Lecture 13: Dynamical Systems based Representation. Contents:
|
|
- Samuel Harris
- 5 years ago
- Views:
Transcription
1 Lecture 13: Dynamical Systems based Representation Contents: Differential Equation Force Fields, Velocity Fields Dynamical systems for Trajectory Plans Generating plans dynamically Fitting (or modifying) plans Imitation based learning Thanks to my collaborator Auke Ijspeert (EPFL) for many of the contents on the slides for this lecture. Lecture 13: RLSC - Prof. Sethu Vijayakumar 1
2 Movement policies as Dynamical Systems des f (,, goal) des Discreet & Rhythmic Movement Primitives Represent complex movements in globally stable attractor landscapes of nonlinear autonomous differential equations Choose kinematic representation for easy reuse in different workspace location Ensure easy temporal and spatial scaling (topological equivalence) Use local learning to modify the attractors according to demonstration of teacher and self-learning Lecture 13: RLSC - Prof. Sethu Vijayakumar 2
3 Discreet & Rhythmic Movement superposition Open loop with oscillators Closed loop control in horizontal plane Open (vertical) + Closed (horizontal) loop control Lecture 13: RLSC - Prof. Sethu Vijayakumar 3
4 What is a Differential Equation? Differential equation: an equation that describes how state variables evolve over time, for instance: y ( c y) Lecture 13: RLSC - Prof. Sethu Vijayakumar 4
5 Some Definitions Ordinary differential equation: differential equation that involves only ordinary derivatives (as opposed to partial derivatives) Autonomous equation: differential equation that does not (explicitly) depend on time Linear differential equation: differential equation in which the state variables only appear in linear combinations Nonlinear differential equation: differential equation in which some state variables appear in nonlinear combinations (e.g. products, cosine, ) Fixed point: point at which all derivatives are zero (can be an attractor, a repeller, or a saddle point, cf later) Limit cycle: periodic isolated closed trajectory (can only occur in nonlinear systems) Lecture 13: RLSC - Prof. Sethu Vijayakumar 5
6 Interesting Regimes of Differential Eqs. From Strogatz 1994 Unstable node saddles Attractor Attractors Limit cycles Chaos Lecture 13: RLSC - Prof. Sethu Vijayakumar 6
7 First Order Linear Systems First order linear system: y ( c y) How to solve this equation, for a given y(t=0), c, and? Two methods: analytical solution or numerical integration Analytical solution: y( t) ( y c)exp( 0 t) Numerical integration: Euler method, Runge-Kutta, c Lecture 13: RLSC - Prof. Sethu Vijayakumar 7
8 First Order Linear Systems y ( c y) y( t) ( y c)exp( 0 t) c Lecture 13: RLSC - Prof. Sethu Vijayakumar 8
9 Second Order Linear Systems y ( ( c x) y) x y Lecture 13: RLSC - Prof. Sethu Vijayakumar 9
10 Second Order Linear Systems y ( ( c x) y) x y Lecture 13: RLSC - Prof. Sethu Vijayakumar 10
11 Second Order Non-linear System y x 2cos 2cos y cos x x cos y Repeller saddles Attractor Lecture 13: RLSC - Prof. Sethu Vijayakumar 11
12 Learning a movement by demonstration Movement recognition Joint angles Trajectory formation system Movement execution (Inv. Dyn.) Task of the trajectory formation system: To encode demonstrated trajectories with high accuracy, To be able to modulate the learned trajectory when: Perceptual variables are varied (e.g. timing, amplitude) Perturbations occur Lecture 13: RLSC - Prof. Sethu Vijayakumar 12
13 Encoding a trajectory Traditionally, the problem of replaying a trajectory has been decomposed into two different issues: One of encoding the trajectory, and One of modifying the trajectory, for instance, in case the movement is perturbed, or when it requires to be modulated. Our approach: combine both abilities in a nonlinear dynamical system Aim: to encode the trajectory in a nonlinear dynamical system with well defined attractor landscape Lecture 13: RLSC - Prof. Sethu Vijayakumar 13
14 Nonlinear Dynamical Systems Approach Discrete movements dy/dt Rhythmic movements dy/dt y y Single point attractor Limit cycle attractor Lecture 13: RLSC - Prof. Sethu Vijayakumar 14
15 Two types of movement recordings Sensuit «Kinesthetic» demonstration Lecture 13: RLSC - Prof. Sethu Vijayakumar 15
16 Pointing Demonstration position velocity Demo for One DOF Lecture 13: RLSC - Prof. Sethu Vijayakumar 16
17 Shaping Attractor Landscapes Goal: g z g y z y z z z Can one create more complex dynamics by nonlinearly modifying the dynamic system: z z? z g y z y f z Lecture 13: RLSC - Prof. Sethu Vijayakumar 17
