Safe Exploration for Reinforcement Learning in Real Unstructured Environments

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F15%3A00230205" target="_blank" >RIV/68407700:21230/15:00230205 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Safe Exploration for Reinforcement Learning in Real Unstructured Environments

Popis výsledku v původním jazyce

In USAR (Urban Search and Rescue) missions, robots are often required to operate in an unknown environment and with imprecise data coming from their sensors. However, it is highly desired that the robots only act in a safe manner and do not perform actions that could probably make damage to them. To train some tasks with the robot, we utilize reinforcement learning (RL). This machine learning method however requires the robot to perform actions leading to unknown states, which may be dangerous. We develop a framework for training a safety function which constrains possible actions to a subset of really safe actions. Our approach utilizes two basic concepts. First, a "core" of the safety function is given by a cautious simulator and possibly also by manually given examples. Second, a classifier training phase is performed (using Neyman-Pearson SVMs), which extends the safety function to the states where the simulator fails to recognize safe states.

Název v anglickém jazyce

Safe Exploration for Reinforcement Learning in Real Unstructured Environments

Popis výsledku anglicky

In USAR (Urban Search and Rescue) missions, robots are often required to operate in an unknown environment and with imprecise data coming from their sensors. However, it is highly desired that the robots only act in a safe manner and do not perform actions that could probably make damage to them. To train some tasks with the robot, we utilize reinforcement learning (RL). This machine learning method however requires the robot to perform actions leading to unknown states, which may be dangerous. We develop a framework for training a safety function which constrains possible actions to a subset of really safe actions. Our approach utilizes two basic concepts. First, a "core" of the safety function is given by a cautious simulator and possibly also by manually given examples. Second, a classifier training phase is performed (using Neyman-Pearson SVMs), which extends the safety function to the states where the simulator fails to recognize safe states.

Druh

D - Stať ve sborníku

CEP obor

JD - Využití počítačů, robotika a její aplikace

OECD FORD obor

—

Projekt

<a href="/cs/project/GA14-13876S" target="_blank" >GA14-13876S: Strojové vnímání pro dlouhodobou autonomii mobilních robotů</a><br>

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2015

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

CVWW 2015: Proceedings of the 20th Computer Vision Winter Workshop

ISBN

978-3-85125-388-7

ISSN

—

e-ISSN

—

Počet stran výsledku

9

Strana od-do

85-93

Název nakladatele

Graz University of Technology

Místo vydání

Graz

Místo konání akce

Seggau

Datum konání akce

9. 2. 2015

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—