A Learning Based Approach for Planning with Safe Actions

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21730%2F20%3A00347586" target="_blank" >RIV/68407700:21730/20:00347586 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1007/978-3-030-58814-4_7" target="_blank" >https://doi.org/10.1007/978-3-030-58814-4_7</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-030-58814-4_7" target="_blank" >10.1007/978-3-030-58814-4_7</a>

Jazyk výsledku

angličtina

Název v původním jazyce

A Learning Based Approach for Planning with Safe Actions

Popis výsledku v původním jazyce

Given a configuration involving some objects in an environment, the planning problem, considered in this paper, is to find a plan that rearranges these objects so as to place a new object. The challenging aspect here involves deciding when an object can be placed on top of another object. Here only defining standard planning operators would not suffice. For instance, using these operators we can come up with actions that may be performed at a state but it should not be performed. So we introduce the notion of safe actions whose outcomes are consistent with the laws of physics, commonsense, and common practice. A safe action can be performed if a robot performing the action knows the knowledge of the situation. We developed a knowledge engine using a supervised learning technique. However, unlike the common task of learning functions, our approach is to learn predicates--that evaluate to binary values. By learning such a predicate a robot would be able to decide whether or not an object A can be placed on top of another object B. We give a method to handle new objects for which the predicates have not been learned. We suggest a nondeterministic planning algorithm to synthesize plans that contain only safe actions. Experimental results show the efficacy of our approach.

Název v anglickém jazyce

A Learning Based Approach for Planning with Safe Actions

Popis výsledku anglicky

Given a configuration involving some objects in an environment, the planning problem, considered in this paper, is to find a plan that rearranges these objects so as to place a new object. The challenging aspect here involves deciding when an object can be placed on top of another object. Here only defining standard planning operators would not suffice. For instance, using these operators we can come up with actions that may be performed at a state but it should not be performed. So we introduce the notion of safe actions whose outcomes are consistent with the laws of physics, commonsense, and common practice. A safe action can be performed if a robot performing the action knows the knowledge of the situation. We developed a knowledge engine using a supervised learning technique. However, unlike the common task of learning functions, our approach is to learn predicates--that evaluate to binary values. By learning such a predicate a robot would be able to decide whether or not an object A can be placed on top of another object B. We give a method to handle new objects for which the predicates have not been learned. We suggest a nondeterministic planning algorithm to synthesize plans that contain only safe actions. Experimental results show the efficacy of our approach.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2020

Kód důvěrnosti údajů

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

Název statě ve sborníku

Computational Science and Its Applications – ICCSA 2020

ISBN

978-3-030-58813-7

ISSN

0302-9743

e-ISSN

1611-3349

Počet stran výsledku

13

Strana od-do

93-105

Název nakladatele

Springer Science+Business Media

Místo vydání

Berlin

Místo konání akce

Cagliari

Datum konání akce

1. 7. 2020

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000719714800007