Xu / C. Lee Human Behavior Learning and Transfer

Erscheinungsjahr 2010
ISBN: 978-1-4200-3697-8
Verlag: Taylor & Francis
Format: PDF
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)

E-Book, Englisch, 352 Seiten

Bridging the gap between human-computer engineering and control engineering, Human Behavior Learning and Transfer delineates how to abstract human action and reaction skills into computational models. The authors include methods for modeling a variety of human action and reaction behaviors and explore processes for evaluating, optimizing, and transferring human skills. They also cover modeling continuous and discontinuous human control strategy and discuss simulation studies and practical real-life situations.

The book examines how to model two main aspects of human behavior: reaction skills and action skills. It begins with a discussion of the various topics involved in human reaction skills modeling. The authors apply machine learning techniques and statistical analysis to abstracting models of human reaction control strategy. They contend that such models can be learned sufficiently to emulate complex human control behaviors in the feedback loop.

The second half of the book explores issues related to human action skills modeling. The methods presented are based on techniques for reducing the dimensionality of data sets, while preserving as much useful information as possible. The modeling approaches developed are applied in real-life applications including navigation of smart wheel chairs and intelligent surveillance.

Written in a consistent, easily approachable style, the book includes in-depth discussions of a broad range of topics. It provides the tools required to formalize human behaviors into algorithmic, machine-coded strategies.

Zielgruppe

Ergonomics and Human Factors Engineers (Human Computer-Interaction), Industrial Engineers, Manufacturing Engineers, Production Engineers, Computer Engineers, those working in Computer Science and Computational Intelligence. Mechanical Engineers (controls & robotics), Electrical Engineers, Process Engineers, Rehabilitation and Biomedical Engineers, Aerospace Engineers

Autoren/Hrsg.

Xu, Yangsheng

C. Lee, Ka Keung

Fachgebiete

• Technische Wissenschaften Technik Allgemein Ergonomie, Barrierefreiheit
• Technische Wissenschaften Elektronik | Nachrichtentechnik Elektronik Überwachungstechnik

Inhaltsverzeichnis

INTRODUCTION
Motivation
Overview

INTRODUCTION TO HUMAN REACTION SKILL MODELING
Motivation
Related Work

LEANING OF HUMAN CONTROL STRATEGY: CONTINUOUS AND DISCONTINUOUS
Experimental Design
Cascade Neural Networks with Kalman Filtering
HCS Models: Continuous Control
HCS Models: Discontinuous Control

VALIDATION OF HUMAN CONTROL STRATEGY MODELS
Need for Model Validation
Stochastic Similarity Measure
Human-to-Model Comparisons

EVALUATION OF HUMAN CONTROL STRATEGY
Introduction
Obstacle Avoidance
Tight Turning
Transient Response
Time Delay
Passenger Comfort
Driving Smoothness
Summary

PERFORMANCE OPTIMIZATION OF HUMAN CONTROL STRATEGY
Introduction
Simultaneously Perturbed Stochastic Approximation
Iterative Optimization Algorithm
Model Optimization and Performance Analysis
Summary

TRANSFER OF HUMAN CONTROL STRATEGY
Introduction
Model Transfer Based on Similarity Measure
Model Compensation
Summary

TRANSFERRING HUMAN NAVIGATIONAL SKILLS TO SMART WHEELCHAIR
Introduction
Methodology
Experimental Study
Analysis
Conclusion

INTRODUCTION TO HUMAN ACTION SKILL MODELING
Learning Action Models
Dimension Reduction Formulation
Related Research

GLOBAL PARAMETRIC METHODS FOR DIMENSION REDUCTION
Introduction
Parametric Methods for Global Modeling
An Experimental Data Set
PCA for Modeling Performance Data
NLPCA
SNLPCA
Comparison
Characterizing NLPCA Mappings

LOCAL METHODS FOR DIMENSION REDUCTION
Introduction
Non-parametric Methods for Trajectory Fitting
Scatter Plot Smoothing
Action Recognition Using Smoothing Splines
An Experiment Using Spline Smoothing
Principal Curves
Expanding the One-Dimensional Representation
Branching
Over-Fitting

A SPLINE SMOOTHER IN PHASE SPACE FOR TRAJECTORY FITTING
Smoothing with Velocity Information
Problem Formulation
Solution
Notes on Computation and Complexity
Similar Parameterizations
Multi-Dimensional Smoothing
Estimation of Variances
Windowing Variance Estimates
The Effect of Velocity Information
Cross-Validation

ANALYSIS OF HUMAN WALKING TRAJECTORIES FOR SURVEILLANCE
Introduction
System Overview
Background Subtraction
Global Trajectory Similarity Estimation
Trajectory Normality Classifier
Experiment 1: Trajectory Normality Classifier
Further Analysis on Global Trajectory Similarity Based on LCSS
Methodology Used in Boundary Modeling
LCSS Boundary Limit Establishment
Experiment 2: Boundary Modeling
Discussion
Conclusions

MODELING OF FACIAL AND FULL-BODY ACTIONS
Facial Expression Intensity Modeling
Full-Body Action Modeling

CONCLUSIONS