search

bradyz / 2020_CARLA_challenge

"Learning by Cheating" (CoRL 2019) submission for the 2020 CARLA Challenge
181 stars 49 forks source link

Velocity, Target As Input #29

Closed omerXfaruq closed 3 years ago

omerXfaruq commented 3 years ago

Hello;

I am confused about target entering to neural net as input. At first I thought it as velocity however after searching the issues I found that it is a 50 meter away future target.

-Don't you use velocity as input like in LBC original repo? -How do you calculate position of the target? And what are the modifications happening to target, do you project the target position onto the camera and heatmap?

bradyz commented 3 years ago

here we don't use velocity as input

the position of the far away target is given during test time - this is essentially the route the ego vehicle is supposed to go, but there's lots of noise in this due to GPS, and it is heavily downsampled (only given every 50m or so)