基于多维加速度传感器的车辆姿态测量系统研究与设计

发布时间：2018-04-03 13:06

本文选题：道路检测　切入点：振动　出处：《哈尔滨工业大学》2015年硕士论文

【摘要】：随着路网的逐渐完善,我国公路建设的工作重心已经逐渐由新公路的建设转移到已建成公路的养护上来。国内外现有的道路检测车多通过线性激光扫描仪实现路面的精确检测,然而,由于路面往往并不平整,而线性激光扫描仪是固定安装在检测车上的,在行驶过程中,由于路面不平度引起的颠簸振动、车辆本身加速、减速以及发动机振动等造成的车身姿态变化会引起线性激光扫描仪的姿态变化,造成扫描精度的降低甚至重大误差的产生。在现有的道路检测车中,多通过搭载组合式惯性姿态测量系统来实现车辆姿态的检测,进而弥补由于车身姿态变化造成的检测误差。然而一方面组合式惯性姿态测量系统往往造价高昂,而更重要的是,现有的道路检测车上的组合式姿态测量系统多着眼于整车姿态的测量,实际中引起测量误差的仅为搭载激光扫描仪部分车身的姿态变化,整车姿态的测量不但造成了系统的冗余,提高了造价,也导致针对部分搭载激光扫描仪部分车身的姿态测量精度不高。针对上述问题,本文对车辆行驶过程中振动加速度特性进行分析,设计了一种基于多维加速度传感器的车辆姿态测量系统,针对性的测量搭载激光扫描仪部分车身的姿态变化。根据实际需要,通过针对性的建立路面及车辆振动模型,对车辆行驶过程中的搭载激光扫描仪部分车身振动加速度特性进行分析,在考虑到道路检测持续时间长特点的基础上,设计了针对车身振动加速度的基于1/2车模型的时域二次积分算法,并通过振动台静态实验以及实际道路测试证明了该算法的有效性及优越性,并最终完成了加速度信号采集电路的PCB设计。本文主要包括以下内容:首先,从高速公路不平度等级出发,建立了路面不平度时域模型,并针对道路检测车的实际需要,分析搭载激光扫描仪部分车身的特点,针对性的建立了四自由度1/2汽车横向振动数学模型,通过MATLAB仿真对车辆行驶过程中的车身振动加速度特性进行了分析,并以此作为二次积分算法设计的重要依据。其次,分析现有的二次积分算法,针对道路检测持续时间长的特点,基于由四自由度1/2汽车横向振动数学模型得到的车身振动加速度特点,设计了能够进行长时间高精度积分的二次积分算法,分析车辆行驶过程中搭载激光扫描仪部分车身姿态变化,根据加速度传感器积分结果给出该部分车身姿态表示方法。然后,通过标准振动台静态测试以及道路实测,对采集的不同频率的加速度数据进行积分,并通过分析积分误差证明了本文所设计二次积分算法的有效性及优越性。最后,根据道路检测实际需要,基于ADS1278芯片,设计了并行8通道加速度信号采集电路,以便同时采集多个方向上的多个加速度传感器数据,并完成了PCB设计。
[Abstract]:With the gradual improvement of road network, the focus of highway construction in our country has gradually shifted from the construction of new highway to the maintenance of completed highway.Most of the existing road detection vehicles at home and abroad use linear laser scanners to accurately detect the road surface. However, because the road surface is often uneven, the linear laser scanner is fixed in the detection vehicle, and in the process of driving,Due to the bumpy vibration caused by the roughness of the road, the vehicle's acceleration, deceleration and engine vibration will cause the attitude change of the linear laser scanner, resulting in the reduction of scanning accuracy and even the generation of major errors.In the existing road detection vehicle, the combined inertial attitude measurement system is used to detect the vehicle attitude, which can make up for the detection error caused by the change of the vehicle body attitude.However, on the one hand, the combined inertial attitude measurement system is often expensive, and more importantly, the existing road detection vehicle combined attitude measurement system mostly focuses on the whole vehicle attitude measurement.In practice, the measurement error is caused only by the attitude change of part of the body carrying the laser scanner. The attitude measurement of the whole vehicle not only results in the redundancy of the system, but also increases the cost.It also leads to the low precision of attitude measurement for part of the body of laser scanner.In order to solve the above problems, this paper analyzes the vibration acceleration characteristics of vehicle driving, designs a vehicle attitude measurement system based on multi-dimension acceleration sensor, and measures the attitude change of part of vehicle body with laser scanner.According to the actual needs, the vibration acceleration characteristics of part of the vehicle body carried on the laser scanner during the vehicle driving are analyzed by establishing the road surface and vehicle vibration model, considering the long duration of the road detection.A time domain quadratic integration algorithm based on 1 / 2 vehicle model for vehicle body vibration acceleration is designed, and the validity and superiority of the algorithm are proved by the static test of the shaking table and the actual road test.Finally, the PCB design of acceleration signal acquisition circuit is completed.The main contents of this paper are as follows: firstly, a time-domain model of road roughness is established from the gradation of highway roughness, and according to the actual needs of road detection vehicle, the characteristics of part of the body carrying laser scanner are analyzed.In this paper, a four-degree-of-freedom (FDOF) 1 / 2 vehicle lateral vibration mathematical model is established, and the acceleration characteristics of vehicle body vibration are analyzed by MATLAB simulation, which is regarded as an important basis for the design of quadratic integration algorithm.Secondly, the existing quadratic integration algorithm is analyzed. Aiming at the long duration of road detection, the acceleration characteristics of vehicle body vibration are obtained from the four-degree-of-freedom 1 / 2 vehicle lateral vibration mathematical model.A quadratic integration algorithm for long time and high precision integration is designed, and the attitude change of part of the body carried by laser scanner is analyzed. According to the integral result of acceleration sensor, the attitude representation method of this part of body is given.Then, through the static test of the standard vibration table and the road measurement, the acceleration data collected at different frequencies are integrated, and the effectiveness and superiority of the quadratic integration algorithm designed in this paper are proved by analyzing the integration error.Finally, according to the actual needs of road detection, based on the ADS1278 chip, a parallel 8-channel acceleration signal acquisition circuit is designed to collect multiple acceleration sensors in multiple directions at the same time, and the PCB design is completed.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U418.3

【参考文献】