[nosequel]

This is really cool. I was on a team for the first and second DARPA Grand Challenge. The teams (like ours) who had to use actuators, pistons, gears oh-my for steering, acceleration, braking were at a huge disadvantage compared to the teams who were given the drive-by-wire access (CMU/Stanford & VW Touareg partnership(s)). It would've been really nice at the time to get the mechanical aspects all worked out and have a nice autonomy layer to use that was ready and tested. I applaud the work you are doing and know it is NOT easy. I also happen to own some construction equipment and can't imagine how hard it would be to interface with a dual pilot joystick configuration (think skids & dozers). It is a completely different way to think about travel, turning, bucket work, tipping weight. Sounds like you all have some really interesting tough problems to work on. Bravo.

[stefan8r]

Thanks for the kind words!

Ya - I can't imagine how hectic it was for you guys. I remember hearing Josh Switkes talk about hand building radars for pre-2010 autonomy.

Re:booms/buckets/etc: right now we're just focusing on mobility (making these systems drive from A to B), specifically because those systems are deceptively hard. FWIW I think the guys at Built Robotics have done some cool stuff on it, as have groups like Moog and FP Innovations.

[iliabara]

Thanks for the kind words, and awesome that you were on a DGC team!

Absolutely, there is a ton of really interesting controls, ML and general software quality work for us to do.