Humanoid Control

8 papers with code • 0 benchmarks • 0 datasets

Control of a high-dimensional humanoid. This can include skill learning by tracking motion capture clips, learning goal-directed tasks like going towards a moving target, and generating motion within a physics simulator.

Most implemented papers

Optical Non-Line-of-Sight Physics-based 3D Human Pose Estimation

marikoisogawa/OpticalNLOSPose CVPR 2020

We bring together a diverse set of technologies from NLOS imaging, human pose estimation and deep reinforcement learning to construct an end-to-end data processing pipeline that converts a raw stream of photon measurements into a full 3D human pose sequence estimate.

Residual Force Control for Agile Human Behavior Imitation and Extended Motion Synthesis

Khrylx/RFC NeurIPS 2020

Our approach is the first humanoid control method that successfully learns from a large-scale human motion dataset (Human3. 6M) and generates diverse long-term motions.

ClipUp: A Simple and Powerful Optimizer for Distribution-based Policy Evolution

nnaisense/pgpelib 5 Aug 2020

In these algorithms, gradients of the total reward with respect to the policy parameters are estimated using a population of solutions drawn from a search distribution, and then used for policy optimization with stochastic gradient ascent.

On the model-based stochastic value gradient for continuous reinforcement learning

facebookresearch/svg 28 Aug 2020

For over a decade, model-based reinforcement learning has been seen as a way to leverage control-based domain knowledge to improve the sample-efficiency of reinforcement learning agents.

Learning to Brachiate via Simplified Model Imitation

brachiation-rl/brachiation 8 May 2022

Key to our method is the use of a simplified model, a point mass with a virtual arm, for which we first learn a policy that can brachiate across handhold sequences with a prescribed order.

Reincarnating Reinforcement Learning: Reusing Prior Computation to Accelerate Progress

google-research/reincarnating_rl 3 Jun 2022

To address these issues, we present reincarnating RL as an alternative workflow or class of problem settings, where prior computational work (e. g., learned policies) is reused or transferred between design iterations of an RL agent, or from one RL agent to another.

Learning Soccer Juggling Skills with Layer-wise Mixture-of-Experts

ZhaomingXie/soccer_juggle_release SIGGRAPH 2022

Learning physics-based character controllers that can successfully integrate diverse motor skills using a single policy remains a challenging problem.

MoCapAct: A Multi-Task Dataset for Simulated Humanoid Control

microsoft/MoCapAct 15 Aug 2022

We demonstrate the utility of MoCapAct by using it to train a single hierarchical policy capable of tracking the entire MoCap dataset within dm_control and show the learned low-level component can be re-used to efficiently learn downstream high-level tasks.