Hi,
Thanks a lot for designing these environments. In your paper, the elasto-plastic environment is hardly mentioned and is not part of the benchmark at all. Is there any reason for that? Has this environment limitations that prevented you from including it in the benchmark?
I ran the demo script for the elasto-plastic environment and I found that one can easily compress the dough almost infinitely (by sequentially applying actions towards the dough), which looks very unrealistic. Is it an issue you are aware of? Do you know where this issue comes from and if there is a way to prevent this?
Thank you