High-Precision GPS for Autonomous Vehicles

Driving Towards Autonomy

The idea of autonomous vehicles sharing the road is slowly becoming a reality due to advances in positioning and sensor integration. High Precision Global Navigation Satellite System (GNSS) technology provides the accuracy, availability and reliability that a vehicle requires to be self-driving.

A fully autonomous vehicle needs an accurate localization solution paired with the confidence that the localization solution is correct. Our GNSS technology is capable of providing decimetre-level accuracy to ensure a vehicle stays in its lane, or a safe distance from other vehicles.

Position Change Tolerance

GNSS receivers use multiple frequencies, multiple GNSS constellations, Synchronous Position Attitude and Navigation (SPAN) technology and GPS Anti-Jam technology to provide the positioning and sensor integration that autonomous vehicles need.

Many technologies onboard vehicles provide local or relative localization. GNSS provides an absolute localization solution, and with the following technologies can achieve the accuracy and availability requirements of a full autonomous driving solution. 


High Precision Image

Multi-Frequency, Multi-Constellation

In order to receive the best possible accuracy it is recommended that two or three frequencies broadcasted by each GNSS constellation (GPS, GLONASS, Beidou and Galileo) be used.

  • Using multi-frequency receivers mitigates errors caused by variable signal delays caused by atmospheric conditions
  • L1/L2 frequency combination is most common; L5 is used for modernized GPS, Beidou and Galileo
  • The more constellations used, the higher the likelihood of observing satellites; particularly important in urban or obstructed environments
  • GNSS receivers and antennas capable of receiving multiple frequencies and constellations must be utilized

Multi Frequency Image

GNSS Corrections

GNSS signals, without corrections, provide positioning accuracy of five to ten metres (16-32 feet). Corrections can be generated by a number of sources or methods, and system developers must choose the corrections method that best meets the demands of their application. NovAtel CORRECT, the algorithm embedded on all NovAtel GNSS receivers, optimizes all correction methods.

Corrections work in combinations with multi-frequency measurements from the GNSS to provide sub-decimetre to centimetre-level accuracy – depending on the correction source.

RTK (Real Time Kinematic)

RTK sends data from reference receivers in the vicinity of the vehicle. Location can be covered by one or more RTK networks, available free of charge or for a subscription fee.

RTK diagram

PPP (Precise Point Positioning) 

PPP uses globally available corrections using a world-wide network of reference receivers. Corrections are transmitted to the vehicle via satellite or by cellular. NovAtel utilizes TerraStar corrections to deliver a sub-deceimetre solution. 

PPP diagram 3

Sensor Integration

Components like radar, LiDAR and cameras are used to provide the distance to objects that surround the vehicle. If the exact location of the surrounding objects is known, this technology can provide the absolute vehicle location with assistance from a high amount of map data. When integrated with complimentary technology such as: ultrasonic, interial motion, digital maps, radar/LiDAR and cameras, GNSS acts as the sixth sense to deliver the positioning performance required by autonomous vehicles. 


Driver Only: The functions of acceleration, handling and control of the vehicle are done by the driver.

Safety Assistance: Either the acceleration, handling or control is executed by the automobile.

Semi-Autonomous Driving: The automobile is able to execute two to three of the following maneuvers; acceleration, handling and control.  

Semi-Autonomous Systems: All acceleration, handling and control would be executed by the automobile, but the driver would be able to take over control in an emergency situation.

Autonomous Driving: All acceleration, handling and control would be executed by the automobile; there would be no involvement from a driver. 

Driving Roadmap

The image below shows the light urban environment the Teseo V SBAS and Teseo V NovAtel PPP tests took place. 

Test Trajectory Map


Horizontal Position Errors Chart

Horizontal Position Errors Chart 

Horizontal Cumulative Error Distribution

Horizontal Error CDF Chart


  CEP [M] 1 σ [M] 2 σ [M]
Teseo V Alone 1.03 1.21 1.81
Teseo V with NovAtel 0.11 0.21 0.26