google-deepmind / google-deepmind/dm_control

Getting started with interactive viewer

未关闭
#353 1 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看
enhancement
主要语言
Python
星标
4.7k
派生
764
PR 合并指标
30 天内没有已合并 PR

描述

I'm new to mujoco and I wish to run my robot model in an interactive viewer.
In various posts I encountered, it is recommended to use mujoco's `simulate` utility, however, to my understanding, `simulate` will not allow me full programmable control of my robots (e.g., applying external forces).

I guess I should use `dm_control.viewer.launch(env)`, but this requires building a `composer.Environment` object as well as defining a `Task`, which seem pretty complex.
Is there sample code showing how to generate a simple `Environment` object? Alternatively, is there an easier way to get started with interactive viewer?

贡献指南

打开贡献指南

评估

这个 Issue 还没有评估数据。

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。