【作者】 吴立珍；

【导师】 曾迎生；

【作者基本信息】 国防科学技术大学 ， 控制科学与工程， 2006， 硕士

【摘要】 本文以陆地自主车(Autonomous Land Vehicle,简称ALV)为应用背景,从超声测距系统、高精度超声测距处理算法以及基于超声传感器的障碍定位和识别技术等三个方面展开了对超声测障技术的深入研究和实验验证。本文首先设计和实现了基于单片机的多通道超声信号采集处理平台。通过上层控制指令,该平台能够完成不同发射方案、不同采样频率以及不同处理算法的测距处理,具有功能强大、便携性好和抗干扰能力强的优点。在超声测距处理算法的研究方面,本文从理论上分析了调制解调技术的引入对测距系统处理增益的影响,根据数字相关法与匹配滤波器的等价性,使用包络相关解调算法实现了伪随机码数字相关测距,并给出了改进的两步相关快速测距算法的计算公式。此外,为进一步提高系统的处理增益和抗干扰能力,本文还引入了双极性M序列以实现PSK相干解调。针对互相关函数峰值检测中的误差来源,本文提出了一种新的基于阈值的两次检测算法,在很大程度上解决了未知环境中阈值设定的难题,提高了超声测距的精度。结合传感器阵列的设计,本文分析了不同情况下超声信息的融合方法,给出了基于多距离信息的障碍定位算法。通过对影响超声回波波形的主要因素进行深入分析,本文建立了超声回波信号模型,并提出了基于回波振幅和时长特征的障碍类型判别算法。更多还原

【Abstract】 In this thesis, the obstacle detection technology based on ultrasonic sensors has been studied for Autonomous Land Vehicle (ALV) applications. The thesis consists of three closely related subjects: ultrasonic range-finding system, ultrasonic ranging processing method and localization and classification of obstacles.At first, an ultrasonic range-finding system based on a single-chip microcomputer is designed and realized. According to different instructions from PC, the system can accomplish ranging work under different transmitting schemes, different sampling rates as well as different processing methods. The system shows high precision, portability and high resistance to interference through practices.With deeply research of both the effect of the introduction of modulation and demodulation technology on processing gain of range-finding system and the equivalence of correlation detection and matching detection, this paper has realized the digital correlation detection arithmetic based on PN code via the envelope demodulation method. Also the calculate formulas for a fast matching detection approach based on two step correlation is concluded. Besides, for the further improvement of the processing gain, M sequence and polar envelopes are utilized to realize PSK demodulation.In order to get the precise position of the echo, a novel peak detection method is proposed. The error due to the fluctuation of the peak pulse of the cross-correlation function could be eliminated effectively in this way, so the assignment problem of threshold in unknown environment can be resolved well.Incorporating the design of sensor array, data fusion of multiple ultrasonic sensors in different situations is discussed, and also the obstacle localization method is presented. Finally, the echo model is built by analyzing the affecting factors of the waveform of echo, and as a result, a novel obstacle recognition method based on the echo amplitude and duration features is proposed.更多还原