核心概念
The author presents a novel contract-based synthesis framework for distributed logical controllers in two-objective parity games, emphasizing the importance of permissive templates. The approach allows for efficient, sound, and complete negotiation processes.
摘要
The content introduces a contract-based synthesis framework for distributed logical controllers in two-objective parity games. It highlights the significance of permissive templates in enabling efficient negotiation mechanisms. The proposed approach aims to provide sound and complete solutions to distributed synthesis problems by locally co-synthesizing contracts and controllers.
The framework relies on the concept of permissive templates to collect an infinite number of controller candidates efficiently. It emphasizes adaptability, computability, and compositionality to achieve an effective negotiation process for contract-based distributed logical control.
By comparing their prototype tool CoSMo with existing tools on a robot motion planning benchmark suite, the authors showcase the superior performance of their approach. The content delves into examples illustrating how strategy templates can be composed and adapted flexibly to meet changing requirements in complex systems.
統計資料
We consider the problem of computing distributed logical controllers for two interacting system components.
Our framework relies on permissive templates that collect an infinite number of controller candidates efficiently.
The proposed negotiation mechanism refines candidates until a solution is found.
The authors showcase the superior performance of their approach through comparisons with existing tools.
Strategy templates are composable and allow for robust control implementation under various scenarios.
引述
"The main contribution of this paper is a new algorithm to solve two-objective parity games arising from distributed logical control problems."
"Our framework relies on the concept of permissive templates which collect an infinite number of controller candidates efficiently."