A novel Q-Learning Based Cartesian Model Reference Compliance Controller Implementation for a Humanoid Robotic Arm

SG Khan; G Herrmann; F.L Lewis; CR Melhuish; AG Pipe

doi:10.1109/RAMECH.2011.6070484

A novel Q-Learning Based Cartesian Model Reference Compliance Controller Implementation for a Humanoid Robotic Arm

SG Khan
, G Herrmann
, F.L Lewis
, CR Melhuish
, AG Pipe

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

6 Citations (Scopus)

Abstract

This paper presents the implementation (real time and simulation) of a model-free Q-learning based discrete model reference compliance controller for a humanoid robot arm. The Reinforcement learning (RL) scheme uses a recently developed Q-learning scheme to develop an optimal policy on-line. The RL Cartesian (x and y) tracking controller with model reference compliance was implemented using two links (shoulder flexion and elbow flexion joints) of the right arm of the humanoid Bristol-Elumotion-Robotic-Torso II (BERT II) torso.

Translated title of the contribution	A novel Q-Learning Based Cartesian Model Reference Compliance Controller Implementation for a Humanoid Robotic Arm
Original language	English
Title of host publication	5th IEEE International Conference on Robotics, Automation and Mechatronics (RAM), Qingdao
DOIs	https://doi.org/10.1109/RAMECH.2011.6070484
Publication status	Published - 2011

Bibliographical note

Conference Organiser: IEEE

Access to Document

10.1109/RAMECH.2011.6070484

http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=6070484

Handle.net

Persistent link

Cite this

@inproceedings{d437a9b19eec4fa1ae88596cdca2303e,

title = "A novel Q-Learning Based Cartesian Model Reference Compliance Controller Implementation for a Humanoid Robotic Arm",

abstract = "This paper presents the implementation (real time and simulation) of a model-free Q-learning based discrete model reference compliance controller for a humanoid robot arm. The Reinforcement learning (RL) scheme uses a recently developed Q-learning scheme to develop an optimal policy on-line. The RL Cartesian (x and y) tracking controller with model reference compliance was implemented using two links (shoulder flexion and elbow flexion joints) of the right arm of the humanoid Bristol-Elumotion-Robotic-Torso II (BERT II) torso.",

author = "SG Khan and G Herrmann and F.L Lewis and CR Melhuish and AG Pipe",

note = "Conference Organiser: IEEE",

year = "2011",

doi = "10.1109/RAMECH.2011.6070484",

language = "English",

booktitle = "5th IEEE International Conference on Robotics, Automation and Mechatronics (RAM), Qingdao",

}

TY - GEN

T1 - A novel Q-Learning Based Cartesian Model Reference Compliance Controller Implementation for a Humanoid Robotic Arm

AU - Khan, SG

AU - Herrmann, G

AU - Lewis, F.L

AU - Melhuish, CR

AU - Pipe, AG

N1 - Conference Organiser: IEEE

PY - 2011

Y1 - 2011

N2 - This paper presents the implementation (real time and simulation) of a model-free Q-learning based discrete model reference compliance controller for a humanoid robot arm. The Reinforcement learning (RL) scheme uses a recently developed Q-learning scheme to develop an optimal policy on-line. The RL Cartesian (x and y) tracking controller with model reference compliance was implemented using two links (shoulder flexion and elbow flexion joints) of the right arm of the humanoid Bristol-Elumotion-Robotic-Torso II (BERT II) torso.

AB - This paper presents the implementation (real time and simulation) of a model-free Q-learning based discrete model reference compliance controller for a humanoid robot arm. The Reinforcement learning (RL) scheme uses a recently developed Q-learning scheme to develop an optimal policy on-line. The RL Cartesian (x and y) tracking controller with model reference compliance was implemented using two links (shoulder flexion and elbow flexion joints) of the right arm of the humanoid Bristol-Elumotion-Robotic-Torso II (BERT II) torso.

U2 - 10.1109/RAMECH.2011.6070484

DO - 10.1109/RAMECH.2011.6070484

M3 - Conference Contribution (Conference Proceeding)

BT - 5th IEEE International Conference on Robotics, Automation and Mechatronics (RAM), Qingdao

ER -

A novel Q-Learning Based Cartesian Model Reference Compliance Controller Implementation for a Humanoid Robotic Arm

Abstract

Bibliographical note

Access to Document

Handle.net

Fingerprint

Cite this