让多智能体系统更可信，关键在实时感知与协同决策。
SymAware: A Software Development Framework for Trustworthy Multi-Agent Systems with Situational Awareness
- 预定义感知数据结构，统一处理多智能体的环境认知
- 支持智能体根据周边风险与知识计算控制指令
- 模块化设计提升仿真开发效率，适合工程落地
构建可信赖的多智能体系统在实际应用中面临挑战，主要源于智能体间情境感知（SA）通信的复杂性。本文提出一种名为 SymAware 的高效易用软件框架，用于多智能体仿真。该框架提供丰富的预定义数据结构，用于计算、存储和传播智能体的情境感知信息。同时，它为智能体提供了抽象接口，使其能够基于对周围环境、知识及风险的感知来计算控制输入。此外，通过一组专用组件，SymAware 将仿真引擎中的物理渲染与通信接口等重负载计算隐藏于控制线程之后，显著提升了从概念原型到实际应用的开发效率。三个多智能体案例研究验证了该框架的有效性与高效性。
原文摘要 · Abstract (English)
Developing trustworthy multi-agent systems for practical applications is challenging due to the complicated communication of situational awareness (SA) among agents. This paper showcases a novel efficient and easy-to-use software framework for multi-agent simulation, named SymAware which provides a rich set of predefined data structures to compute, store, and communicate SA for agents. It also provides an abstract interface for the agents to compute their control inputs taking into account the awareness of the situation, knowledge, and risk of surrounding agents. Besides, utilizing a cluster of specialized components, SymAware hides the heavy computation of physical rendering and communication interfacing of simulation engines behind the control threads, resulting in high implementation efficiency in bridging the gap between conceptual prototyping and practical applications. Three multi-agent case studies are used to validate the efficacy and efficiency of this software framework.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。