What Is GNSS (Global Navigation Satellite System)?
A Global Navigation Satellite System (GNSS) is a satellite-based positioning system that allows accurate estimation of a receiver’s absolute position on Earth. In this article, we will explore GNSS: what it is and how it works.
The most well-known implementation is GPS (Global Positioning System, USA), but there are also Galileo (Europe), GLONASS (Russia), and BeiDou (China).
How GNSS Positioning Works
- Each satellite transmits signals containing:
- orbital parameters
- a high-precision timestamp
- navigation data
- The GNSS receiver measures the travel time of each signal and computes a pseudorange. The distance between the receiver and each satellite can be written as:
where:
- (x, y, z) are the receiver coordinates
- (xs, ys, zs) are the satellite coordinates at the time of signal transmission
- c is the speed of light
- δt is the receiver clock bias
Common Sources of GNSS Errors
GNSS positioning accuracy can be affected by:
- ionospheric and tropospheric delays,
- satellite orbital and clock errors,
- multipath interference (signal reflections),
- electromagnetic interference near the antenna.
GNSS Correction Technologies and Sensor Fusion
To overcome GNSS intrinsic limitations, modern robotics relies on a combination of GNSS correction technologies for precision navigation and sensor fusion algorithms for autonomous systems, transforming noisy measurements into a robust and reliable state estimation.
Multi-Constellation GNSS Receivers
Modern receivers use multiple constellations (GPS, Galileo, GLONASS, BeiDou) to increase the number of visible satellites. This improves availability, accuracy, and positioning reliability.
SBAS: Satellite-Based Augmentation Systems
SBAS corrections (e.g., EGNOS in Europe, WAAS in the U.S.) improve accuracy by broadcasting real-time atmospheric and orbital error corrections via satellites.
RTK GNSS for Centimeter-Level Accuracy
RTK GNSS for precision agriculture and autonomous drones: the gold standard for centimeter-level accuracy in outdoor robotics.
GNSS Sensor Fusion for Reliable Navigation
Sensor Fusion with IMU, LiDAR, and cameras: widely used in GNSS/INS integration for self-driving cars and autonomous UAV navigation. Algorithms include the Extended Kalman Filter (EKF), Unscented Kalman Filter (UKF), and Particle Filters for localization in robotics.
GNSS Applications in Robotics and Autonomous Vehicles
GNSS receivers act as absolute localization sensors for mobile robots and are employed in virtually all outdoor robotic platforms operating in open-sky environments.
Applications include:
Precision Agriculture with GNSS RTK
In precision farming, RTK GNSS ensures centimeter accuracy for:
- automatic tractor guidance,
- optimized seeding, fertilization, and harvesting,
- efficient drone mapping and targeted spraying.
Logistics and Autonomous Transport with GNSS
GNSS is essential for:
- autonomous container trucks,
- yard automation,
- AMRs (Autonomous Mobile Robots) transitioning between indoor and outdoor environments,
- last-mile delivery robots navigating sidewalks and urban canyons.
GNSS for Aerial Drones (UAVs)
Unmanned aerial vehicles rely on GNSS for:
- waypoint navigation,
- photogrammetry and infrastructure inspection,
- environmental monitoring,
- search and rescue missions,
- safety features like Return-to-Home.
Marine and Underwater Robotics with GNSS
- Unmanned Surface Vehicles (USVs) use GNSS for bathymetric mapping, water quality monitoring, and surveillance.
- Autonomous Underwater Vehicles (AUVs) cannot receive GNSS signals underwater, so they rely on inertial systems. Periodically, they surface to reset their position using GNSS fixes.
Practical GNSS Case Studies: Common Issues and How We Solved Them
Aitronik has deployed GNSS solutions for outdoor mobile robots in multiple industrial scenarios, facing and solving real-world challenges.
Electromagnetic Interference in GNSS Antennas
One of the first challenges Aitronik faced with mobile robots was GNSS signal interference.
Despite excellent satellite visibility, position estimates were often inaccurate or completely missing. After thorough investigation, the root cause was identified as electromagnetic interference from a USB3 connector located just a few centimeters from the GNSS antenna.
How We Solved GNSS Interference Problems
The solution was simple but effective:
- Relocating the GNSS antenna approximately 1.5 meters away from the USB3 connector.
- Adding conductive shielding (ground plane) under the antenna. A circular shield with a radius of 6–7 cm greatly reduces unwanted noise and improves signal quality.
GNSS Communication Protocol Compatibility Issues
Another recurring problem in real-world robotics projects is incompatibility between GNSS communication protocols.
Although the NMEA 0183 protocol is an industry standard, many manufacturers add proprietary sentences or even replace standard ones. This often leads to parser incompatibility, forcing developers to write custom software for each GNSS receiver.
Best Practices to Avoid Protocol Problems
- Always verify that the full protocol documentation is available from the manufacturer.
- Check for proprietary messages before integration, even if the receiver claims to be “NMEA compliant.”
GNSS Filters and Position Estimation Errors
A less obvious but equally critical source of errors lies in the filters embedded in GNSS receivers. Most receivers use an Extended Kalman Filter (EKF) to estimate position, but implementations vary significantly between models.
Why GNSS Filter Configuration Matters
Different motion models are available depending on the application:
- Static use (fixed receiver),
- Pedestrian navigation,
- High-speed vehicles.
If the wrong model is selected, the filter may interpret normal vehicle speeds as invalid, resulting in completely unrealistic position estimates.