18 Discreet Control Policy Output System Goal: g Amplitude and phase system Lecture 13: RLSC - Prof. Sethu Vijayakumar 18
19 Shaping Attractor Landscapes A globally stable learnable nonlinear point attractor: Trajectory Plan Dynamics Canonical Dynamics Local Linear Model Approx. z z g y z y f x, v z y where v g x v x f v v x x, v z v k i1 k i1 i i w b v w i 1 wi exp di x ci and x 2 x x g x Lecture 13: RLSC - Prof. Sethu Vijayakumar 19
20 Learning the Attractor Given a demonstrated trajectory y(t) demo and a goal g Extract movement duration Adjust time constants of canonical dynamics to movement duration Use LWL to learn supervised problem ydemo ytarget z f x, v y Also extended to rhythmic primitives : [ Stefan Schaal, Sethu Vijayakumar et al, Proc. of Intl. Symp. Rob. Res.(ISRR) (2001) ] Lecture 13: RLSC - Prof. Sethu Vijayakumar 20
21 Trajectory following & Generalization Backhand Demonstration Backhand Reproduction Lecture 13: RLSC - Prof. Sethu Vijayakumar 21
22 Lecture 13: RLSC - Prof. Sethu Vijayakumar 22 Basic oscillator: Output signal: 1 ( 0) r r r N i i N i T i i m I w z y z y y z 1 1 ) ) ( ( ],, [ v x T v x I ], [ Rhythmic Control Policies
23 Modulation of Goal: Anchor Lecture 13: RLSC - Prof. Sethu Vijayakumar 23
24 Drumming: Modulating Frequency Drumming: Kinesthetic Demo Imitation and Modulation Lecture 13: RLSC - Prof. Sethu Vijayakumar 24
25 Biped Locomotion using DMPs 5-link biped robot [Nakanishi, Morimoto, Endo, Cheng, Schaal, and Kawato, 2004] Height 40cm Mass hip 2.0kg knee 0.64kg foot 0.06kg Approach Movement primitives as a central pattern generator (CPG) Synchronization between CPG and robot - phase resetting and frequency adaptation - phase regulation between coupled oscillators Lecture 13: RLSC - Prof. Sethu Vijayakumar 25
26 Control Architecture Rhythmic Movement primitives periodic pattern generation learning from demonstrated trajectories Phase resetting [Nakanishi, Morimoto, Endo, Cheng, Schaal, and Kawato, 2004] CPG reset the phase of the oscillator at heel contact oscillator output with local models Dynamical primitives (index: i=1~# of oscillators) Frequency adaptation Phase regulation among oscillators coupling among oscillators Robot and Environment phase resetting frequency update Lecture 13: RLSC - Prof. Sethu Vijayakumar 26
27 Experiments: Bipedal Locomotion Lecture 13: RLSC - Prof. Sethu Vijayakumar 27
Learning Attractor Landscapes for Learning Motor Primitives
Learning Attractor Landscapes for Learning Motor Primitives Auke Jan Ijspeert,, Jun Nakanishi, and Stefan Schaal, University of Southern California, Los Angeles, CA 989-5, USA ATR Human Information Science
More informationLearning Manipulation Patterns
Introduction Dynamic Movement Primitives Reinforcement Learning Neuroinformatics Group February 29, 2012 Introduction Dynamic Movement Primitives Reinforcement Learning Motivation shift to reactive motion
More informationCS545 Contents XIII. Trajectory Planning. Reading Assignment for Next Class
CS545 Contents XIII Trajectory Planning Control Policies Desired Trajectories Optimization Methods Dynamical Systems Reading Assignment for Next Class See http://www-clmc.usc.edu/~cs545 Learning Policies
More informationLearning Movement Primitives
S. Schaal, J. Peters, J. Nakanishi, and A. Ijspeert, "Learning Movement Primitives," in International Symposium on Robotics Research (ISRR2003), Springer Tracts in Advanced Robotics. Ciena, Italy: Springer,
More informationOnline Learning in High Dimensions. LWPR and it s application
Lecture 9 LWPR Online Learning in High Dimensions Contents: LWPR and it s application Sethu Vijayakumar, Aaron D'Souza and Stefan Schaal, Incremental Online Learning in High Dimensions, Neural Computation,
More informationEncoding of Periodic and their Transient Motions by a Single Dynamic Movement Primitive
Encoding of Periodic and their Transient Motions by a Single Dynamic Movement Primitive Johannes Ernesti, Ludovic Righetti, Martin Do, Tamim Asfour, Stefan Schaal Institute for Anthropomatics, Karlsruhe
More informationCITEC SummerSchool 2013 Learning From Physics to Knowledge Selected Learning Methods
CITEC SummerSchool 23 Learning From Physics to Knowledge Selected Learning Methods Robert Haschke Neuroinformatics Group CITEC, Bielefeld University September, th 23 Robert Haschke (CITEC) Learning From
More informationDynamical Movement Primitives: Learning Attractor Models for Motor Behaviors
