Ali Ghadirzadeh, Atsuto Maki, Mårten Björkman Sept 28- Oct 2 2015. Hamburg Germany Presented by Jen-Fang Chang 1.

Slides:



Advertisements
Similar presentations
Kinematic Synthesis of Robotic Manipulators from Task Descriptions June 2003 By: Tarek Sobh, Daniel Toundykov.
Advertisements

Effective Reinforcement Learning for Mobile Robots Smart, D.L and Kaelbing, L.P.
Patch to the Future: Unsupervised Visual Prediction
SA-1 Robotic Self-Perception and Body Scheme Learning Jürgen Sturm Christian Plagemann Wolfram Burgard University of Freiburg Germany.
Real-Time Human Pose Recognition in Parts from Single Depth Images Presented by: Mohammad A. Gowayyed.
4/15/2017 Using Gaussian Process Regression for Efficient Motion Planning in Environments with Deformable Objects Barbara Frank, Cyrill Stachniss, Nichola.
Yiannis Demiris and Anthony Dearden By James Gilbert.
1Notes  Handing assignment 0 back (at the front of the room)  Read the newsgroup!  Planning to put 16mm films on the web soon (possibly tomorrow)
SA-1 Body Scheme Learning Through Self-Perception Jürgen Sturm, Christian Plagemann, Wolfram Burgard.
Quantifying Generalization from Trial-by-Trial Behavior in Reaching Movement Dan Liu Natural Computation Group Cognitive Science Department, UCSD March,
Vision-Based Motion Control of Robots
Tracking a moving object with real-time obstacle avoidance Chung-Hao Chen, Chang Cheng, David Page, Andreas Koschan and Mongi Abidi Imaging, Robotics and.
The City College of New York 1 Jizhong Xiao Department of Electrical Engineering City College of New York Manipulator Control Introduction.
MEAM 620 Project Report Nima Moshtagh.
Continuum Crowds Adrien Treuille, Siggraph 王上文.
Associative Learning in Hierarchical Self Organizing Learning Arrays Janusz A. Starzyk, Zhen Zhu, and Yue Li School of Electrical Engineering and Computer.
High Speed Obstacle Avoidance using Monocular Vision and Reinforcement Learning Jeff Michels Ashutosh Saxena Andrew Y. Ng Stanford University ICML 2005.
Pattern Recognition. Introduction. Definitions.. Recognition process. Recognition process relates input signal to the stored concepts about the object.
Reinforcement Learning Game playing: So far, we have told the agent the value of a given board position. How can agent learn which positions are important?
Computational aspects of motor control and motor learning Michael I. Jordan* Mark J. Buller (mbuller) 21 February 2007 *In H. Heuer & S. Keele, (Eds.),
Face Recognition Using Neural Networks Presented By: Hadis Mohseni Leila Taghavi Atefeh Mirsafian.
역운동학의 구현과 응용 Implementation of Inverse Kinematics and Application 서울대학교 전기공학부 휴먼애니메이션연구단 최광진
Music and Interaction Development Using the Countess Quanta Robot Brad Pitney Yin Shi Chal McCollough.
Radial Basis Function Networks
Goal Directed Design of Serial Robotic Manipulators
Function Approximation for Imitation Learning in Humanoid Robots Rajesh P. N. Rao Dept of Computer Science and Engineering University of Washington,
Definition of an Industrial Robot
Resilient Machines Through Continuous Self-Modeling Pattern Recognition Seung-Hyun Lee Soft Computing Lab. Josh Bongard,Victor Zykov, and Hod.
IMPLEMENTATION ISSUES REGARDING A 3D ROBOT – BASED LASER SCANNING SYSTEM Theodor Borangiu, Anamaria Dogar, Alexandru Dumitrache University Politehnica.
Construction and Motion control of a mobile robot using Visual Roadmap
Constraints-based Motion Planning for an Automatic, Flexible Laser Scanning Robotized Platform Th. Borangiu, A. Dogar, A. Dumitrache University Politehnica.
Optimization-Based Full Body Control for the DARPA Robotics Challenge Siyuan Feng Mar
Lecture 2: Introduction to Concepts in Robotics
INVERSE KINEMATICS ANALYSIS TRAJECTORY PLANNING FOR A ROBOT ARM Proceedings of th Asian Control Conference Kaohsiung, Taiwan, May 15-18, 2011 Guo-Shing.
