Differences
This shows you the differences between two versions of the page.
Both sides previous revisionPrevious revision | Last revisionBoth sides next revision | ||
doc:common-sense_reasoning_about_object_locations [2014/06/05 11:38] – external edit 127.0.0.1 | doc:common-sense_reasoning_about_object_locations [2014/11/28 12:43] – admin | ||
---|---|---|---|
Line 1: | Line 1: | ||
====== Commonsense knowledge about object locations ====== | ====== Commonsense knowledge about object locations ====== | ||
+ | \\ | ||
+ | ^ This page describes the ' | ||
+ | \\ | ||
Commonsense knowledge about typical object locations was acquired by the Open Mind Indoor Common Sense (OMICS [[http:// | Commonsense knowledge about typical object locations was acquired by the Open Mind Indoor Common Sense (OMICS [[http:// | ||
===== Getting Started ===== | ===== Getting Started ===== | ||
- | First you have to build the rospackage | + | You have to build the //knowrob_omics// package in // |
< | < | ||
- | | ||
| | ||
</ | </ | ||
Line 12: | Line 14: | ||
===== Queries about object locations ===== | ===== Queries about object locations ===== | ||
To query the probability of finding an object in a given room type you can use the following query: | To query the probability of finding an object in a given room type you can use the following query: | ||
- | < | + | < |
- | ?- probability_given(knowrob:' | + | ?- probability_given(knowrob:' |
- | Obj = ' | + | Obj = knowrob:'CitrusFruit', |
- | Pr = 0.003865979381443299 | + | Pr = 0.0038510911424903724 |
- | Obj = 'http://ias.cs.tum.edu/ | + | Obj = knowrob:' |
- | Pr = 0.006443298969072165 | + | Pr = 0.006418485237483954 ; |
+ | Obj = knowrob:' | ||
+ | Pr = 0.037227214377406934 ; | ||
+ | Obj = knowrob:' | ||
+ | Pr = 0.006418485237483954 | ||
</ | </ | ||
If you are interested in what type of room you could find a given object use the following query: | If you are interested in what type of room you could find a given object use the following query: | ||
< | < | ||
- | ?- bayes_probability_given(knowrob:' | + | ?- bayes_probability_given(knowrob:' |
- | knowrob:' | + | Room = knowrob:' |
- | Room = 'http:// | + | Pr = 0.20109219415371862 |
- | Pr = 0.21657432923639094 | + | Room = knowrob:' |
- | Room = 'http:// | + | Pr = 0.046915489441673196 |
- | Pr = 0.05065150074968034 | + | Room = knowrob:' |
- | Room = 'http:// | + | Pr = 0.06843635615803703 |
- | Pr = 0.07410126961527308 | + | Room = knowrob:' |
- | Room = ' | + | Pr = 0.07637777632654648 |
- | Pr = 0.08278900467361545 ; | + | |
- | Room = 'http:// | + | |
- | Pr = 0.08394689285086883 ; | + | |
- | Room = ' | + | |
- | Pr = 0.24008811355348483 | + | |
</ | </ | ||
You can see the processed database table entries with: | You can see the processed database table entries with: | ||
< | < | ||
- | ?- rdf_has(Entry, | + | ?- rdf_has(Entry, |
- | | + | |
- | | + | |
- | Entry = 'http:// | + | Entry = knowrob:' |
- | Obj = ' | + | Obj = knowrob:'CitrusFruit', |
- | Room = ' | + | Room = knowrob:'Kitchen' |
- | Entry = 'http:// | + | Entry = knowrob:' |
- | Obj = ' | + | Obj = knowrob:'Bowl-Eating', |
- | Room = 'http:// | + | Room = knowrob:'Kitchen' |
+ | Entry = knowrob:' | ||
+ | Obj = knowrob:' | ||
+ | Room = knowrob:' | ||
+ | Entry = knowrob:' | ||
+ | Obj = knowrob:' | ||
+ | Room = knowrob:' | ||
</ | </ | ||
Get the list of Room types: | Get the list of Room types: | ||
- | < | + | < |
- | ?- findall(Room, | + | ?- findall(Room, |
- | | + | rdf_has(Entry, |
- | | + | |
+ | RoomSet =[knowrob:' | ||
+ | knowrob:' | ||
+ | knowrob:' | ||
+ | knowrob:' | ||
+ | knowrob:' | ||
+ | knowrob:' | ||
+ | knowrob:' | ||
+ | knowrob:' | ||
+ | knowrob:' | ||
+ | knowrob:' | ||
</ | </ | ||
+ | The findall may produce duplicate results, which are eliminated by the // | ||
+ |