【作者】 李卓；

【导师】 刘经南； 牛小骥； 张提升；

【作者基本信息】 武汉大学 ， 电路与系统， 2019， 硕士

【摘要】 自动驾驶、无人机等技术领域的兴起,标志着高精度定位需求从专业领域进入大众市场。全球卫星导航系统(GNSS)是室外定位的重要手段之一,利用GNSS载波相位差分定位技术能实现厘米级的定位精度。由于载波相位跟踪的脆弱性,目前连续可靠的厘米级定位精度只有在观测条件良好的开阔场景才能保证。因此,研究复杂信号环境下载波相位跟踪与观测值质量具有重要意义。传统GNSS接收机在复杂信号场景面临动态、弱信号等诸多挑战与参数设计的局限性,而GNSS/INS深组合接收机具有优良的动态性能,能有效提高载波环的跟踪性能。本文基于对传统标量深组合与矢量深组合结构的分析,实现一种基于标量深组合结构,借鉴矢量接收机通道间信息共享思想的深组合软件接收机。基于深组合软件接收机平台,针对复杂信号条件下载波基带跟踪层面和观测值提取层面面临的各种挑战,提出相应的优化方法,进而改善载波相位质量与接收机定位精度。在载波相位基带跟踪层面,首先,针对信号衰弱导致载波相位跟踪不连续问题,提出一种深组合辅助下的COOP跟踪环结构。利用强信号通道估计出接收机钟漂对各跟踪环路的影响,从而使载波环克服动态应力与跟踪灵敏度之间的矛盾,提高跟踪灵敏度。其次,针对信号频繁断续条件下载波相位跟踪不连续问题,提出一种载波相位预测方法。通过相位预测方法维持被遮挡卫星在短暂遮挡时间内的载波相位连续与信号恢复后的快速锁定,主要优化了对被遮挡卫星通道的接收机钟漂影响的多普勒估计。在载波相位观测值层面,针对载波环鉴相器对180°翻转不敏感导致载波相位测量值存在半周模糊度的问题,提出比特预测方法快速确定半周模糊度值。然后,提出基于惯导和里程计辅助的载波相位观测值周跳检测和修复方法,基于惯导短期精度高的特点,利用RTK/INS/里程计的高精度定位结果对载波相位周跳进行检测。经过与真值文件的检测结果对比,该方法能一定程度上改善载波相位观测值连续性。最后,设计实验对本文的深组合软件接收机进行了测试与评估。首先,基于仿真测试场景对载波相位跟踪性能进行定量测试。结果表明:在部分信号较弱的动态仿真测试中,本文的载波环跟踪灵敏度较传统锁相环提高了12 dB,较经典深组合锁相环提高了4~7 dB。在信号断续测试中,在静态和动态场景下,模拟部分卫星遮挡时,90%以上的测试结果都能在15 s的预测时间内维持载波相位发散不超过1/4周;在动态场景下测试了相位预测方法对RTK定位的影响,结果表明,使用相位预测方法能有效提高部分卫星信号断续条件下的RTK固定率和定位精度。然后,在实测车载环境下对深组合软件接收机的载波相位观测值质量和RTK定位精度进行了测试与评估。测试结果表明,深组合软件接收机的载波相位观测值连续性和RTK固定情况,以及整体定位效果均优于ublox接收机,证明了本文提出的载波相位优化方法的有效性。更多还原

【Abstract】 With the rise of technical fields such as autonomous driving and unmanned aerial Vehicle(UAV),the demand for high-precision positioning has entered the mass market from the professional field.Global Satellite Navigation System(GNSS)is one of the important means of outdoor positioning.The centimeter-level positioning accuracy can be achieved by using GNSS carrier phase differential positioning technology.Due to the fragility of carrier phase tracking,the continuous and reliable centimeter-level positioning accuracy can only be guaranteed in open-sky scenes with good observation conditions.Therefore,it is important to study carrier phase tracking and observation quality in complex signal environment.Traditional GNSS receiver faces many challenges and parameter design limitations of dynamic and weak signals in complex signal scenarios,while the GNSS/INS deep integration receiver has excellent dynamic performance and can effectively improve the tracking performance of the carrier loop.Based on the analysis of traditional scalar deep integration and vector deep integration structure,this paper implements a deep integrated software receiver based on scalar deep integration structure and the idea of information sharing between vector receiver channels.Based on the deep integration software receiver platform,aiming at the challenges faced by carrier baseband tracking level and observation extraction level under complex signal conditions,corresponding optimization methods are proposed to improve carrier phase quality and receiver positioning accuracy.At the carrier phase baseband tracking level,firstly,for the problem of carrier phase tracking discontinuity caused by signal weakening,a COOP tracking loop structure with deep integration assistance is proposed.The strong signal channels are used to estimate the influence of the receiver clock drift on each tracking loop,so that the carrier loop overcomes the contradiction between dynamic stress and tracking sensitivity,and improves tracking sensitivity.Secondly,aiming at the discontinuity of carrier phase tracking under the condition of frequent discontinuity of signals,a carrier phase prediction method is proposed.The phase prediction method is used to maintain the carrier phase continuity of the occluded satellite during the short occlusion time and the fast lock after the signal recovery,and the Doppler estimation of the receiver clock drift of the occluded satellite channel is optimized.At the carrier phase observation level,for the problem of half-cycle ambiguity of carrier phase measurement due to the insensitivity of carrier loop phase detector to 180 degree flip,a bit prediction method is proposed to determine half-cycle ambiguity quickly.Then,a cycle slip detection and repair method of carrier phase observation based on INS and odometer is proposed.Based on the short-term high accuracy of INS,carrier phase cycle slip is detected by using the high precision positioning results of RTK/INS/odometer.Compared with the result of the true value file,this method can improve the continuity of carrier phase observation to a certain extent.Finally,the deep integration software receiver is tested and evaluated by design experiment.Firstly,the carrier phase tracking performance is quantitatively tested based on simulation test scenarios.The results show that the tracking sensitivity of the carrier loop in this paper is 12 dB higher than that of the traditional PLL and 4 ~ 7 dB higher than that of the classical deep combination PLL in the dynamic weak signal test.In signal intermittent test,more than 90% of the test results can maintain carrier phase divergence for less than 1/4 cycles in 15 s prediction time when simulating partial satellite occlusion in static and dynamic scenarios.The effect of phase prediction method on RTK positioning in dynamic scenes is tested.The results show that the phase prediction method can effectively improve the fixed rate and positioning accuracy of RTK under intermittent satellite signals.Then,the carrier phase observation quality and RTK positioning accuracy of the deep integrated software receiver are tested and evaluated in the real vehicle environment.The test results show that the carrier phase observation continuity,RTK fixing and overall positioning effect of the deep integration software receiver are better than those of ublox receiver,which shows the effectiveness of the carrier phase optimization method proposed in this paper.更多还原