基于车载惯性传感器的驾乘人员计数和体重识别方法研究

发布时间：2018-02-11 23:02

本文关键词： 驾乘人员计数体重识别惯性传感器车辆动力学　出处：《重庆大学》2016年硕士论文　论文类型：学位论文

【摘要】：为了促进出租车管理机构对出租车数量的合理安排和部署,完善道路交通死亡统计数据的分类,紧随车辆轻量化、无人化和智能化的新趋势以及增强汽车个性化设计和防盗设计等,本文提出了一种效率较高而成本极低的,针对驾乘人员进行计数和体重识别的解决方案。首先,对所研究内容的数学模型进行了研究。为了合理准确地表征研究对象,本文借鉴了以往的车辆动力学模型。并在此基础上,结合车辆垂向动力学模型和所研究内容的车辆运动情况,本文采用了七自由度、四自由度和三自由度的车辆垂向动力学模型。其次,通过对研究内容和研究对象的初步分析,筛选出了合理的车辆垂向动力学模型,并基于MATLAB/Simulink模块对单个人上下车的场景进行了仿真分析。为了对所筛选动力学模型进行验证,本文依据实验设计的基本规则,对单个人上下车的情况进行了整车实验。并通过比较和分析仿真实验和整车实验的数据,验证了所建立数学模型的正确性。最后,借助整车实验数据,分析并筛选了对模型输入激励最为敏感的响应信号。凭借以上所完成的研究内容,再次分析并结合整车实验数据和仿真数据,本文提出了基于阈值法的驾乘人员计数方法的解决方案。此外,通过MATLAB/Sim ulink仿真开发工具,建立了计数系统的仿真模型并利用实验数据进行了测试。最后,为了实现对驾乘人员体重的识别,本文推导并分析了系统的传递函数。在此基础上,依据线性时不变系统的齐次性,提出了基于车载惯性传感器的驾乘人员体重识别方法的解决方案。
[Abstract]:In order to promote the rational arrangement and deployment of taxi numbers by taxi management agencies, and to improve the classification of road traffic death statistics, closely following the light weight of vehicles, The new trend of inhumanization and intelligence, as well as the enhancement of individualized design and anti-theft design of the automobile, this paper presents a solution for counting and weight identification of drivers with high efficiency and extremely low cost. The mathematical model of the research is studied. In order to represent the object of study reasonably and accurately, this paper draws lessons from the previous vehicle dynamics model, and on this basis, Combined with the vehicle vertical dynamics model and the vehicle motion of the research content, this paper adopts the vehicle vertical dynamics model with seven degrees of freedom, four degrees of freedom and three degrees of freedom. Secondly, through the preliminary analysis of the research content and research object, A reasonable vertical vehicle dynamic model is selected, and the scene of single person getting on and off is simulated and analyzed based on MATLAB/Simulink module. In order to verify the selected dynamic model, this paper bases on the basic rules of experimental design. Through comparing and analyzing the data of the simulation experiment and the whole vehicle experiment, the correctness of the mathematical model is verified. Finally, with the aid of the whole vehicle experiment data, The response signal which is most sensitive to the input excitation of the model is analyzed and screened. Based on the above research content, the experimental data and simulation data of the whole vehicle are analyzed and combined again. In this paper, the solution of driver counting method based on threshold method is put forward. In addition, the simulation model of counting system is established by MATLAB/Sim / ulink simulation development tool and tested by experimental data. In order to recognize the weight of the driver, the transfer function of the system is deduced and analyzed. On this basis, according to the homogeneity of the linear time-invariant system, This paper presents a method to identify the weight of the driver based on the vehicle inertial sensor.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U463.6

【相似文献】