Problem with actuator control and step change constraint #2451
Replies: 1 comment 2 replies
-
Did you try a position actuator? |
Beta Was this translation helpful? Give feedback.
2 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Intro
Hi!
I am a applied computer science student and I'm new to MuJoCo. I use MuJoCo for a project on reinforcement learning.
My setup
MuJoCo version 3.2.7, Python 3.13, arm64, macOS
My question
I'm currently working on a maze designed to be solved using reinforcement learning (RL). However, I'm facing an issue with actuator control: the board's motors are set up to apply torque, but the applied force isn't proportional to the input. Specifically, only about 0.15 of the control range (which is between -1 and 1) actually produces any noticeable tilt change, small inputs result in disproportionately large changes, while larger inputs hardly have any effect.
I'm looking for a solution that makes the board's response more predictable and consistent, with a uniform mapping from the control input to tilt angle.
Additionally, I was wondering if it is possible to impose a constraint so that the change in tilt from one step to the next is limited to a maximum value.
Minimal model and/or code that explain my question
Confirmations
Beta Was this translation helpful? Give feedback.
All reactions