Online-Learning in Humanoid Robots
- Art: Diplomarbeit
- Autor: Jörg Conradt
- Abgabedatum: Februar 2001
- Umfang: 114 Seiten
- Dateigröße: 2,1 MB
- Note: 1,0
- Institution / Hochschule: Technische Universität Berlin Deutschland
- ISBN (eBook): 978-3-8324-4858-5
-
ISBN (Paperback) :
978-3-8324-4858-5 P - ISBN (CD) :978-3-8324-4858-5 CD
- Sprache: Englisch
- Prämierung:
- Arbeit zitieren: Conradt, Jörg Februar 2001: Online-Learning in Humanoid Robots, Hamburg: Diplomica Verlag
- Schlagworte: Robotik, Robot-Motion-Control, Neuroinformatics
In den Warenkorb
58,00 €
Diplomarbeit von Jörg Conradt
Abstract:
Humanoid Robotic Systems have gained an increasing significance in the research world within the last few years. Just five years ago, there were hardly any human-like robots in the world, and those available did not represent human properties at all. They neither looked nor behaved like human beings. Today, a variety of research groups around the world is starting to work on topics related to humanoid robots, and it is very likely that these robots will become important within the upcoming decades even beyond the realm of science.
Trying to determine what humanoid robots are, a first draft of a definition might read as follows: such robots are to be called humanoid robots which - to some extent - are able to live and interact with the everyday human world, and represent certain human features, like cognitive or acting abilities. The main strength of such humanoid robots lies in their ability to operate in surroundings that have been designed for humans in the first place. Humanoid robots can be imagined to become useful assistants for every-day life in areas as diverse as:
- Rescue and clearing of dangerous situations.
- Janitorial services, Housekeeping.
- Security services.
- Care-taking in hospitals, recreational facilities.
- Entertainment.
In all these fields, close human interaction is a core issue and can be regarded as the minimum common basis. The interaction happens on many different levels, from physical touch to gesture recognition and the processing of spoken language. On cognitive issues like the two last named, much research has been done in the past few years. One has, however, to keep in mind that also the physical appearance, e.g. smoothness of motions, is an important issue when designing humanoid robots.
Table of Contents:
| FOREWORD | 1 | |
| 1. | INTRODUCTION | 2 |
| 1.1 | INTRODUCING THE AREA OF HUMANOID ROBOTICS | 2 |
| 1.2 | MECHANICAL DESIGN FOR HUMANOID ROBOTS | 3 |
| 1.3 | CONTROLLING HUMANOID ROBOTS | 4 |
| 1.4 | EXAMPLES OF TODAY'S HUMANOID ROBOTS | 5 |
| 1.5 | THE PURPOSE OF THE THESIS | 9 |
| 2. | A BRIEF RECAPITULATION OF BASIC ROBOT CONTROL | 10 |
| 2.1 | INTRODUCTION TO ROBOT CONTROL | 10 |
| 2.2 | CONTROLLING THE EXECUTION OF DESIRED TRAJECTORIES | 12 |
| 2.3 | THE FEEDBACK CONTROL FUNCTION | 14 |
| 2.4 | THE FEED-FORWARD CONTROL FUNCTION | 17 |
| 2.5 | ESTIMATING DYNAMICS USING RIGID BODY ASSUMPTIONS | 18 |
| 2.6 | RECAPITULATION | 22 |
| 2.7 | CONTROL OF HUMANOID ROBOTS | 23 |
| 3. | INTRODUCTION TO ROBOT LEARNING | 25 |
| 3.1 | GENERAL REMARKS ON ROBOT LEARNING | 25 |
| 3.2 | THE BIAS / VARIANCE TRADEOFF | 27 |
| 3.3 | GLOBAL VERSUS LOCAL LEARNING STRATEGIES | 29 |
| 3.4 | THE CURSE OF DIMENSIONALITY | 31 |
| 3.5 | ONLINE-LEARNING | 32 |
| 3.6 | LEARNING INVERSE DYNAMICS | 34 |
| 3.7 | RESULTS | 34 |
| 4. | THE LEARNING ALGORITHM LWPR | 36 |
| 4.1 | ADVANTAGES OF LEARNING APPROACHES | 36 |
| 4.2 | DESIRED ALGORITHMIC PROPERTIES | 38 |
| 4.3 | INPUT DATA PREPROCESSING | 45 |
| 4.4 | PREDICTING OUTPUT DATA USING LWPR | 47 |
| 4.5 | LEARNING PARAMETERS | 50 |
| 4.6 | THE FINAL LWPR ALGORITHM | 57 |
| 4.7 | PROPERTIES OF LWPR | 59 |
| 5. | EVALUATION OF LWPR'S PERFORMANCE ON ARTIFICIAL DATA | 61 |
| 5.1 | INTRODUCTION | 61 |
| 5.2 | ONE DIMENSIONAL FUNCTION FITTING | 61 |
| 5.3 | TWO-DIMENSIONAL FUNCTION FITTING WITH SHIFTING INPUT DISTRIBUTIONS | 65 |
| 5.4 | LIMITING THE COMPUTATIONAL COMPLEXITY | 68 |
| 5.5 | DISCUSSION | 69 |
| 6. | THE HUMANOID ROBOT | 71 |
| 6.1 | HARDWARE FOR ROBOT CONTROL | 71 |
| 6.2 | SOFTWARE FOR ROBOT CONTROL | 74 |
| 6.3 | THE SARCOS ROBOTIC ARM | 77 |
| 7. | EVALUATION OF THE ARM'S MOTION | 81 |
| 7.1 | GENERATING TRAJECTORIES FOR LEARNING | 81 |
| 7.2 | LEARNING INVERSE DYNAMICS | 85 |
| 7.3 | LEARNING RESULTS ON SAMPLED DATA | 87 |
| 7.4 | VERIFYING THE RESULTS ON REAL MOTION | 91 |
| 8. | CONCLUSION | 96 |
| 8.1 | DISCUSSION | 96 |
| 8.2 | BRIEF INTRODUCTION TO USING LWPR FOR INVERSE KINEMATICS ESTIMATION | 97 |
| 8.2 | FUTURE RESEARCH | 99 |
| REFERENCES | 101 | |
| SHORT GERMAN SUMMARY | 104 | |
| EIDESSTATTLICHE ERKLÄRUNG | 105 |
In den Warenkorb
58,00 €
Link zur Arbeit:
http://www.diplom.de/ean/9783832448585
Arbeit zitieren:
Conradt, Jörg Februar 2001: Online-Learning in Humanoid Robots, Hamburg: Diplomica Verlag
Schlagworte:
Robotik, Robot-Motion-Control, Neuroinformatics



