Aerospace and Electronic Systems - May 2019 - 82

Feature Article:

DOI. No. 10.1109/MAES.2019.2906971

Navigation Systems Panel Report
Navigation Systems for Autonomous and
Semi-Autonomous Vehicles: Current Trends
and Future Challenges
Zaher M. Kassas, University of California, Irvine, CA, USA
Pau Closas, Northeastern University, Boston, MA, USA
Jason Gross, West Virginia University, Morgantown, WV, USA

NAVIGATION SYSTEMS FOR AUTONOMOUS VEHICLES
The world is abuzz with semi-autonomous and fully
autonomous vehicles. From unmanned aerial vehicles
(UAVs) to self-driving cars, integrating these vehicles
into our daily lives will have astounding societal and economic impacts. As we endow these vehicles with higher
levels of autonomy, the requirements on their navigation
system become more stringent than ever before. Undoubtedly, navigation system failure for these vehicles could
have intolerable consequences.
Navigation systems for future semi-autonomous and
fully autonomous vehicles must possess the following
attributes:
(1) Assured performance: specify the uncertainty
associated with the navigation solution and alert the
human to take over, when needed
(2) Tamper-proof: detect and recover from malicious
attacks (e.g., jamming and spoofing)
(3) Redundancy: robustness to sensor failure and/or
signal degradation

Authors' current address: Z. M. Kassas, University of
California, Irvine, Irvine, CA 92697, USA, E-mail:
(zkassas@ieee.org). P. Closas, Northeastern University,
Boston, MA 02115, USA. E-mail: (closas@northeastern.
edu). J. Gross, West Virginia University, Morgantown,
VA, USA. E-mail: (jason.gross@mail.wvu.edu ).
Manuscript received January 26, 2019, revised January 31,
2019, and ready for publication January 31, 2019.
Review handled by P. Willett.
0885-8985/19/$26.00 ß 2019 IEEE
82

(4) High levels of accuracy: meet the accuracy requirement as dictated by the application; for example,
achieve lane-level accuracy for self-driving cars
Navigation systems can be broadly categorized into:
(1) Sensor-based: provide a local navigation solution by
utilizing dedicated on-board hardware that senses
the surrounding environment (e.g., inertial navigation systems, cameras, lidar, etc.)
(2) Signal-based: provide a global navigation solution
and rely on receiving external signals from either
(i) dedicated transmitters (e.g., global navigation
satellite systems (GNSS), eLoran, pseudolites, etc.)
or (ii) nondedicated transmitters, also known as signals of opportunity (SOPs) (e.g., AM/FM, cellular,
television, WiFi, communication satellites, etc.)
Current autonomous vehicle navigation system design
trends fuse GNSS receivers with a suite of sensor-based
technologies. By adding more and more sensors, designers
are throwing "everything but the kitchen sink" to prepare the
autonomous vehicle navigation system for the inevitable
scenario when GNSS signals become unavailable or unreliable. High-grade sensors may violate cost, size, weight, and
power (C-SWaP) constraints. Also, these sensors may not
properly function in all environments (e.g., fog, snow, rain,
dust, etc.) and are still susceptible to malicious attacks.
In what follows, future navigation system challenges
brought forth by autonomous vehicles are discussed.

SENSORS
Recent decades have enjoyed rapid maturation of navigation alternatives to GNSS. Many of these advances

IEEE A&E SYSTEMS MAGAZINE

MAY 2019

Aerospace and Electronic Systems - May 2019

Table of Contents for the Digital Edition of Aerospace and Electronic Systems - May 2019