COMMON EVALUATION FINAL PROJECT Vira Oleksyuk ECE 8110: Introduction to machine Learning and Pattern Recognition.
Towards Cognitive Robotics Biointelligence Laboratory School of Computer Science and Engineering Seoul National University Christian.
Vision-Based Reach-To-Grasp Movements From the Human Example to an Autonomous Robotic System Alexa Hauck.
Robotica Lecture 3. 2 Robot Control Robot control is the mean by which the sensing and action of a robot are coordinated The infinitely many possible.
Vrobotics I. DeSouza, I. Jookhun, R. Mete, J. Timbreza, Z. Hossain Group 3 “Helping people reach further”
HARDWARE INTERFACE FOR A 3-DOF SURGICAL ROBOT ARM Ahmet Atasoy 1, Mehmed Ozkan 2, Duygun Erol Barkana 3 1 Institute of Biomedical Engineering, Bogazici.
M.S. Thesis Defense Jason Anderson Electrical and Computer Engineering Dept. Clemson University.
Reinforcement Learning 主講人:虞台文 Content Introduction Main Elements Markov Decision Process (MDP) Value Functions.
DARPA ITO/MARS Project Update Vanderbilt University A Software Architecture and Tools for Autonomous Robots that Learn on Mission K. Kawamura, M. Wilkes,
A Passive Approach to Sensor Network Localization Rahul Biswas and Sebastian Thrun International Conference on Intelligent Robots and Systems 2004 Presented.
Abstract This presentation questions the need for reinforcement learning and related paradigms from machine-learning, when trying to optimise the behavior.
Reinforcement Learning 主講人:虞台文 大同大學資工所 智慧型多媒體研究室.
Image Tracing Laser System Jason Duarte Azmat Latif Stephen Sundell Tim Weidner.
Inverse Kinematics for Robotics using Neural Networks. Authors: Sreenivas Tejomurtula., Subhash Kak
CSE 185 Introduction to Computer Vision Feature Matching.
Chapter 8. Learning of Gestures by Imitation in a Humanoid Robot in Imitation and Social Learning in Robots, Calinon and Billard. Course: Robots Learning.
Vision-based SLAM Enhanced by Particle Swarm Optimization on the Euclidean Group Vision seminar : Dec Young Ki BAIK Computer Vision Lab.
Randomized Kinodynamics Planning Steven M. LaVelle and James J
Learning Photographic Global Tonal Adjustment with a Database of Input / Output Image Pairs.
Semantic Alignment Spring 2009 Ben-Gurion University of the Negev.
Robot Intelligence Technology Lab. 10. Complex Hardware Morphologies: Walking Machines Presented by In-Won Park
Shape2Pose: Human Centric Shape Analysis CMPT888 Vladimir G. Kim Siddhartha Chaudhuri Leonidas Guibas Thomas Funkhouser Stanford University Princeton University.
1 Algoritmos Genéticos aplicados em Machine Learning Controle de um Robo (em inglês)
Introduction to Machine Learning, its potential usage in network area,
Parametric calibration of speed–density relationships in mesoscopic traffic simulator with data mining Adviser: Yu-Chiang Li Speaker: Gung-Shian Lin Date:2009/10/20.
Minor Project - Human Interaction Robot Arm
Presented by Aayush Mudgal
Date of download: 1/2/2018 Copyright © ASME. All rights reserved.
Dynamical Statistical Shape Priors for Level Set Based Tracking
Parametric calibration of speed–density relationships in mesoscopic traffic simulator with data mining Adviser: Yu-Chiang Li Speaker: Gung-Shian Lin Date:2009/10/20.
Presentation by Ryan Brand
Ch 14. Active Vision for Goal-Oriented Humanoid Robot Walking (1/2) Creating Brain-Like Intelligence, Sendhoff et al. (eds), Robots Learning from.
2-DOF Manipulator Now, given the joint angles Ө1, Ө2 we can determine the end effecter coordinates x and y.
Emir Zeylan Stylianos Filippou
Humanoid Motion Planning for Dual-Arm Manipulation and Re-Grasping Tasks Nikolaus Vahrenkamp, Dmitry Berenson, Tamim Asfour, James Kuffner, Rudiger Dillmann.
Presentation transcript:

