| VRSTA GRADIVA | analitična raven (sestavni del), tekstovno gradivo, tiskano, 1.03 - kratki znanstveni prispevek |
| DRŽAVA IZIDA | Slovenija |
| LETO IZIDA | 2006 |
| JEZIK BESEDILA/IZVIRNIKA | angleški |
| PISAVA | latinica |
| AVTOR | Kaučič, Branko - avtor |
| ODGOVORNOST | Žalik, Borut - avtor |
| NASLOV | Comparison of heuristics for terrain guarding |
| V PUBLIKACIJI | Contributions to geometric modelling and multimedia. - ISSN 1580-5689. - ǂVol. ǂ6, ǂno. ǂ2 (2006), str. 1-17. |
| KRATKA VSEBINA | Guarding polyhedral terrains is a well known problem. It has a wide area of applications and many problems have been identified - problems with only one guard and more complex problems with many guards. Many of them represent NP-hard problems and since 1986 when first solution for terrain guarding was proposed, several heuristics have been developed. Extensive set of heuristics was published in different sources and consequently a realistic comparison between them (which is better and which worse) is missing. ln this contribution we present the results of comprehensive testing of terrain guarding. Basic terrain guarding problem called watch lower problem is considered and the problem of guarding where number of guards is upward limited. To the authors' knowledge, all known heuristics are used, and comparison is done on actual terrain surfaces. The most important result of this contribution is showingthat long lasting belief that the best results can be obtained by greedy add heuristics is wrong. Better results can be obtained by our new heuristic called improved greedy add and i-technique. // Varovanje poliedrskih površij predstavlja sožitje mnogih raziskovalnih področij. Ima precej področij uporabe in podanih je precej nalog varovanja terenov - od nalog z enim samim stražarjem, do kompleksnejših z večstražarji. Večina tehprestavlja NP-težek zalogaj in od leta 1986, ko je bila predstavljena prva rešitev, je bilo razvitih kar nekaj hevristik. Obsežni množici hevristik, objavljenih v najrazličnejših virih manjka primerjava njihove medsebojne konkurenčnosti. V prispevku bomo podali rezultate testiranja varovanja terenov za osnovni problem varovanja terenov, imenovan watch lower problem in problem varovanja, kjer smo omejeni z največjim številom uporabljenih stražarjev. Uporabili bomo vse znane hevristike, testiranje pa izvedli izključno na dejanskih topografskih površjih. Kot najpomembnejši rezultat prispevka bomo pokazali, da dolgoletno prepričanje, da najboljše rezultate dobimo s hevristiko greedy add ne drži. Boljše rezultate namreč dobimo z našo lastno hevristiko improved greedy add oziroma i-tehniko odstranjevanjaredundantnih stražarjev. |
| OPOMBE | Bibliografija: str. 16-17 |
| OSTALI NASLOVI | Comparison of heuristics for guarding polyhedral terrains // Rezultati varovanja realnih poliedrskih terenov |
| PREDMETNE OZNAKE | // poliedrski teren // računalniška geometrija // kombinatorična optimizacija // GIS |
| UDK | 004.9 |