The research paper presents a solution for complex planning problem-solving that ensures completeness. The authors propose a hybrid search method that combines high-level and low-level actions, called a complete subgoal search. This approach guarantees the completeness of low-level search with the practical efficiency of high-level search. The method is applied and evaluated on a recently proposed subgoal search algorithm, demonstrating improved performance and the ability to apply subgoal-level planning where completeness is a critical requirement. This work is beneficial for systems that require complete solutions, such as robotics and autonomous driving.
Publication date: 19 Oct 2023
Project Page: https://arxiv.org/abs/2310.12819
Paper: https://arxiv.org/pdf/2310.12819
