Abstract

In urban environments, Global Navigation Satellite Systems (GNSS) signals are frequently attenuated, blocked or reflected, which degrades the positioning accuracy of GNSS receivers significantly. To improve the performance of GNSS receiver for vehicle urban navigation, a GNSS/INS deeply-coupled software defined receiver (GIDCSR) with a low cost micro-electro-mechanical system (MEMS) inertial measurement unit (IMU) ICM-20602 is presented, in which both GPS and BDS constellations are supported. Two key technologies, that is, adaptive open-close tracking loops and INS aided pseudo-range weight control algorithm, are applied in the GIDCSR to enhance the signal tracking continuity and positioning accuracy in urban areas. To assess the performance of the proposed deep couple solution, vehicle field tests were carried out in GNSS-challenged urban environments. With the adaptive open-close tracking loops, the deep couple output carrier phase in the open sky, and improved pseudo-range accuracy before and after GNSS signal blocked. Applying the INS aided pseudo-range weight control, the pseudo-range gross errors of the deep couple decreased caused by multipath. A popular GNSS/INS tightly-coupled vehicle navigation kit from u-blox company, M8U, was tested side by side as benchmark. The test results indicate that in the GNSS-challenged urban areas, the pseudo-range quality of GIDCSR is at least 25% better than that of M8U, and GIDCSR’s horizontal positioning results are at least 69% more accurate than M8U’s.

Highlights

  • Autonomous vehicles in urban environments require accurate and timely access to information such as car position, speed and so on [1,2]

  • As an important sensor of the vehicle navigation solution, the Inertial Navigation System (INS) information can be assisted into Global Navigation Satellite Systems (GNSS) receiver baseband processing to implement a deep couple design, which could eliminate the impact of dynamics caused by vehicle and permit the receiver baseband to work in a quasi-static scene [18,19,20,21,22,23]

  • GNSS/INS deeply-coupled software defined receiver (GIDCSR) is developed on basis of non-coherent deep couple structure

Read more

Summary

Introduction

Autonomous vehicles in urban environments require accurate and timely access to information such as car position, speed and so on [1,2]. As an important sensor of the vehicle navigation solution, the Inertial Navigation System (INS) information can be assisted into GNSS receiver baseband processing to implement a deep couple design, which could eliminate the impact of dynamics caused by vehicle and permit the receiver baseband to work in a quasi-static scene [18,19,20,21,22,23] In this way the bandwidth can be compressed and the integration time can be lengthened, which will reduce the noise and improve the sensitivity and accuracy of the receiver under dynamics [24,25,26]. In 2017, iMAR’s high precision integration navigation product iTraceRT-F402 applied scalar GNSS/INS deeply-coupled technology to improve its performance of carrier phase reacquisition and tracking [39]. Implemented a vector tracking GNSS/INS deeply coupled receiver to improve the performance in GNSS-challenged environments They developed a maximum-likelihood bit decoding algorithm for bit wipe-off to extend the integration time coherently and achieved good results. The performances of pseudo-range, GNSS position and integrated position of GIDCSR and M8U are compared

Overview
Adaptive Open-Close Loop Strategy
INS Aided Pseudo-Range Weight Control Algorithm
Experiment Setup
Experiment Results and Analysis
Deep Couple Performance Analysis
Adaptive Open-Close Tracking Loops Analysis
INS Aided Pseudo-Range Weight Control Algorithm Analysis
Deep Couple Positioning Analysis
Experiment Results Comparison Between GIDCSR and M8U
Pseudo-Range Error Analysis
GNSS Positioning Analysis
Conclusions and Future Work
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call