Closed lambdavi closed 6 months ago
Hi @lambdavi, there is a known issue with the Unitree A1 environment not allowing custom reward functions yet. I will add this feature today, and push it to the master branch. I would ask you to switch to an editable installation until the next release, which is going to happen soon! Sorry for the inconvenience, and thanks for reporting this! I will notify you once added.
You probably know it already, but that same behaviour is extended to other envs, such as Unitree H1. I will switch to the editable version. I am going to check out better the code for the simulation, if you need help in developing some not critical feature, just let me know.
Thanks for the prompt reply and help!
Yeah, with the latest release, there was a problem with the custom rewards for other robots as well. But that was fixed already, it is just not released yet. The only one missing is the Unitree A1, which requires a few more changes. Thanks for being willing to help out!
Sorry for the delay, this has been fixed now!
Hello. Great work done with the simulator. However, I am encountering a problem in using the environment with custom reward function. I installed loco_mujoco through pip.
Following your documentation I encounter this error:
It looks like I can't override the reward function. I don't know if you have any suggestions in this sense or if you encountered this problem as well and you know an easy fix.