AIJan 2, 2019

Learning Humanoid Robot Motions Through Deep Neural Networks

arXiv:1901.00270v19 citations
Originality Synthesis-oriented
AI Analysis

This addresses the challenge of robot motion control for robotics researchers, but appears incremental as it builds on existing learning approaches without major breakthroughs.

The paper tackles the problem of controlling high degrees of freedom humanoid robots by proposing a neural network-based learning framework to mimic movements, applied in the RoboCup 3D Soccer Simulation domain with promising results.

Controlling a high degrees of freedom humanoid robot is acknowledged as one of the hardest problems in Robotics. Due to the lack of mathematical models, an approach frequently employed is to rely on human intuition to design keyframe movements by hand, usually aided by graphical tools. In this paper, we propose a learning framework based on neural networks in order to mimic humanoid robot movements. The developed technique does not make any assumption about the underlying implementation of the movement, therefore both keyframe and model-based motions may be learned. The framework was applied in the RoboCup 3D Soccer Simulation domain and promising results were obtained using the same network architecture for several motions, even when copying motions from another teams.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes