Sequencing operator counts with state-space search

• Ano de defesa: 2020
• Autor(a) principal: Kaizer, Wesley Luciano
• Orientador(a): Pereira, André Grahl
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Palavras-chave em Português: Inteligência artificial; Busca heurística; Algoritmos; Complexidade computacional
• Palavras-chave em Inglês: Classical planning; State-space heuristic search; Integer programming; Operator-counting framework
• Link de acesso: http://hdl.handle.net/10183/211457

A search algorithm with an admissible heuristic function is the most common approach to optimally solve classical planning tasks. Recently DAVIES et al. (2015) introduced the solver OpSeq using Logic-Based Benders Decomposition. In this approach to planning, the master problem is an integer program derived from the operator-counting framework that generates operator counts, i.e., an assignment of integer counts for each task operator. Then, the operator counts sequencing subproblem verifies if a plan satisfying these operator counts exists, or generates a violated constraint to strengthen the master problem. In OpSeq, the subproblem is solved by a SAT solver. In this thesis, we show that this subproblem can be better solved by state-space search. We introduce OpSearch, an A -based algorithm to solve the operator counts sequencing problem: it either finds an optimal plan, or uses the frontier of the search, i.e., the set of generated but yet unexpanded states, to derive a violated constraint. We show that using a standard search framework has three advantages: i) the search scales better than a SAT-based approach for solving the operator counts sequencing, ii) explicit information in the search frontier can be used to derive stronger constraints, and iii) stronger heuristics generate more informed constraints. We present results using the benchmark of the International Planning Competition, showing that this approach solves more planning tasks, using less memory. On tasks solved by both methods, OpSearch usually requires solving fewer operator counts sequencing problems than OpSeq, evidencing the stronger constraints generated by OpSearch.