R H I N O S Railway High Integrity Navigation Overlay System

Transcription

1 R H I N O S Railway High Integrity Navigation Overlay System RHINOS Integrity Monitoring and Augmentation Subsystem Reference Architectures Cosimo Stallo RadioLabs Rome SOGEI premises June 21 th 2017

2 Table of Content 1. Requirements Review for Rail Community 2. Candidate AIMN Architectures Definition following current and future Augmentation System Trends 3. PRO-CON Analysis and Ranking Results 4. SWOT Analysis applied to two ranked Architectures 5. Selected Augmentation Architecture Definition 6. Architectural Blocks Definition 7. Fault Tree Analysis 8. Experimental results

3 Main Goals to select the Integrity Monitoring and Augmentation subsystem architecture after collection and comparison of a set of candidates; to define the infrastructure elements that can be shared with the corresponding avionics solutions; to review the analytical models to evaluate system performances; to perform Common Mode Failure (CMF)/ Common Cause Failure (CCF) analysis and safety design of selected architectures.

4 Participants RadioLabs (WP7 Leader): Collection of the candidate solutions for the AIMN subsystem and selection of the AIMN reference architecture SOGEI: PRO-CON Analysis on AIMN Architecture Candidate Solutions Ansaldo STS: Rail Sector requirements review and Definition of Architectural components and Interfaces with reference to the existing TCS for High Integrity LA University of Pardubice: Performance comparison of the candidate solutions for the AIMN subsystem DLR: Multiconstellation and/or multi-frequency receivers. ARAIM. University of Nottingham: Environmental Faults Modelling

5 Requirements Review Third RHINOS Workshop Rome SOGEI 06/21/2017

6 System Architecture

7 Candidate Augmentation Architectures Application Domain Accuracy Integrity SBAS Aviation < 2 m 10-7 /150s GBAS/DGNSS Aviation/Maritime <16 m /150s RTK/NRTK Surveying < 5 cm (AF) * PPP Surveying Maritime Autonomous Driving *Under study (CRAIM and AR Integrity) 10 cm (Float) < 10 cm (AR) No single system can currently meet Rail Requirements: Longitudinal: Max balise detection error: 1 m THR=10-9 /h Lateral: accuracy: 3m THR=10-9 /h *

8 List of Identified Candidate Solutions GBAS: it represents GBAS solution; HA/DGNSS: it represents a cross transport modes solution following the HA/DGPS initiative; NRTK+CRAIM: it represents existing Network RTK solutions extended through CRAIM (Carrier Phase RAIM) capabilities for covering Integrity requirements 2-Tiers solution: it is the 2-Tiers solution integrating Local Augmentation and SBAS NRTK+ARAIM: it is the NRTK solution augmented with ARAIM capabilities at Network and user level for SIS IM THA/RAIM: it represents a high Accuracy Network for Rail improved with OBU RAIM for achieving high integrity performances; relevant costs can be shared with other transport modes requiring the same performance levels PPP/ARAIM/CRAIM: it represents the integration of FAST PPP (including Accurate Real-Time Ionospheric Monitoring) and ARAIM+CRAIM for improving Integrity performances

9 Ranking Criterias and Weigths Suitability for Rail Deployment Costs THR or Confidence Interval for Rail Maintenance cost Accuracy for normal operations Ionospheric Monitoring Accuracy for parallel track discrimination OBU Complexity Time To Fix Ambiguities (Ambiguity Resolution) Communication needs Availability Coverage Continuity Standardisation TTA or equivalent Maturity Weigths Description Weight Very Adequate 4 Adequate 3 Discrete 2 Sufficient 1 Not appropriate 0

10 PRO-CON Analysis - Ranking Results PRO CON Difference Ranking GBAS HA-DGNSS NRTK+IM Tiers (SBAS+LADGNSS) NRTK+ARAIM THA/RAIM PPP/ARAIM/CRAIM SBAS Main Message: Integration is the Winning Card!

