Emergence of Human-comparable Balancing Behaviors by Deep Reinforcement Learning

by   Chuanyu Yang, et al.

This paper presents a hierarchical framework based on deep reinforcement learning that learns a diversity of policies for humanoid balance control. Conventional zero moment point based controllers perform limited actions during under-actuation, whereas the proposed framework can perform human-like balancing behaviors such as active push-off of ankles. The learning is done through the design of an explainable reward based on physical constraints. The simulated results are presented and analyzed. The successful emergence of human-like behaviors through deep reinforcement learning proves the feasibility of using an AI-based approach for learning humanoid balancing control in a unified framework.


A Multi-Objective Deep Reinforcement Learning Framework

This paper presents a new multi-objective deep reinforcement learning (M...

Emergence of Different Modes of Tool Use in a Reaching and Dragging Task

Tool use is an important milestone in the evolution of intelligence. In ...

Modeling Human Cognition with a Hybrid Deep Reinforcement Learning Agent

Human cognition model could help us gain insights in how human cognition...

Learning Whole-body Motor Skills for Humanoids

This paper presents a hierarchical framework for Deep Reinforcement Lear...

On the Emergence of Whole-body Strategies from Humanoid Robot Push-recovery Learning

Balancing and push-recovery are essential capabilities enabling humanoid...

Diversity-based Deep Reinforcement Learning Towards Multidimensional Difficulty for Fighting Game AI

In fighting games, individual players of the same skill level often exhi...

Learning natural locomotion behaviors for humanoid robots using human knowledge

This paper presents a new learning framework that leverages the knowledg...

Please sign up or login with your details

Forgot password? Click here to reset