[Proposal] Request for Implementation of a More Practical Terrain Level Curriculum Function

[H-Hisamichi]

Proposal

Hello everyone,

I would like to propose the implementation of a terrain curriculum function that supports more accurate movement distance estimation and command resampling during episodes.

Motivation

I am working on addressing the issue of insufficient terrain level progression by implementing a more accurate method for measuring actor movement distance. (There have also been excellent suggestions in the discussion here)
However, implementing a function that requires managing time-series data is a challenging task for someone new to IsaacLab. Additionally, I believe the existing terrain_levels_vel does not account for command updates during an episode.
Since proper terrain level updates significantly impact policy performance, I kindly request consideration for implementing this new terrain curriculum function!

Checklist

• I have checked that there is no similar issue in the repo (required)

[RandomOakForest]

Thank you for posting this proposal. The team will review it.