11 SWOT Analysis Results First Two Ranked Architectures from PRO-CON Analysis

12 Reference GNSS Local Augmentation Architecture

13 Interfaces vs External Sources To TALS SBAS Ground Services RIMS Network Raw data/sbas Msgs Terrestrial Broadcasting External Network GPS and Galileo Ground Services GRSP IGS IGS-RTS RTCM /1005 RTCM /1005 MSM High QoS/security Communication Network RBC Costituent GNSS Local Augmentation Rail RS Network Fundamental role of High QoS (low Latency) High Security Communication Network: Very Hazardous Hackering consequences EoP SP3 at global level! RTCM SSR Messages (Sat biases Precise Eph/Clock STEC)

14 Integrity Monitoring: GBAS and 2-Tiers Purpose: it determines Faulty SISs Constellations and RSs and passes clean raw data and Integrity Masks to other blocks Processing: GBAS algorithms 2-Tiers algorithms Interfaces: RTCM/MSM New RTCM proposals (RTCM prop.) Data Acquisition Rail RS Network Spoofing Detection RTCM 3 100x MSM Spoofing Mask (RTCM-like) Integrity Monitoring RS FDE SIS FDE Integrity Parameters Augmentation Processing and Msgs Generation Healthy Sat and Const Mask (RTCM prop.)

15 Integrity Parameters Purpose: it calculates parameters needed for Integrity Monitoring Augmentation messages generation Processing: RS and SIS statistics Estimators par. Interfaces: RTCM RTCA-246D RTCM prop. Data Acquisition RTCM 3 100x MSM Integrity Monitoring RS FDE SIS FDE Integrity Parameters P Sat F P i i Augmentation Processing and Msgs Generation RTCA-246D Type Tiers par. RTCM prop. Const F P LA F

16 Local Atmospheric Monitoring Purpose: it performs a Local Atmospheric conditions monitoring and detectes anomalous conditions Processing: Gradient Estimation Vert. Iono Grad. Std est. Tropo refractivity est. Interfaces: GBAS Type 2 RTCM prop. Integrity Monitoring RS FDE SIS FDE Local Atmospheric Monitoring Augmentation Processing and Msgs Generation GBAS Type 2 RTCM prop. Measurement Corrections

17 Network Processing for High Accuracy Purpose: it performs needed computations for allowing RTK NRTK and be enabled for RT PPP AR Processing: Network AR starting from widelaning meas. (RTK NRTK) POD using IGS-RTS ultrarapid RT SSR RTCM mesgs (PPP) Satellites DCBs estimation using GNSS Ground Services (PPP) Accurate STEC (PPP) Measurement Corrections Network AR Network Processing for High Accuracy Precise Orbit/Clock* Satellite Biases* Accurate STEC* RTCM SSR RTCM prop

18 Measurements Corrections Purpose: generation of measurement corrections (PRC RRC RTK/NRTK raws or Area corrections) to be sent to the TALS for transferring to the OBU Processing: PRC and RRC calculation RTK/NRTK/VRS raws PPP errors estimation Interfaces: RTCM 1000x RTCM SSR RTCM prop. RTCA-246D RTCM 100x RTCM SSR RTCM prop RTCA-246D TALS Augmentation Processing and Msgs Generation Measurement Corrections

19 Spoofing Detection Purpose: it detects spoofed satellites and constellations and sends Spoofing Masks to other blocks Processing: Network Monitoring: P/Y-C/A correlations PRS C/N0 CRPA Antennas Others Interfaces: RTCM 100x MSM RTCM Prop. (Spoofing Mask) RTCM 100x MSM Spoofing Detection Data Acquisition RTCM prop (Spoofing Mask) Integrity Monitoring RS FDE SIS FDE

