【作者】 张伟；

【导师】 齐华；

【作者基本信息】 西安工业大学 ， 通信与信息系统， 2019， 硕士

【摘要】 随着经济的快速发展和人民生活条件的改善,电器化产品不断增加,由于用电不规范或者线路老化等问题,经常导致火灾爆炸等灾难发生,而大型建筑楼宇内部结构复杂,因此对消防救援工作提出了更高的要求。消防员作为火灾等应急救援的中坚力量,肩负着快速及时救援的重任,一旦消防员发生意外,指挥端不能得到消防员具体位置和实时情况,错过最佳营救时机,这将直接威胁消防员生命安全甚至影响整个救援工作的进度。针对以上问题本文研究了消防现场信息记录系统关键技术和设计了系统方案,该系统的研究与设计对保障消防员安全和辅助救援决策具有重要的意义。本文以消防救援工作为应用背景,研究和设计了消防救援过程中能够实时获取消防员位置、运动姿态及生命体征等信息的消防现场信息记录系统。首先根据系统实际需求对系统总体方案进行设计,然后重点对系统中消防员室内定位和姿态识别两种关键技术分析与研究,最后对系统进行测试。在综合分析了现有的室内定位技术后,选择室外辅助的RSSI定位和惯性定位两种定位技术,其中对RSSI加权质心算法进行权值修正并对信标节点的部署问题进行分析,实验结果表明,相比于未修正权值和节点优化前定位性能提高了 19.5%,惯性定位采用加速度双重积分估计位移和四元数姿态解算得到航向角实现室内定位,通过对加速度进行零偏修正和滤波处理,减少了加速度计系统误差和噪声对位移估计的影响。由于RSSI模型定位易受环境的影响,且惯性定位会随着时间的增加会出现累计误差,所以选择EKF模型对两种定位技术进行融合处理,发挥两种定位技术各自的优点,实现优势互补。其次,对人体姿态识别算法进行分析,选择BP神经网络算法对人体运动姿态进行分类识别,通过对姿态数据进行预处理、特征提取和实验测试,由测试结果可知,本文中算法能够满足姿态识别的需求。然后,对硬件系统中各功能模块选择合适的型号和相关电路设计,搭建系统软件开发平台和上位机用户界面的设计。最后,对系统的功能进行了测试并且对融合定位算法和姿态识别算法进行了实验验证。实验结果表明,本文的研究和设计能够满足系统的基本需求,具有一定的实用价值和研究意义。更多还原

【Abstract】 With the rapid development of the economy and the improvement of people’s living conditions,electrical products have been increasing.Due to problems such as irregular electricity consumption or aging of the line,disasters such as fire and explosion often occur,Since the internal structure of large buildings is complex,it puts forward higher requirements to the fire rescue Work.As the backbone of emergency rescue such as fire,firefighters shoulder the responsibility of quick and timely rescue.Once a firefighter has an accident,the commander cannot get the specific location and real-time situation of the firefighter,and miss the best rescue opportunity,which will directly threaten the life of the firefighter and even affect the progress of the entire rescue work.In view of the above problems,this paper studies the key technology of the fire scene information recording system and designs the system scheme.The research and design of the system is of great significance to ensure the safety of firefighters and assist rescue decisions.Based on the background of fire rescue work,this paper studies and designs a fire scene information recording system that can obtain real-time information such as firefighter position,movement posture and vital signs in the fire rescue process.Firstly,according to the actual needs of the system,the overall scheme of the system is designed.Then,the key technologies of the firefighters indoor positioning and attitude recognition are analyzed and studied,and finally the system is tested.After comprehensive analysis of the existing indoor positioning technology,Two positioning technologies including the outdoor assisted RSSI positioning and inertial positioning are selected.Among them,the RSSI weighted centroid algorithm is modified and the beacon node deployment problem is analyzed.The experimental results show that the positioning performance is improved by 19.5%,comparing with the uncorrected weight and node optimization.The inertial positioning uses the acceleration double integral estimation displacement and the quaternion attitude solution to obtain the heading angle to achieve indoor positioning,and the offset is corrected and filtered.Processing reduces the effects of accelerometer system errors and noise on displacement estimation.Since the positioning of the RSSI model is susceptible to environment and the inertial positioning will accumulate error with time,the EKF model is selected to fuse the two positioning technologies,and the advantages of the two positioning technologies are utilized to achieve complementary advantages.Secondly,the human body gesture recognition algorithm is analyzed,and the BP neural network algorithm is selected to classify and recognize the human body motion posture.By preprocessing,feature extraction and experimental testing of the attitude data,the test results show that the algorithm can satisfy the gesture recognition demand.Then,select the appropriate model and related circuit design for each functional module in the hardware system,and build the system software development platform and the design of the host computer user interface.Finally,the function of the system is tested and the fusion location algorithm and gesture recognition algorithm are experimentally verified.The experimental results show that the research and design of this paper can meet the basic needs of the system,and it has certain practical value and research significance.更多还原