Dr. Weisong Wen

Dr. Weisong Wen

Assistant Professor at Hong Kong Polytechnic University. GNSS, trustworthy navigation and control of autonomous systems.

Zhong Yihan (PhD Student): GNSS Collaborative Positioning

3 minute read

Published:

Targeted potential project (2023)

Director Chan, road doctor, discuss on potential support for the ITF project.

Abstract

This project aims to improve the urban GNSS positioning using the collaborion of the raw GNSS measurements from multiple agents. Specifically, the relative positioning constraints is free of the atmospheric and clock errors. In addition, the relative constraints can lead to the measurements redundancy which even enables the estimation of the atmospheric errors.

Dr. Wen, W. and Zhong Yihan.

Scholarship Plan

  • (1st year) 09/2022-08/2023: FTE (Dr LT Hsu’s quota)
  • (2nd year) 09/2023-08/2024: Pending
  • (3rd year) 09/2024-08/2025: Pending

Recent Research Plan: Zhong Yihan (PolyU TPS)

  • Half year research plan (by Dec 2022)
    • Solid-state LiDAR matching with map (IET ITS, first revision)
    • GNSS/PDR FGO (IEEE T-VT, under preparation)
    • GNSS FGO GNC (ION GNSS+ 2022; GPS Solutions)
    • First demo of the GNSS collaborative positioning
  • One year research plan (by Jun 2023)
    • Double-differenced Doppler modeling for relative positioning (Huawei Phase 3)
    • Tripple differenced carrier-phase modeling for relative positioning (Huawei Phase 3)
  • 1.5 years research plan (by Dec 2023)
    • Outlier-aware GNSS CP algorithms in urban canyons

Student Supervision

  • Supervise Edward? (undergraduate student)

News

  • 1.	ION GNSS+ abstract the experiment part revision: add noise to investigate if the inter-frame DD factor could make our system more robust.
2.	Estimation of ionosphere and troposphere delay 
3.	Map matching work
4.	Extension of Shaomin’s work on FGO PPP-RTK 
5.	Study if sharing measurements could make PPP convergence faster.
6.	Aiding vector map for FGO-based GNSS-RTK/INS system (try to find the scientific problem).
7.	Recording the data using UAVs.
    • Performance analysis of the collaborative positioning inter-frame constraints under different window sizes
    • agent 2 and 3 add the interframe constraints
    • plot the satellite skyplot
    • add the SNR filering
    • 2 and 3 relative constraints does not have been added
    • Performance analysis of the collaborative positioning with factor graph model (mainly evaluate the relative positioning)
    • Cross-epoch relative positioning from same/different trajectories
    • Update the GraphGNSSLib information
    • Collaborative map matching to improve the positioning performance of the pedestrian
    • Using the simulation (An Open-Source Gazebo Plugin for GNSS Multipath Signal Emulation in Virtual Urban Canyons) to simulate the collaborative positioning.
    • Using the high accuracy pedestrian positioning to update the Openstreet map 人行道
    • Smart Summon using the GNSS collaborative positioning
    • Performance analysis of the collaborative GNSS positioning in urban canyons
  • [12/01/2022]
    • Can we use the UAV in the sky as the ancle point to provide the relative positioning concerning the user in the dense urban canyons?
    • Yihan is working on a problem of how to accurate estimate the relative location between the user and the vehicle. We discussed the data collection about the relative positioning between the pedestrian and the vehicle.
      • Can we also combine the map to constrain the position of the user, so-called the map matching.
    • Yihan will try to found the full name of the application.
    • Can we submit this work to the ION GNSS+ in terms of the performance analysis of the GNSS collaborative positioning with FGO?
      • How will GNSS collaborative posiitoning improve the absolute positioning?
      • How will GNSS collaborative posiitoning improve the relative positioning?
      • How will GNSS collaborative posiitoning performs under different receivers/scenarios?
    • Comments from Yihan about the student supervision?
      • Xiaxiao is willing to share and very outgoing.
  • Discussion on 25th, Nov 2022
    • Slight improved positioning result is obtained with help of the double-differenced relative constraints
    • Next step is to use the static and open area dataset to validate the effectiveness of the proposed method.
    • Add the time-correlated and spatial-correlated relative constraints based on the DD difference
    • Based on all these constraints, can we also estimate the astmosphere error?

You May Also Enjoy