首页|麦克风集群网络模型及其障碍物识别中的应用

麦克风集群网络模型及其障碍物识别中的应用

扫码查看
基于声学基本原理和抽象空间的网络结构方法,开发了一种创新的障碍物检测系统,其结构包括硬件终端、多声路网络模型和分类器.硬件终端由立体式分布的麦克风节点、数据采集卡和工控机组成.多声路网络模型能够基于多路声波信号建立网络结构,并提取和计算多路声波复杂信号中的有效信息.分类器采用主成分分析方法(Principal Component Analysis,PC A)对提取的有效信息进行降维处理,再采用监督学习方法实现对环境的分类和识别.该系统的开发不仅为汽车自动驾驶、机器人导航和无人机等领域提供了重要的技术支持,而且为智能导航和自动化领域的技术发展提供了新的可能性.未来的工作将进一步优化系统性能,提高其应用的广泛性和经济效益.
Microphone Array Network Model and Its Application in Obstacle Recognition
An innovative obstacle detection system is developed based on basic principles of acoustics and a network structure method in abstract space.This system is comprised of hardware terminal,multi-sonic network model,and classifiers.The hardware terminal includes stereoscopic distributed microphone nodes,data acquisi-tion card,and industrial personal computer.The network structure is established based on signals from multiple sound waves,and valuable information is extracted and computed from the complex signals of multiple sound waves.The principal component analysis(PCA)is used to reduce the dimensionality of the extracted informa-tion,then supervised learning methods is applied to achieve the ability of classifing and recognizing environ-ments.The creation of this system not only provides significant technical support for autonomous driving cars,robot navigation and drones,but also provides new possibilities for the technological advancement in intelligent navigation and automation fields.Future works will focus on further optimizing the system's performance and improve its application scope and economic benefits.

microphone arrayspatial acoustic wave patternmulti-sonic network modelhigh-dimensional fea-ture extractionobstacle recognition application

张子明、许劭晟、李凯、王纬国

展开 >

国营芜湖机械厂,安徽芜湖 241007

南京航空航天大学 自动化学院,江苏南京 211106

安徽师范大学计算机与信息学院,安徽芜湖 241000

麦克风阵列 空间声波纹路 多声路网络模型 高维特征提取 障碍物识别应用

2021年安徽省重点研究与开发计划第三批立项项目

2021008

2024

测控技术
中国航空工业集团公司北京长城航空测控技术研究所

测控技术

CSTPCD
影响因子:0.5
ISSN:1000-8829
年,卷(期):2024.43(10)