Publicação

Poolimmersion: a realistic physically-based billiards simulation

Ver documento

Detalhes bibliográficos
Resumo:A realistic CG billiards simulation can be accomplished when based on a physical foundation which provides data for a 3D simulation. Artificial opponents are always a welcome addition when it is a challenging opponent. This paper reports on our approach to develop such a simulation. To achieve the required accuracy when computing the shots and ball motion we used as a starting development point the FastFiz library. FastFiz is a physically accurate library designed for billiard simulation, however, it does not fully serve the purposes of a real-time CG simulation and several enhancements were required for it to serve our purpose. A novel approach for an artificial player, based on simulated annealing, is also presented and we compare it to the winners of international computational billiards competitions. Our simulator is fully configurable and can be used both as a game and as a training tool.
Autores principais:Silva, Alexandre
Outros Autores:Malhadas, Daniel; Fernandes, António Ramires
Assunto:Artificial intelligence Mathematical model Computational modeling Physics Games Libraries Artificial intelligence player Computational billiards FastFiz Simulated annealing Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática
Ano:2017
País:Portugal
Tipo de documento:comunicação em conferência
Tipo de acesso:acesso restrito
Instituição associada:Universidade do Minho
Idioma:inglês
Origem:RepositóriUM - Universidade do Minho
Descrição
Resumo:A realistic CG billiards simulation can be accomplished when based on a physical foundation which provides data for a 3D simulation. Artificial opponents are always a welcome addition when it is a challenging opponent. This paper reports on our approach to develop such a simulation. To achieve the required accuracy when computing the shots and ball motion we used as a starting development point the FastFiz library. FastFiz is a physically accurate library designed for billiard simulation, however, it does not fully serve the purposes of a real-time CG simulation and several enhancements were required for it to serve our purpose. A novel approach for an artificial player, based on simulated annealing, is also presented and we compare it to the winners of international computational billiards competitions. Our simulator is fully configurable and can be used both as a game and as a training tool.