NNF Webinars October - November 2022
30. 09The NNF is pleased to announce two upcoming webinars in the months of October and November. Attendance is open to all interested parties at the zoom links included below.
On Thursday October 27th at 15:00 CET doctoral researcher Akpo Siemuri from the University of Vaasa in Finland will give his presentation titled "Utilization of Machine learning in improving accuracy of low-cost localization devices - experiences from the Google Smartphone Decimeter Challenge."
Smartphone Decimeter Challenge
Google, the Institute of Naviation's Satellite Division, and Kaggle sponsored a Smartphone Decimeter Challenge at the just concluded ION GNSS+ 2022 held in Denver, Colorado, USA. Teams developed high precision GNSS positioning using a pool of GNSS and IMU datasets collected from smartphones, and accompanied by high accuracy ground truth. They compete to achieve the best location accuracy with the datasets provided. More information: https://www.kaggle.com/
Abstract:
The global navigation satellite system (GNSS) data from smartphones have lower signal levels and higher noise in GNSS observations compared to commercial GNSS receivers. Therefore, it is difficult to directly apply the existing GNSS high-precision positioning methods like precise point positioning (PPP) and real-time kinematic (RTK). Multi-sensor fusion technology has become very prominent for seamless navigation systems due to its complementary capabilities to GNSS positioning. This talk presents a way to improve the positioning estimate of smartphones by utilizing a machine learning (ML) based adaptive positioning approach (MAP) with post-processed kinematic (PPK) techniques to process the GNSS datasets. The ML model is used to predict the driving paths (highways, tree-lined streets, or downtown areas). Depending on the predicted driving path, the PPK technique uses the carrier phase to compute the user position using differential corrections from known GNSS base stations. A Rauch–Tung–Striebel (RTS) smoother, which consists of a forward pass Kalman Filter (KF) and a backward recursion smoother is used to achieve a loosely coupled GNSS/IMU integration to estimate the position of the smartphone. We refer to this method as LC-GNSS/IMU/ML. Using the proposed method, we estimated the location of the smartphone and tackled the competition. The results were validated using the provided high accuracy ground truth giving an accuracy of 2.29 m compared to the weighted least square (WLS) score of 9 m.
A recording of the presentation is here.
On Thursday November 24th at 15:00 CET Gustaf Hendeby from Linköping University will present "Results on GNSS Spoofing Migitation Using Multiple Receivers"
Abstract:
GNSS receivers are vulnerable to spoofing attacks in which false satellite signals deceive receivers to compute false position and/or time estimates. This work derives and evaluates algorithms that perform spoofing mitigation by utilizing double differences of pseudorange or carrier phase measurements from multiple receivers. The algorithms identify pseudorange and carrier-phase measurements originating from spoofing signals, and omit these from the position and time computation.
The algorithms are evaluated with simulated and live-sky meaconing attacks. The simulated spoofing attacks show that mitigation using pseudoranges is possible in these tests when the receivers are separated by five meters or more. At 20 meters, the pseudorange algorithm correctly authenticates six out of seven pseudoranges within 30 seconds in the same simulator tests. Using carrier phase allows mitigation with shorter distances between receivers, but requires better time synchronization between the receivers. Evaluations with live-sky meaconing attacks show the validity of the proposed mitigation algorithms
A recording of the presentation is here.
