Algorithms for mobile agents, autonomous robots, and sensor networks

来自人工智能、机器人学和软件工程等多个领域的自主移动计算实体的研究领域专注于能够在网络(例如移动软件代理)或地形(例如自主机器人或移动传感器)上移动的计算实体。人们对这一领域的兴趣迅速增长。一方面，在从互联网到数据网格的网络环境中进行计算时，移动代理的使用正变得越来越流行，无论是作为一种理论范式，还是作为一种系统支持的编程平台。另一方面，考虑到与移动传感器网络的联系，自主机器人的理论和算法研究现在更加重要。从计算和算法的角度来看，对移动代理、自主机器人(特别是移动传感器网络)的研究还处于非常早期的阶段。理论框架含糊其辞，有时甚至缺失，算法技术和工具也很原始。为了开始涵盖这种缺乏理论和算法背景的问题，这个项目专注于移动实体系统中产生的计算和算法问题。主要目标是：更好地了解移动实体的计算能力(即，实体可以执行什么任务，在什么条件下，以什么成本)，并设计有效的算法来解决基本的协调、控制和安全问题。还将在环境中存在可能不安全的组件(故障、有害元素)的情况下研究计算和算法任务。特别是，在移动代理的情况下，项目会考虑一些主机是有害的(例如主机攻击)，以及当不想要的入侵者在网络中移动的情况(例如间谍软件或病毒)。*在第一种情况下，代理的目标是本地化有害主机，努力不被其破坏，在第二种情况下，它是捕获入侵者。就机器人而言，该项目的重点是可计算性问题，以及在对机器人能力的不同假设下对各种算法的最佳协调设计；特别强调和关注的是未来移动传感器网络技术背景下的应用。

Originating from fields as diverse as AI, robotics, and software engineering,  the research area on  autonomous mobile  entities  focuses on computational entities capable of moving over  a network (e.g., mobile software agents) or a terrain (e.g., autonomous robots or mobile sensors). This area has seen a rapid growth of interest. On one hand,  the use of mobile agents is becoming increasingly popular when computing in networked environments, ranging from Internet to the Data Grid, both as a theoretical paradigm and as a system-supported programming platform. On the other hand, the theoretical and algorithmic  study of autonomous robots is now even more important considering  the link with a network of mobile sensors. From   computational and   algorithmic viewpoints, research on mobile agents, on autonomous robots (and especially on mobile sensor networks) is still at a very early stage. Theoretical frameworks are vague and sometimes missing, algorithmic techniques and tools   are primitive.  To start covering this lack of theoretical and algorithmic background,  this project focuses on computational and algorithmic issues arising  in systems of mobile entities. The main goals are: to reach a better understanding of the computational power of mobile entities (i.e., what tasks can be performed by the entities, under what conditions, and at what cost), and to design efficient algorithms for solving basic coordination, control, and security problems. Both the computational and the algorithmic tasks will be studied also in presence of possibly unsafe components in the environments (faults, harmful elements). In particular,  in the case of mobile agents  the project  considers situations when some  of the hosts are harmful (e.g., host attack), and when an unwanted intruder  is moving   in the network(e.g., a spyware or a virus).  In the first case,  the goal of the agents is to localize the harmful host trying not to be  damaged by it, in the second case, it is to capture the intruder. In the case of robots, the project focuses on computability issues and on the   design of coordination   algorithms, under different assumptions on the capabilities of the robots; particular emphasis and focus is on applications in the  context of  mobile sensor networks.