This repository was archived by the owner on Aug 17, 2025. It is now read-only.

Description
Hi!
Appreciate for the great work!
I noticed that in your RULE-BASED TRAJECTORY SCORING STRATEGY, you need to use target positions, target heading, and target speed. Is this 'target' the ground truth of the target trajectory? However, in actual online trajectory prediction or test set inference, this ground truth might not be available, so how are the target values obtained?