's Mars Emulation Terrain.  We conducted two long mission scenarios where we hunted for a methane source using a spectrometer.  We also used our spare Husky rover to do a little extra testing on some of the more extreme terrain in the test area.  In this photo the rover is fighting hard to track a path along a steep crosshill slope using our stereo visual teach and repeat algorithm.  It almost lost its balance here, but managed to hold on to complete the drive.  We head back to Toronto tomorrow. 

CH4 Hunter

We've had a productive first two days at the Canadian Space Agency testing our methane-hunting scenario.  The rover drove to many locations on the Mars Emulation Terrain and gathered measurements from an active spectrometer, which bounces a laser off retroreflectors, to try to locate a hidden source of methane.  Tomorrow we'll switch over to a passive spectrometer scenario where we'll be using the sun as a light source (instead of bouncing a laser off a retroreflector).

Arriva!!!

We've arrived at the Canadian Space Agency for rover testing and it's looking to be a sunny week.

Touchdown!

The NASA/JPL-Caltech rover, Curiosity, has successfully landed on Mars this morning and sent back its first images!  You can read more about it on the Mars Science Laboratory webpage.  

Congratulations to everybody involved in this ground-breaking mission.  It is a truly remarkable achievement!

The Countdown Has Begun

ASRL rover being prepared
for mock Mars mission

With the NASA Mars Science Laboratory landing just three days away, we are also counting down to our next field season, coincidentally also centred around a Mars mission concept.  Our robot will be hunting for a methane source in the Canadian Space Agency's Mars Emulation Terrain.  Scientists believe methane may play an important role in understanding Mars.  We'll have two different spectrometers mounted on the robot, capable of detecting methane.  One will bounce a laser off a reflective sign, the other will point at the sun.  Both spectrometers can detect the presence of methane in the direction they're pointing, so the robot will have to wander around until it finds the source; it will be a bit like the game, Battleship.  The test will occur August 20-24, hopefully just as the Curiosity rover is taking its first steps.  Stay tuned for updates on our progress!