多智能传感器的有限分散自治体系研究

DECENTRALIZED AUTONOMOUS MULTIPLE SMART SENSORS SYSTEM

智能系统多传感器信息融合的复杂性迫切需要开发一套合适的结构体系，目前大多数结构体系都通过融合中心对分散在不同点的多个传感器进行信息处理，而底层传感器之间缺乏必要的联系．这样导致融合中心计算和通信的负担过重而造成瓶颈，且不能使传感器之间互相启发以提高任务环境认知的效率．针对这些问题本文首先提出智能传感器的新概念，指出智能传感器须具备的５个基本能力即预测、规划、刷新、通信和同化，并在此基础上讨论了多智能传感器组成系统时的算法及信息流程．

To tackle the complexity of robotic multiple sensor fusion, a suitable system structure is required. Most structures at present have a fusion center to deal with information from various sensors located at different points , but unfortunately they have no necessary connections between low level sensors. Consequently fusion center has more responsibility for communication and signal interpretation, which resulted in bottle neck problem . Moreover, sensors cannot help each other to improve task environment recognition efficiency. To overcome the above problems, we define the concept of smart sensor at first . Each smart sensor should have five abilities : prediction, planning, updating, communication and assimilation. Then we discuss the algorithm of a perception system which consists of several smart sensors . Finally we take active tactile and active stereo camera as an example to illustrate the principle of our new idea.