LETTER Communicated by Hirokazu Tanaka Dynamical Movement Primitives: Learning Attractor Models for Motor Behaviors Auke Jan Ijspeert auke.ijspeert@epfl.ch Ecole Polytechnique Fédérale de Lausanne, Lausanne
More informationTrajectory encoding methods in robotics
JOŽEF STEFAN INTERNATIONAL POSTGRADUATE SCHOOL Humanoid and Service Robotics Trajectory encoding methods in robotics Andrej Gams Jožef Stefan Institute Ljubljana, Slovenia andrej.gams@ijs.si 12. 04. 2018
More informationRobotics: Science & Systems [Topic 6: Control] Prof. Sethu Vijayakumar Course webpage:
Robotics: Science & Systems [Topic 6: Control] Prof. Sethu Vijayakumar Course webpage: http://wcms.inf.ed.ac.uk/ipab/rss Control Theory Concerns controlled systems of the form: and a controller of the
More informationCoordinate Change Dynamic Movement Primitives A Leader-Follower Approach
Coordinate Change Dynamic Movement Primitives A Leader-Follower Approach You Zhou, Martin Do, and Tamim Asfour Abstract Dynamic movement primitives prove to be a useful and effective way to represent a
More informationLearning Dynamical System Modulation for Constrained Reaching Tasks
In Proceedings of the IEEE-RAS International Conference on Humanoid Robots (HUMANOIDS'6) Learning Dynamical System Modulation for Constrained Reaching Tasks Micha Hersch, Florent Guenter, Sylvain Calinon
More informationDynamic Movement Primitives for Cooperative Manipulation and Synchronized Motions
Dynamic Movement Primitives for Cooperative Manipulation and Synchronized Motions Jonas Umlauft, Dominik Sieber and Sandra Hirche Abstract Cooperative manipulation, where several robots jointly manipulate
More informationStiffness and Temporal Optimization in Periodic Movements: An Optimal Control Approach
Stiffness and Temporal Optimization in Periodic Movements: An Optimal Control Approach Jun Nakanishi, Konrad Rawlik and Sethu Vijayakumar Abstract We present a novel framework for stiffness and temporal
More informationA Probabilistic Representation for Dynamic Movement Primitives
A Probabilistic Representation for Dynamic Movement Primitives Franziska Meier,2 and Stefan Schaal,2 CLMC Lab, University of Southern California, Los Angeles, USA 2 Autonomous Motion Department, MPI for
More informationDiscovery Channel. Biped
Discovery Channel Biped Schaal S, Sternad D, Osu R, Kawato M: Rhythmic arm movement is not discrete Nature Neuroscience, 7, 11371144 (004) Nakanishi J, Morimoto J, Endo G, Cheng G, Schaal S, Kawato M:
More informationLearning Motor Skills from Partially Observed Movements Executed at Different Speeds
Learning Motor Skills from Partially Observed Movements Executed at Different Speeds Marco Ewerton 1, Guilherme Maeda 1, Jan Peters 1,2 and Gerhard Neumann 3 Abstract Learning motor skills from multiple
More informationCo-evolution of Morphology and Control for Roombots
Co-evolution of Morphology and Control for Roombots Master Thesis Presentation Ebru Aydın Advisors: Prof. Auke Jan Ijspeert Rico Möckel Jesse van den Kieboom Soha Pouya Alexander Spröwitz Co-evolution
More informationLecture XI Learning with Distal Teachers (Forward and Inverse Models)
Lecture XI Learning with Distal eachers (Forward and Inverse Models) Contents: he Distal eacher Problem Distal Supervised Learning Direct Inverse Modeling Forward Models Jacobian Combined Inverse & Forward
More informationLearning robot control
Learning robot control Max Mindt Autonomous Learning System Seminar Department of Computer Science TU Darmstadt Darmstadt, Germany maxmindt@msn.com Abstract Learning is essential to expand the capabilities
More informationBiped Locomotion and Stability
Biped Locomotion and Stability A Practical Approach Jesse van den Kieboom s141474 March 29 Master Thesis Artificial Intelligence Dept of Artificial Intelligence University of Groningen, The Netherlands
More informationLearning Rhythmic Movements by Demonstration using Nonlinear Oscillators
Learning Rhthmic Movements b Demonstration using Nonlinear Oscillators Auke Jan Ijspeert, Jun Nakanishi, and Stefan Schaal, Dept. of Computer Science, Universit of Southern California, Los Angeles, USA
More informationImitation Learning of Globally Stable Non-Linear Point-to-Point Robot Motions using Nonlinear Programming
Imitation Learning of Globally Stable Non-Linear Point-to-Point Robot Motions using Nonlinear Programming S. Mohammad Khansari-Zadeh and Aude Billard Abstract This paper presents a methodology for learning
More informationStability of Dynamical systems
Stability of Dynamical systems Stability Isolated equilibria Classification of Isolated Equilibria Attractor and Repeller Almost linear systems Jacobian Matrix Stability Consider an autonomous system u