Ali Ghadirzadeh, Atsuto Maki, Mårten Björkman Sept 28- Oct Hamburg Germany Presented by Jen-Fang Chang 1

Outline Introduction Proposed method Experiment results Conclusion and Future work 2

Introduction Sensorimotor contingencies(SMC) based method Sensory awareness are the result of integrating sensorimotor coupling into the planning system. Grounded directly to the environment, give the flexibility to designs systems with self-learning. 3

Visual servoing Visual servoing is an approach to control a robot’s motion using feedback. Unlike tradition methods to design system, there is no need to calibrate the robot by SMC based approach. The task is grounded to the environment, enabling self- learning without intervention. 4

Main contribution Eliminate the need for prior knowledge of kinematic or dynamic models of the robot Use forward model to search for proper actions to solve tasks by minimizing a cost function, instead of training a separate inverse model, to speed up training Encode 3D spatial positions of a target project to avoid calibration with external coordinate system 5

Related Works Forward model with Distal Supervised learning (DSL) Combined forward-inverse model learning Affordance theory learning Reinforcement learning PILCO 6

Proposed method Training a forward model Finding inverse output Gaussian Process regression Kinematic system Visuomotor tasks 7

Training a forward model START Target S* cost function initial train forward model with randomly generated action-observation pairs Target state reached? No Yes Gaussian Process regression Find inverse output General form forward model: 8

Inverse output The cost function defined as (1) To find the optimum motor command that minimize the cost function, we use gradient based method: (2) Equation (2) can be written as: (3) 9

Apply gradient descent repeatedly gets an optimum unbounded value: (4) Insert the optimum value to (5) We can get the motor commands that optimize the cost function 10

Gaussian Process regression A non-parametric model where the representation is given by the training samples. The key factor to successfully learn the forward models in real-time. 11

Kinematic system The camera joints consist Jnp (neck pan), Jnt (neck tilt) and Jv (Vergence angle) The robot arms joints consist Jsp(shoulder pan), Jsl(shoulder lift), Jar (arm roll) and Je(elbow) The control architect is hierarchical and can be regarded as a look and move architecture A position is defined by camera joints that would be required for the robot to fixate onto the object 12

Visuomotor tasks Fixation task Fixate onto an object which can be observed with the maximum overlap between the two camera view The process serves as a mean to probe the 3D position Reaching task Use the information of 3D position and try to reduce the distance between the end-effector and target 13

14 Fixation forward model structure Reaching forward model structure Jv missing

Experiment results Learning performance for the training and test phases of the fixation task, Implemented on the PR2 robot. 15 Model training

Learning performance for the reaching task, (a) with the same target used for training and test, and (b) for two new target. 16

Cost functions evaluation 17

3D position encoding 18 The Euclidean distance between two points in the joint space against the 3D distance in Metric space

Tolerance against image distortion The cost functions are given as the average of 10 different trials, performed in the simulation environment. Image with no distortion, moderate(λ= 0.9) and considerable(λ= 0.7) 19

Conclusion The most important feature of the proposed method are the real-time learning and the fact that it requires the robot to only have a few interaction with the environment. One key factor to speed up the training is that we applied the forward model to search for the motor commands by minimizing a given cost, instead of training a separate inverse model. 20

Future work Control a robot arm while avoiding obstacles Utilize other regression model for different event 21

22