خط مشی دسترسیدرباره ما
ثبت نامثبت نام
راهنماراهنما
فارسی
ورودورود
صفحه اصلیصفحه اصلی
جستجوی مدارک
تمام متن
منابع دیجیتالی
رکورد قبلیرکورد بعدی
Document Type : Latin Dissertation
Language of Document : English
Record Number : 150142
Doc. No : ET21934
Main Entry : Kevin R. Dixon
Title Proper : Inferring User Intent for Learning by Observation
Note : This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract : Despite the numerous advances in human-robot interaction, most development systems still require that users havesubstantial knowledge of procedural-programming techniques as well as the specific robot system at hand. Forthe vast majority of the population, this effectively precludes the use of robots in most cases. If robots are tomake headway into everyday situations, then users must be able to program robots in a more natural and intuitivemanner. This dissertation explores a method of programming robots to automate motor tasks by inferring theintent of users based on demonstrations of a task. In order to understand such a system, we decompose it intosimpler components: modeling user subgoal selection and the response of users to different conditions.We have developed a learning algorithm that constructs a statistical model of user subgoal selection based onprevious observations. After deriving the algorithm, we provide theoretical guarantees about the model. To vali-date the theoretical underpinnings of the algorithm, we isolate the performance of modeling user subgoal selectionby removing extraneous factors such as sensor noise and environment considerations. To this end, we presentexperimental results in predicting the waypoints of manipulator-robot programs. We show that the algorithmproduces submillimeter prediction errors on real-world data.We hypothesize about the response of users to different conditions with a model of sequenced linear dynamicalsystems. We first develop the concept that a single dynamical system can represent a simple trajectory using aclosed-form least-squares procedure. Since our approach is based on the least-squares principle, it is simple tocombine multiple demonstrations, giving the system a better generalization of "what the user would have done"in novel conditions. To represent more complicated trajectories, we segment it and represent each segment by asingle dynamical system.These algorithms form the core of a mobile-robot system that learns motor skills by observing users demon-strating a task. From these observations, the system extracts task subgoals and automatically associates them withobjects in the environment, so that as the objects move, the subgoals are updated accordingly. This system canlearn from multiple demonstrations, as well as demonstrations performed in different environment configurations.In laboratory experiments, we show that the system accurately infers user intent..,..tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject : Electericl tess
: برق
electronic file name : TL45128.pdf
Title and statement of responsibility and : Inferring User Intent for Learning by Observation [Thesis]
 
 
 
(در صورت عدم وضوح تصویر اینجا را کلیک نمایید)