Cyclist Detection from Volvo

[mtbxplorer]

http://www.google.com/url?sa=t&rct=j...43287494,d.dmQ

[hunter006]

http://www.google.com/url?sa=t&rct=j...43287494,d.dmQ

Cascade Bikes posted this as well, and I had a quick chat with them. Given how difficult it was for the DARPA project to kick off autonomous driving without colliding in to stuff, I think this will work well 90% of the time and cause problems 10%. It's a sliding scale, but I doubt it's foolproof.

Cases that I'd like to see tested:
1) Driving over loose surface.
2) Driving over split surface (assuming ABS will kick in, but sometimes these things don't play nicely together)
3) When approaching a curve at high speed and a cyclist is ahead (i.e. cyclist has just entered the curb, car is approaching from the rear).
4) Whether it has problems handling the relatively close proximity of cyclists in bike lanes. I'm assuming this is a, "no" but stranger things have happened.

Basically, I think it will work pretty well, but not be a substitute for old fashioned paying attention. I think it will fail pretty well under #3, which would make driving US101 through Oregon during the summer, when there's lots of cyclists out there, pretty nightmarish.