Resilient and Accurate Autonomous Vehicle Navigation via Signals of Opportunity Zak M. Kassas Autonomous Systems Perception, Intelligence, and Navigation (ASPIN) Laboratory University of California, Riverside 1 Stanford University Stanford, CA November 8, 2017
3 Future Autonomous Vehicles
4 Future Autonomous Vehicles
Navigation with Signals of Opportunity (SOPs): COpNav 5 Kassas (2013). Collaborative opportunistic navigation. IEEE Aerospace and Electronic Systems Magazine, (28)6, 38 41.
SOPs: Opportunities and Challenges 6 Abundant Opportunities Available at varying geometries & frequencies Free to use Are significantly more powerful than GPS Challenges States may be unknown a priori Observables need to be extracted Clocks are not as stable and not synchronized Signal models and error budgets are unavailable
SOP Signal Landscape 8 The SOP signal landscape state space is NOT stationary
9 Software- Defined Radio for SOP- Based Navigation (MATRIX)
10 Cellular CDMA Navigation SDR (LabVIEW)
Experimental Setup Ground Vehicle 11 MATRIX Khalife, Shamaei, & Kassas (2016). A software- defined receiver architecture for cellular CDMA- based navigation. IEEE/ION Position, Location, & Navigation Symposium (PLANS), 816 826, (Best student paper).
12 SOP Mapping
13 SOP Mapping
14 SOP = Fake Tree!
Experimental Demo: UAV Navigation with Cellular CDMA 15 https://www.youtube.com/watch?v=gkfuxie2wna
Cellular LTE Navigation SDR (LabVIEW) 16 Shamaei, Khalife, & Kassas (2016). Performance characterization of positioning in LTE systems. in Proceedings of ION GNSS Conference (IONGNSS+), 2262-2270, (Bestpaper presentation).
Cellular LTE Navigation SDR (LabVIEW) 17 Shamaei, Khalife, & Kassas (2016). Exploiting LTE signals for navigation: Theory to implementation, IEEE Trans. on Wireless Communications, submitted.
LTE Receiver Structure 18 ESPRIT: estimation of signal parameters by rotational invariance techniques Shamaei, Khalife, Bhattacharya, & Kassas (2017). Computationally efficient receiver design for mitigating multipath for positioning with LTE signals. in Proceedings of ION GNSS Conference (ION GNSS+), 3751-3760, (Best paper presentation).
Experimental Demo: Ground Vehicle Navigation with LTE 19 https://www.youtube.com/watch?v=fidgngrjuzq
20 SOP- Aided Inertial Navigation
SOP- Aided INS Framework 21 EKF Prediction Tightly-coupled data SOP Receiver IMU Inertial Navigation System EKF Update flag Detector PPS GNSS Receiver Traditional tightly- coupled GNSS- aided INS Objectives: 1. Estimate SOPs states when GNSS pseudoranges are available (mapping) 2. When GNSS psuedoranges become unavailable, continue to estimate SOPs states and use to correct INS errors (SLAM)
Simulator Overview 22 Mission Planning & Scripting Accelerate/ Decelerate, Climb/ Descend, Turn, Loops, Rolls, Aerobatics Kinematic Models 6DOF Aircraft 3DOF & 6DOF Automobile Sensor Models GPS L1/L2 (constellation & sub- frames) IMU Magnetometer Air Data: Pitot & Static (and several others )
SOP- Aided INS: Simulated Environment 23 SOP SOP SOP SOP
24 SOP- Aided INS: Simulated Environment
25 SOP- Aided INS: Simulated Environment
EKF Results: Vehicle Position and Velocity 26 GPS Cut-off Traditional GPS-aided INS Error SOP-aided INS* Error *With consumer grade IMU 0 50 100 150 200 0 50 100 150 200 Morales, Roysdon, & Kassas (2016). Signals of opportunity aided inertial navigation. ION Global Navigation Satellite System (ION GNSS+) Conference, 1492-1501, (Bestpaper presentation).
Navigating with Cellular CDMA & LTE 27 Kassas, Morales, Shamaei, & Khalife (2017). LTE Steers UAV. GPS World Magazine, (28)4, 18-25 (Cover Article).
Experimental Results 28 Kassas, Morales, Shamaei, & Khalife (2017). LTE Steers UAV. GPS World Magazine, (28)4, 18-25 (Cover Article).
29 Collaborative Navigation
Centralized Collaborative Framework 30 Central fusion center Modes of operation: 1. Collaborative Mapping 2. Collaborative SLAM (C- SLAM) o Single point of failure o Large communication bandwidth Morales & Kassas (2016). Collaborative autonomous vehicles with signals of opportunity aided inertial navigation systems. ION International Technical Meeting (ION ITM), 805-818.
Distributed Collaborative Framework 31 AV 3 AV 2 AV 4 This is only 32 elements! AV 1 AV N Replace: ) 289 elements Morales & Kassas (2017). A Low communication rate distributed inertial navigation architecture with cellular signal aiding. IEEE Vehicular Technology Conference (VTC), submitted.
Monte Carlo Analysis 32 Performance robustness Approximation robustness Morales & Kassas (2017). Distributed signals of opportunity aided inertial navigation with intermittent communication. ION Global Navigation Satellite System (ION GNSS+) Conference, 2519-2530. (Best paper presentation).
Experimental Demo: Distributed UAV Navigation with SOP- Aided Inertial 33 https://www.youtube.com/watch?v=gljrk2ogspm
34 Precise Carrier Phase Navigation
Experimental Demo: UAV Navigation with Carrier Phase Cellular Signals 35 https://www.youtube.com/watch?v=wsqduolktwo
36 Acknowledgment