More informationPath Integral Stochastic Optimal Control for Reinforcement Learning
Preprint August 3, 204 The st Multidisciplinary Conference on Reinforcement Learning and Decision Making RLDM203 Path Integral Stochastic Optimal Control for Reinforcement Learning Farbod Farshidian Institute
More informationA Simple Reinforcement Learning Algorithm For Biped Walking
A Simple Reinforcement Learning Algorithm For Biped Walking Jun Morimoto, Gordon Cheng, Department of Humanoid Robotics and Computational Neuroscience ATR Computational Neuroscience Labs xmorimo@atr.co.jp,
More informationPoincaré-Map-Based Reinforcement Learning For Biped Walking
Poincaré-Map-Based Reinforcement Learning For Biped Walking Jun Morimoto,, Jun Nakanishi,, Gen Endo,3, and Gordon Cheng, Computational Brain Project, ICORP, JST ATR Computational Neuroscience Laboratories
More informationProbabilistic Movement Primitives under Unknown System Dynamics
To appear in Advanced Robotics Vol., No., Month XX, 8 FULL PAPER Probabilistic Movement Primitives under Unknown System Dynamics Alexandros Paraschos a, Elmar Rueckert a, Jan Peters a,c, and Gerhard Neumann
More informationLearning Control Lyapunov Function to Ensure Stability of Dynamical System-based Robot Reaching Motions
Learning Control Lyapunov Function to Ensure Stability of Dynamical System-based Robot Reaching Motions S. Mohammad Khansari-Zadeh, Aude Billard 1 EPFL-STI-IMT-LASA, CH-1, Lausanne, Switzerland Abstract
More informationPhase Locking. 1 of of 10. The PRC's amplitude determines which frequencies a neuron locks to. The PRC's slope determines if locking is stable
Printed from the Mathematica Help Browser 1 1 of 10 Phase Locking A neuron phase-locks to a periodic input it spikes at a fixed delay [Izhikevich07]. The PRC's amplitude determines which frequencies a
More informationLecture 5: Linear Genetic Programming
Lecture 5: Linear Genetic Programming CIU036 Artificial Intelligence 2, 2010 Krister Wolff, Ph.D. Department of Applied Mechanics Chalmers University of Technology 41296 Göteborg, Sweden krister.wolff@chalmers.se
More informationLong-Short Term Memory and Other Gated RNNs
Long-Short Term Memory and Other Gated RNNs Sargur Srihari srihari@buffalo.edu This is part of lecture slides on Deep Learning: http://www.cedar.buffalo.edu/~srihari/cse676 1 Topics in Sequence Modeling
More informationProbabilistic Prioritization of Movement Primitives
Probabilistic Prioritization of Movement Primitives Alexandros Paraschos 1,, Rudolf Lioutikov 1, Jan Peters 1,3 and Gerhard Neumann Abstract Movement prioritization is a common approach to combine controllers
More informationACQUISITION OF A BIPED WALKING PATTERN USING A POINCARE MAP
ACQUISITION OF A BIPED WALKING PATTERN USING A POINCARE MAP Jun Morimoto, Jun Nakanishi, Gen Endo, Gordon Cheng ICORP Computational Brain Project, JST ATR Computational Neuroscience Labs 2-2-2 Hikaridai
More informationFull Paper Proceeding ECBA-2017, Vol Issue. 37, 1-11 ECBA-17. Optimization of Central Patterns Generators. 1, 2,3 Atılım University, Turkey
Full Paper Proceeding ECBA-017, Vol. 00 - Issue. 37, 1-11 ECBA-17 Optimization of Central Patterns Generators FULL PAPER PROCEEDING Multidisciplinary Studies Available online at www.academicfora.com Abdalftah
More informationLesson 4: Non-fading Memory Nonlinearities
Lesson 4: Non-fading Memory Nonlinearities Nonlinear Signal Processing SS 2017 Christian Knoll Signal Processing and Speech Communication Laboratory Graz University of Technology June 22, 2017 NLSP SS
More informationWe propose a model-based reinforcement
Learning Biped Locomotion Application of Poincaré-Map-Based Reinforcement Learning IMAGESTATE We propose a model-based reinforcement learning (RL) algorithm for biped walking in which the robot learns
More informationReconstructing Null-space Policies Subject to Dynamic Task Constraints in Redundant Manipulators
Reconstructing Null-space Policies Subject to Dynamic Task Constraints in Redundant Manipulators Matthew Howard Sethu Vijayakumar September, 7 Abstract We consider the problem of direct policy learning
More informationFurther Analysis of the Period-Three Route to Chaos in Passive Dynamic Walking of a Compass-Gait Biped Robot
Further Analysis of the Period-Three Route to Chaos in Passive Dynamic Walking of a Compass-Gait Biped Robot Hassène Gritli Direction Générale des Etudes Technologiques Institut Supérieur des Etudes Technologiques
More informationTutorial on Policy Gradient Methods. Jan Peters