20 Local Augmentation Fault Cases Hints Wrong Receiver channels behaviour (2%) Reference Receiver MotherBoard Failures (4%) Unexpected Multipath (2%) Antenna displacement-environmental causes (2%) Failures in the communication link between the Control Centre and the RR (56%) Failures in the Reference Station Power System (20%) Human Intervention (2%) Failures in the Control Centre Software Processing (5%) Severe Lighting (1%) Others (6%) Approximated values from empirical data

21 Current NRTK Specificities Local Augmentation coverage: National Level NRTK/MAC: Master Station for Corrections Calculation Multi-Reference System: Recovery of Single Receiver faults Based on continuous clusters of 4-5 Reference Stations OSR (Observation State Representation) or SSR (State Space Representation) corrections Baselines <= 70 Km Cluster 1 Cluster 2 Master1 Master2

22 GNSS Augmentation Fault Analysis 5 RSs/Cluster COTS RS MTBF (nominal): h COTS RS MTBF (not nominal: comm. link signal deformations..): about 2 years Pr Fault RS (shared among 5 RSs): /h P MD =10-4 /h THR= /h Generalised B-Values for H1 (NRTK) Ephemeris Fault MDE to be revised (longer baselines) H2 FDE perfomance can be improved through 2-Tiers

23 Generalised B-Values for LA and NRTK Measurement Corrections Software (MC SW) calculates corrections through all possible Subnetworks of M-1 RSs in a cluster excluding at each step one the Master Station of the whole set ) ( 4 1 ) ( ) ( ) ( 4 1 ) ( ) ( z y x PHCN z y x PHCN z y x BN z y x PRCN z y x PRCN z y x BN m j j n j n n m m j j n j n n Pm Master ) ( z y x PRCN n PRC calculated by the MC SW ) ( z y x PHCN n Phase Range Raws/Corrections calculated by the MC SW Network Ambiguities Fixed by the NRTK MC SW xyz= Prediction Point close to the user (e.g. VRS) or Cluster centroid NRTK

24 THR for VB detection

25 Experimental Results from ERSAT-EAV Test Environment: Scenario 1 - not ERTMS Sardinia: Track Cagliari San Gavino October TAAN Augmentation Network messages are processed by TALS implemented on a Laptop (COTS equipment). The OBU equipment is on the diesel automotrice ALN Both constellations GPS and GALILEO are used. Case 1 - Three TAAN RSs TAAN-CC integrity monitoring TALS and OBU processing and positioning for rides in nominal condition without any faults using GPS and Galileo satellites. Case 2 - Three TAAN RSs TAAN-CC integrity monitoring TALS and OBU processing and positioning for ride where satellite faults are injected in Real-Time by TAAN. The faults are simulated on GPS PRN 01 PRN 03 PRN 06 PRN 07 PRN 09 and PRN 17 along the track San Gavino Cagliari on 25th of October Case 3 - Three TAAN RSs TAAN-CC integrity monitoring TALS and OBU processing and positioning for ride using only GPS constellation.

26 Protection Level (m) Number of Points per Pixel Experimental Results from ERSAT-EAV Case 1: Nominal Conditions GPS+ Galileo 2016/10/24 San Gavino - Cagliari 50 Stanford Diagram (1260 epochs) System Unavailable Alarm Epochs: 0 Normal Operation MI epochs: MI epochs: 0 HMI epochs: Error (m)

27 Experimental Results from ERSAT-EAV Case 2: Faulty satellites GPS+ Galileo 2016/10/25 San Gavino - Cagliari

28 Experimental Results from ERSAT-EAV Case 3: GPS only 2016/10/26 San Gavino - Cagliari

29 Conclusions and Future Steps Conclusions: Integration of High Accuracy and High Integrity Architectures for meeting requirements 2-Tiers algorithms for SIS and RSs for Augmentation FDE to be adopted by Internal and External Networks Openess to innovative HP algorithms Standardisation: existing and new messages Importance of High QoS/Secure Communication Links Ongoing and future steps: Feedbacks from POC and Performance Analysis RHINOS Architecture Refinement

30 Thank you for your attention