面向月球探测多机器人通信系统的Ad hoc路由协议优化设计

Optimization Design of Ad hoc Routing Protocol for the Communication of Lunar Exploration Multi-robots System

Ad hoc网络作为一种不依赖于任何固定设施的临时性多跳自治系统,是解决月球探测过程中多机器人通信问题的可行选择。由于月球探测多机器人通信系统对节能和实时性等方面的特殊要求,现有的Ad hoc网络路由协议还难以满足工程设计和实际应用的需求。因而,结合月球探测环境下的实际工程需要,对现有的Ad hoc网络路由协议进行优化设计就成为关键。在分析月球探测多机器人通信系统拓扑结构的基础上,给出了系统中各节点的能量消耗模型;进而,建立了路径寻优的数学模型,在动态源路由协议DSR的基础上设计了一种基于能量约束的Ad hoc网络路由协议。该协议综合考虑了通信系统对于节能和实时性两方面的需求,将路由跳数、路由传送功率、节点剩余电池能量作为路由度量,采用改进的离散Bellman-Ford算法进行最优路径的求解。仿真实验结果表明,该协议在保证通信系统实时性的前提下,可延长网络生存时间,能为月球探测多机器人系统提供更可靠的通信保障。

Mobile ad hoe networks are a kind of self-organizing and self-reconfiguring networks. They can be established anytime and anywhere without the presence of stationary radio stations or fixed infrastructures. Thus they are regarded as an effective support tool for the communication of a lunar exploration multi-robots system. Generally speaking, there are two issues, the efficient energy conservation and the real-time communication, that are important for the lunar exploration multi-robots system. However, these two issues do not have a satisfactory answer with concurrently existing Ad hoc routing algorithms. Hence, an optimization design of the routing protocol is the key for the Ad hoc networks with the objective of the energy conservation and the constraint of the realtime communication. For that purpose, a node＇s energy consumption model is built based on the analysis of the network topology of the communication system. Then, a mathematical model is proposed to carry out the optimization for the optimal path. Moreover, a new power-aware routing scheme is presented based on the Dynamic Source Routing （DSR） according to the engineering requirements of the lunar exploration muhi-robots communication system. With constraints of hops, total transmission power, and node＇s remaining energy, the modified distributed Bellman-Ford algorithm is employed to find the optimal path. Simulation results show that the routing protocol designed in this paper performs better for the lunar exploration multi-robots system than others.