人工智能集群控制演示验证系统
人工智能集群控制演示验证系统安梅岩1,2，王兆魁1，张育林1（1. 清华大学航天航空学院，北京100084； 2. 中国航天员科研训练中心，北京100193）摘要：为了对人工智能集群的自组织控制方法进行验证，提出了一种在实验室环境下低成本建立人工智能集群控制演示验证系统的方法．系统由演示场、多个移动个体、合作标识及识别单元、控制及信息分配单元共同组成．合作标识及识别单元能准确获取多个个体的身份信息和高精度的位姿信息．控制及信息分配单元对智能集群中个体间的信息交换和个体控制策略进行模拟．最后，以机动自组织探测集群为验证对象，在人工智能集群控制演示验证系统中对基于人工势场法的自组织控制策略进行演示验证．验证结果表明，该智能集群控制演示验证系统能够在实验室环境下对智能集群的运行过程进行演示和验证，可以更真实地体现智能集群的实际表现．
关键词：人工智能集群；演示验证系统；自组织；视觉标识识别
中图分类号：TP23 文献标识码：A 文章编号：1002-0446(2016)-03-0265-11
Demonstration and Verification System for Artificial Intelligent Swarm ControlAN Meiyan1,2 ，WANG Zhaokui1 ，ZHANG Yulin1(1. School of Aerospace Engineering, Tsinghua University, Beijing 100084, China;2. Astronaut Center of China, Beijing 100193, China)
Abstract: A low-cost demonstration and verification system is developed and established in laboratory condition to demonstrate and verify the self-organizing control strategy of an intelligent swarm. The system consists of an arena, multiple mobile
individuals, the cooperative identification logo and the identification unit, and the control and information allocating unit. The cooperative identification and the identification unit provides the accurate identification and the high-precision position and
direction data of the multiple mobile individuals. The control and information allocating unit simulates the rules of the information exchanging and control among the individuals of an intelligent swarm. Finally, the mobile self-organized detecting swarm is taken as a demonstration case, and the self-organizing control strategy based on artificial potential field is demonstrated and verified in the system. The demonstration results show that the proposed system can demonstrate and verify the operation process of an intelligent swarm in laboratory condition, and the demonstration can provide the actual performance of an intelligent swarm.
Keywords: artificial intelligent swarm; demonstration and verification system; self-organizing; visual fiducial identification
你也看这个杂志？棒棒哒:)
页:
[1]