Tutorial on Policy Gradient Methods Jan Peters Outline 1. Reinforcement Learning 2. Finite Difference vs Likelihood-Ratio Policy Gradients 3. Likelihood-Ratio Policy Gradients 4. Conclusion General Setup
More informationOpen loop vs Closed Loop. Example: Open Loop. Example: Feedforward Control. Advanced Control I
Open loop vs Closed Loop Advanced I Moor Command Movemen Overview Open Loop vs Closed Loop Some examples Useful Open Loop lers Dynamical sysems CPG (biologically inspired ), Force Fields Feedback conrol
More informationRobotics Part II: From Learning Model-based Control to Model-free Reinforcement Learning
Robotics Part II: From Learning Model-based Control to Model-free Reinforcement Learning Stefan Schaal Max-Planck-Institute for Intelligent Systems Tübingen, Germany & Computer Science, Neuroscience, &
More informationMinimax Differential Dynamic Programming: An Application to Robust Biped Walking
Minimax Differential Dynamic Programming: An Application to Robust Biped Walking Jun Morimoto Human Information Science Labs, Department 3, ATR International Keihanna Science City, Kyoto, JAPAN, 619-0288
More informationLecture 1: A Preliminary to Nonlinear Dynamics and Chaos
Lecture 1: A Preliminary to Nonlinear Dynamics and Chaos Autonomous Systems A set of coupled autonomous 1st-order ODEs. Here "autonomous" means that the right hand side of the equations does not explicitly
More informationBringing the Compass-Gait Bipedal Walker to Three Dimensions
October 14, 2009 IROS 2009, St. Louis, MO 1 1 Bringing the Compass-Gait Bipedal Walker to Three Dimensions Robert D. Gregg* and Mark W. Spong Coordinated Science Laboratory University of Illinois at Urbana-Champaign
More information(Refer Slide Time: 00:32)
Nonlinear Dynamical Systems Prof. Madhu. N. Belur and Prof. Harish. K. Pillai Department of Electrical Engineering Indian Institute of Technology, Bombay Lecture - 12 Scilab simulation of Lotka Volterra
More informationNonlinear systems, chaos and control in Engineering
Nonlinear systems, chaos and control in Engineering Module 1 block 3 One-dimensional nonlinear systems Cristina Masoller Cristina.masoller@upc.edu http://www.fisica.edu.uy/~cris/ Schedule Flows on the
More informationMovement Primitives with Multiple Phase Parameters
Movement Primitives with Multiple Phase Parameters Marco Ewerton 1, Guilherme Maeda 1, Gerhard Neumann 2, Viktor Kisner 3, Gerrit Kollegger 4, Josef Wiemeyer 4 and Jan Peters 1, Abstract Movement primitives
More informationChapter 1 Introduction
Chapter 1 Introduction 1.1 Introduction to Chapter This chapter starts by describing the problems addressed by the project. The aims and objectives of the research are outlined and novel ideas discovered
More informationApplication of Neural Networks for Control of Inverted Pendulum
Application of Neural Networks for Control of Inverted Pendulum VALERI MLADENOV Department of Theoretical Electrical Engineering Technical University of Sofia Sofia, Kliment Ohridski blvd. 8; BULARIA valerim@tu-sofia.bg
More informationA Study of Morphological Computation by using Probabilistic Inference for Motor Planning
A Study of Morphological Computation by using Probabilistic Inference for Motor Planning Elmar Rückert Institute for Theoretical Computer Science Graz University of Technology Graz, 8010, Austria rueckert@igi.tugraz.at
More informationMULTISTABILITY IN A BUTTERFLY FLOW
International Journal of Bifurcation and Chaos, Vol. 23, No. 12 (2013) 1350199 (10 pages) c World Scientific Publishing Company DOI: 10.1142/S021812741350199X MULTISTABILITY IN A BUTTERFLY FLOW CHUNBIAO
More informationDynamics of Heel Strike in Bipedal Systems with Circular Feet
Dynamics of Heel Strike in Bipedal Systems with Circular Feet Josep Maria Font and József Kövecses Abstract Energetic efficiency is a fundamental subject of research in bipedal robot locomotion. In such
More informationUsing Probabilistic Movement Primitives in Robotics
Noname manuscript No. (will be inserted by the editor) Using Probabilistic Movement Primitives in Robotics Alexandros Paraschos Christian Daniel Jan Peters Gerhard Neumann the date of receipt and acceptance
More informationPath Integral Policy Improvement with Covariance Matrix Adaptation
JMLR: Workshop and Conference Proceedings 1 14, 2012 European Workshop on Reinforcement Learning (EWRL) Path Integral Policy Improvement with Covariance Matrix Adaptation Freek Stulp freek.stulp@ensta-paristech.fr
More informationWe consider modeling of point-to-point motions, i.e.
