Interdisciplinary approaches to robot learning / edited by J. Demiris, A. Birk.

Format:

Book

Contributor:

Demiris, John, 1969-

Birk, Andreas, 1969-

Series:

World Scientific series in robotics and intelligent systems ; v. 24.

World Scientific series in robotics and intelligent systems ; vol. 24

Language:

English

Subjects (All):

Robots--Control systems.

Robots.

Machine learning.

Physical Description:

1 online resource (220 p.)

Place of Publication:

Singapore ; River Edge, N.J. : World Scientific, c2000.

Language Note:

English

Summary:

Robots are being used in increasingly complicated and demanding tasks, often in environments that are complex or even hostile. Underwater, space and volcano exploration are just some of the activities that robots are taking part in, mainly because the environments that are being explored are dangerous for humans. Robots can also inhabit dynamic environments, for example to operate among humans, not just in factories, but also taking on more active roles. Recently, for instance, they have made their way into the home entertainment market. Given the variety of situations that robots will be placed in, learning becomes increasingly important. Robot learning is essentially about equipping robots with the capacity to improve their behaviour over time, based on their incoming experiences. The papers in this volume present a variety of techniques. Each paper provides a mini-introduction to a subfield of robot learning. Some also give a fine introduction to the field of robot learning as a whole. There is one unifying aspect to the work reported in the book, namely its interdisciplinary nature, especially in the combination of robotics, computer science and biology. This approach has two important benefits: first, the study of learning in biological systems can provide robot learning scientists and engineers with valuable insights into learning mechanisms of proven functionality and versatility; second, computational models of learning in biological systems, and their implementation in simulated agents and robots, can provide researchers of biological systems with a powerful platform for the development and testing of learning theories.

Contents:

CONTENTS; Program Committee; INTERDISCIPLINARY APPROACHES TO ROBOT LEARNING: INTRODUCTION; 1 Overview; 2 Robot Learning - what and why?; 3 Roadmap; Acknowledgements; References; BOOTSTRAPPING THE DEVELOPMENTAL PROCESS: THE FILTER HYPOTHESIS; 1 Introduction; 2 The Filter Hypothesis; 3 The neural model; 3.1 SAM model; 3.2 Noise and time-constants; 3.3 Temporal learning rule; 3.4 Network; 4 Categorization; 4.1 Experimental setup; 4.2 Experiment; 5 Self-Entrainment; 5.1 Experimental setup; 5.2 Experiment; 6 Conclusion; Acknowledgments; References; BIOMIMETIC GAZE STABILIZATION; 1 Introduction

2 Biological VOR-OKR Models2.1 The VOR and the OKR; 2.2 An Overview of Previous Learning Models of VOR and OKR; 3 A Computational Model of VOR-OKR; 3.1 Research Objectives; 3.2 A Simple VOR System; 3.3 Merging the VOR with an OKR-like Pathway; 3.4 Adding a Learning Network; 3.5 Fast and Stable Learning; 3.6 Learning with the Delayed-Error Signal; 3.7 Learning in Nonlinear Systems; 4 Experimental Setup; 4.1 Mimicing the eye-muscle system; 4.2 Retinal slip acquisition; 5 Experimental Results; 6 Conclusion; References; EXPERIMENTS AND MODELS ABOUT COGNITIVE MAP LEARNING FOR MOTIVATED NAVIGATION

1 Introduction2 A simple homing mechanism; 3 Learning a ""cognitive map""; 4 Optimizing the use of the map: learning the best route; 4.1 Learning to avoid ""dangerous"" areas; 4.2 Learning to choose between goals; 4.3 Learning new paths in a changing environment; 5 Conclusion; Acknowledgments; Appendix; A Navigation algorithm; B Proof of the equivalence between the neural planning algorithm and Bellman and Ford 's shortest path finding algorithm; References; LEARNING SELECTION OF ACTION FOR CORTICALLY-INSPIRED ROBOT CONTROL; 1 Introduction; 2 The cortical model; 2.1 The map level

2.2 The maxicolumn level3 Learning mechanisms; 3.1 Non temporal learning mechanisms; 3.2 Temporal learning mechanism; 3.3 Functioning rules: Using learned regularities; 3.4 Formalization of the temporal learning mechanism in the model; 4 Experiments; 4.1 The framework; 4.2 Architecture of the animat; 4.3 Experimental results; 5 Discussion; References; TRANSFERRING LEARNED KNOWLEDGE IN A LIFELONG LEARNING MOBILE ROBOT AGENT; 1 Introduction; 2 Lifelong Learning; 2.1 SMTL; 2.2 EBNN; 3 Sequential Multitask Learning - Experimental Results; 3.1 Domain 1; 3.2 Domain 2: SYN-XAVIER; 3.3 Results

4 Transferring multiple types of knowledge - Experimental Results5 Discussion; 6 Related Work; 7 Conclusion; Acknowledgments; References; OF HUMMINGBIRDS AND HELICOPTERS: AN ALGEBRAIC FRAMEWORK FOR INTERDISCIPLINARY STUDIES OF IMITATION AND ITS APPLICATIONS; 1 Introduction; 2 Imitation; 2.1 Action-Level, Program-Level, and Effect-Level Imitation; 3 Dissimiliar Bodies; 3.1 Imitation for Some Purpose; 3.2 Abstraction of Action; 4 Affordances from a Situation; 5 The Correspondence Problem and Attempted Imitation: Overview of Rigorous Foundations; 6 Imitation and Learning

6.1 Learning by Imitation

Notes:

Description based upon print version of record.

Includes bibliographical references.

ISBN:

9786611934163

9781281934161

128193416X

9789812792747

9812792740

OCLC:

824362752