IEEE TRANSACTION ON ROBOTICS, VOL.?, NO.?,? 211 1 Learning Stable Non-Linear Dynamical Systems with Gaussian Mixture Models S. Mohammad Khansari-Zadeh and Aude Billard Abstract This paper presents a method
More informationChimera State Realization in Chaotic Systems. The Role of Hyperbolicity
Chimera State Realization in Chaotic Systems. The Role of Hyperbolicity Vadim S. Anishchenko Saratov State University, Saratov, Russia Nizhny Novgorod, July 20, 2015 My co-authors Nadezhda Semenova, PhD
More informationWorksheet 8 Sample Solutions
Technische Universität München WS 2016/17 Lehrstuhl für Informatik V Scientific Computing Univ.-Prof. Dr. M. Bader 19.12.2016/21.12.2016 M.Sc. S. Seckler, M.Sc. D. Jarema Worksheet 8 Sample Solutions Ordinary
More informationMixed control for perception and action: timing and error correction in rhythmic ball-bouncing
DOI 10.1007/s00221-013-3475-7 RESEARCH ARTICLE Mixed control for perception and action: timing and error correction in rhythmic ball-bouncing I. A. Siegler C. Bazile W. H. Warren Received: 9 October 2012
More informationNonlinear dynamics & chaos BECS
Nonlinear dynamics & chaos BECS-114.7151 Phase portraits Focus: nonlinear systems in two dimensions General form of a vector field on the phase plane: Vector notation: Phase portraits Solution x(t) describes
More informationMethods for Supervised Learning of Tasks
Methods for Supervised Learning of Tasks Alper Denasi DCT 8.43 Traineeship report Coaching: Prof. dr. H. Nijmeijer Technische Universiteit Eindhoven Department Mechanical Engineering Dynamics and Control
More informationComputational Physics (6810): Session 8
Computational Physics (6810): Session 8 Dick Furnstahl Nuclear Theory Group OSU Physics Department February 24, 2014 Differential equation solving Session 7 Preview Session 8 Stuff Solving differential
More informationIterative Motion Primitive Learning and Refinement by Compliant Motion Control
Iterative Motion Primitive Learning and Refinement by Compliant Motion Control Dongheui Lee and Christian Ott Abstract We present an approach for motion primitive learning and refinement for a humanoid
More informationRESEARCH ARTICLE. Sensorimotor Coordination and Metastability in a Situated HKB Model
Connection Science Vol. 99, No. 99, Month 9999, 1 2 RESEARCH ARTICLE Sensorimotor Coordination and Metastability in a Situated HKB Model Bruno Santos a,b, Xabier Barandiaran c, Philip Husbands a, Miguel
More informationThe Nonlinear Pendulum
The Nonlinear Pendulum Evan Sheridan 11367741 Feburary 18th 013 Abstract Both the non-linear linear pendulum are investigated compared using the pendulum.c program that utilizes the trapezoid method for
More informationLearning of Passive Motor Control Strategies with Genetic Algorithms
Stefan Schaal and Dagmar Sternad Department of Brain and Cognitive Sciences and the Artificial Intelligence Laboratory, MIT, 545 Technology Square, Cambridge, MA, 239 (Email: sschaal@ai.mit.edu); Center
More informationIntroduction to Reinforcement Learning. CMPT 882 Mar. 18
Introduction to Reinforcement Learning CMPT 882 Mar. 18 Outline for the week Basic ideas in RL Value functions and value iteration Policy evaluation and policy improvement Model-free RL Monte-Carlo and
More informationSection 9.3 Phase Plane Portraits (for Planar Systems)
Section 9.3 Phase Plane Portraits (for Planar Systems) Key Terms: Equilibrium point of planer system yꞌ = Ay o Equilibrium solution Exponential solutions o Half-line solutions Unstable solution Stable
More informationRhythmic Robot Arm Control Using Oscillators
Rhythmic Robot Arm Control Using Oscillators Matthew M. Williamson MIT AI Lab, 545 Technology Square, Cambridge, MA 2139 http://www.ai.mit.edu/people/matt Abstract This paper presents an approach to robot
More informationTO date, most robots are still programmed by a smart
IEEE ROBOTICS & AUTOMATION MAGAZINE, VOL. 17, NO. 2, JUNE 2010 1 Imitation and Reinforcement Learning Practical Learning Algorithms for Motor Primitives in Robotics by Jens Kober and Jan Peters TO date,
More informationAsymptotically Stable Gait Primitives for Planning Dynamic Bipedal Locomotion in Three Dimensions
May 4, 2010 ICRA, Anchorage, AK 1 1 Asymptotically Stable Gait Primitives for Planning Dynamic Bipedal Locomotion in Three Dimensions Robert D. Gregg *, Timothy W. Bretl, Mark W. Spong Coordinated Science
More informationCoupling Movement Primitives: Interaction with the Environment and Bimanual Tasks
IEEE T-RO Coupling Movement Primitives: Interaction with the Environment and Bimanual Tasks Andrej Gams, Bojan Nemec, Auke J Ijspeert and Aleš Ude Abstract The framework of dynamic movement primitives
More informationNonlinear System Analysis
Nonlinear System Analysis Lyapunov Based Approach Lecture 4 Module 1 Dr. Laxmidhar Behera Department of Electrical Engineering, Indian Institute of Technology, Kanpur. January 4, 2003 Intelligent Control
More informationLearning Operational Space Control
Robotics: Science and Systems 26 Philadelphia, PA, USA, August 16-19, 26 Learning Operational Space Control Jan Peters, Stefan Schaal Departments of Computer Science & Neuroscience University of Southern
More informationMODELING AND CONTINUOUS SONIFICATION OF AFFORDANCES FOR GESTURE-BASED INTERFACES. Yon Visell and Jeremy Cooperstock
MODELING AND CONTINUOUS SONIFICATION OF AFFORDANCES FOR GESTURE-BASED INTERFACES Yon Visell and Jeremy Cooperstock Centre for Intelligent Machines and CIRMMT McGill University, Montreal, Canada yon@cim.mcgill.ca
More informationNeural Network Control of an Inverted Pendulum on a Cart
Neural Network Control of an Inverted Pendulum on a Cart VALERI MLADENOV, GEORGI TSENOV, LAMBROS EKONOMOU, NICHOLAS HARKIOLAKIS, PANAGIOTIS KARAMPELAS Department of Theoretical Electrical Engineering Technical
More informationFundamental problems in the analysis of skilled action (Saltzman 1995) Degrees of freedom Spatiotemporal form Invariance
Fundamental problems in the analysis of skilled action (Saltzman 1995) Degrees of freedom Spatiotemporal form Invariance 1 Degrees of Freedom Actions require the the coordination of a large number of (potentially)
More informationη 5 ξ [k+2] η 4 η 3 η 1 ξ [k]
On-line and Hierarchical Design Methods of Dynamics Based Information Processing System Masafumi OKADA, Daisuke NAKAMURA and Yoshihiko NAKAMURA Dept. of Mechano-Informatics, University of okyo 7-- Hongo
More informationTWO DIMENSIONAL FLOWS. Lecture 5: Limit Cycles and Bifurcations
TWO DIMENSIONAL FLOWS Lecture 5: Limit Cycles and Bifurcations 5. Limit cycles A limit cycle is an isolated closed trajectory [ isolated means that neighbouring trajectories are not closed] Fig. 5.1.1
More informationCOMPARISON OF TWO METHODS TO SOLVE PRESSURES IN SMALL VOLUMES IN REAL-TIME SIMULATION OF A MOBILE DIRECTIONAL CONTROL VALVE
COMPARISON OF TWO METHODS TO SOLVE PRESSURES IN SMALL VOLUMES IN REAL-TIME SIMULATION OF A MOBILE DIRECTIONAL CONTROL VALVE Rafael ÅMAN*, Heikki HANDROOS*, Pasi KORKEALAAKSO** and Asko ROUVINEN** * Laboratory
More informationVariable Impedance Control A Reinforcement Learning Approach
Variable Impedance Control A Reinforcement Learning Approach Jonas Buchli, Evangelos Theodorou, Freek Stulp, Stefan Schaal Abstract One of the hallmarks of the performance, versatility, and robustness
More informationThe Mixed States of Associative Memories Realize Unimodal Distribution of Dominance Durations in Multistable Perception
The Mixed States of Associative Memories Realize Unimodal Distribution of Dominance Durations in Multistable Perception Takashi Kanamaru Department of Mechanical Science and ngineering, School of Advanced
More informationRobotics. Control Theory. Marc Toussaint U Stuttgart
Robotics Control Theory Topics in control theory, optimal control, HJB equation, infinite horizon case, Linear-Quadratic optimal control, Riccati equations (differential, algebraic, discrete-time), controllability,
More informationMinimum Angular Acceleration Control of Articulated Body Dynamics
Minimum Angular Acceleration Control of Articulated Body Dynamics Javier R. Movellan UCSD Abstract As robots find applications in daily life conditions it becomes important to develop controllers that
More informationLecture 6. Lorenz equations and Malkus' waterwheel Some properties of the Lorenz Eq.'s Lorenz Map Towards definitions of:
Lecture 6 Chaos Lorenz equations and Malkus' waterwheel Some properties of the Lorenz Eq.'s Lorenz Map Towards definitions of: Chaos, Attractors and strange attractors Transient chaos Lorenz Equations
More informationLearning Partially Contracting Dynamical Systems from Demonstrations
Learning Partially Contracting Dynamical Systems from Demonstrations Harish Ravichandar Department of Electrical Engineering and Computer Sciences University of Connecticut United States harish.ravichandar@uconn.edu
More informationSaturation of Information Exchange in Locally Connected Pulse-Coupled Oscillators
Saturation of Information Exchange in Locally Connected Pulse-Coupled Oscillators Will Wagstaff School of Computer Science, Georgia Institute of Technology, Atlanta, Georgia 30332, USA (Dated: 13 December
More informationLecture 14: Ordinary Differential Equation I. First Order
Lecture 14: Ordinary Differential Equation I. First Order 1. Key points Maple commands dsolve 2. Introduction We consider a function of one variable. An ordinary differential equations (ODE) specifies
More information2 Discrete growth models, logistic map (Murray, Chapter 2)
2 Discrete growth models, logistic map (Murray, Chapter 2) As argued in Lecture 1 the population of non-overlapping generations can be modelled as a discrete dynamical system. This is an example of an
More informationDynamical Systems with Applications using Mathematica
Stephen Lynch Dynamical Systems with Applications using Mathematica Birkhäuser Boston Basel Berlin Contents Preface xi 0 A Tutorial Introduction to Mathematica 1 0.1 A Quick Tour of Mathematica 2 0.2 Tutorial
More information7 Two-dimensional bifurcations
7 Two-dimensional bifurcations As in one-dimensional systems: fixed points may be created, destroyed, or change stability as parameters are varied (change of topological equivalence ). In addition closed
More informationC 2 Continuous Gait-Pattern Generation for Biped Robots
C Continuous Gait-Pattern Generation for Biped Robots Shunsuke Kudoh 1 Taku Komura 1 The University of Tokyo, JAPAN, kudoh@cvl.iis.u-tokyo.ac.jp City University of ong Kong, ong Kong, taku@ieee.org Abstract
More informationMechanical energy transfer by internal force during the swing phase of running
Available online at www.sciencedirect.com Procedia Engineering 34 (2012 ) 772 777 9 th Conference of the International Sports Engineering Association (ISEA) Mechanical energy transfer by internal force
More informationRobust Low Torque Biped Walking Using Differential Dynamic Programming With a Minimax Criterion
Robust Low Torque Biped Walking Using Differential Dynamic Programming With a Minimax Criterion J. Morimoto and C. Atkeson Human Information Science Laboratories, ATR International, Department 3 2-2-2
More informationMovement Generation with Circuits of Spiking Neurons
no. 2953 Movement Generation with Circuits of Spiking Neurons Prashant Joshi, Wolfgang Maass Institute for Theoretical Computer Science Technische Universität Graz A-8010 Graz, Austria {joshi, maass}@igi.tugraz.at
More informationFeedback Basics. David M. Auslander Mechanical Engineering University of California at Berkeley. copyright 1998, D.M. Auslander
Feedback Basics David M. Auslander Mechanical Engineering University of California at Berkeley copyright 1998, D.M. Auslander 1 I. Feedback Control Context 2 What is Feedback Control? Measure desired behavior
More informationCoupling Movement Primitives: Interaction With the Environment and Bimanual Tasks
816 IEEE TRANSACTIONS ON ROBOTICS, VOL. 30, NO. 4, AUGUST 2014 Coupling Movement Primitives: Interaction With the Environment and Bimanual Tasks Andrej Gams, Bojan Nemec, Auke Jan Ijspeert, Member, IEEE,
More informationLecture 19: Calculus of Variations II - Lagrangian
Lecture 19: Calculus of Variations II - Lagrangian 1. Key points Lagrangian Euler-Lagrange equation Canonical momentum Variable transformation Maple VariationalCalculus package EulerLagrange 2. Newton's
More informationEnhancing a Model-Free Adaptive Controller through Evolutionary Computation
Enhancing a Model-Free Adaptive Controller through Evolutionary Computation Anthony Clark, Philip McKinley, and Xiaobo Tan Michigan State University, East Lansing, USA Aquatic Robots Practical uses autonomous
More informationProbabilistic Model-based Imitation Learning
Probabilistic Model-based Imitation Learning Peter Englert, Alexandros Paraschos, Jan Peters,2, and Marc Peter Deisenroth Department of Computer Science, Technische Universität Darmstadt, Germany. 2 Max
More information