Scroll Top

Challenge

In November 2021, we began collaborating with a technology solutions integrator that was seeking a method for precisely identifying and locating containers at a container port. The port environment poses exceptionally challenging conditions for GPS systems: containers are stacked in towers reaching up to 5 units high and arranged in long rows, creating dense radio corridors. This results in limited sky visibility and a large number of signals reflected off metal walls.

Container ports are places where positioning accuracy directly impacts operational efficiency and safety. A positioning error of just a few meters can mean handling the wrong container. We decided to use RTK (real-time kinematic positioning) GPS technology, which achieves centimeter-level accuracy under favorable conditions.

Technological Context

GNSS (Global Navigation Satellite System) technology includes systems such as GPS, Glonass, Beidou, and QZSS. They determine the receiver’s position based on the distance between the satellites and the receiver, calculated using the time it takes for the signal to travel from the transmitter to the receiver, multiplied by the speed of the signal. In practice, however, the calculated distance is rarely the actual distance. As the satellite signal passes through successive layers of the atmosphere, it undergoes refraction, which introduces measurement errors, including orbital, clock, and tropospheric errors.

Under ideal conditions, a GPS antenna has an unobstructed view of the sky within a range of approximately 150°, which allows it to receive signals from a sufficient number of satellites – a minimum of 4 are required for accurate position calculation. The more satellites, the more accurate and stable the reading.

In fact, any terrain obstacle between the satellite and the receiver limits this line of sight and degrades the quality of the signal. The more urbanized or densely built-up the environment, the worse the conditions for satellite systems.

Signal reflections are another source of errors. This occurs when the signal reaches the receiver not only via a direct path, but also after being reflected off terrain obstacles such as buildings, trees, or, as in our case, the metal walls of shipping containers. Each additional signal path introduces interference and causes position drift.

GPS RTK (real-time kinematic positioning) is a technology that addresses these limitations by achieving centimeter-level accuracy in real time. It works by using a base station with a known reference position, which measures the difference between its actual and calculated positions and, based on this, generates corrections for each satellite signal. These corrections are then transmitted to the end receiver via radio (e.g., LoRa, Wi-Fi) or cellular networks (NTRIP protocol) and allow the calculated position to be corrected on the fly. Key limitation: the base station and the end receiver cannot be more than 10 km apart. This technology is only useful on a local scale.

Tests – Phase One: Single-Frequency System at the Port

For the initial tests, we selected the Ublox Neo M8P single-frequency receiver. The base station communicated with the mobile module via LoRa wireless technology.

The port environment quickly revealed its limitations. We encountered three main problems:

  • the container stacks blocked a significant portion of the sky, reducing the number of available satellites below the threshold required to maintain RTK mode,
  • the metal walls of the containers strongly reflected the satellite signal, generating interference and position drift,
  • the receiver regularly switched to standard GPS mode without RTK corrections, which negated the main advantage of the chosen technology.

The single-frequency system did not meet our expectations in this environment.

Tests - Phase Two: Verification in an Urban Environment

Before returning to port, we decided to conduct comparative tests in an urban environment-a less extreme setting, but one that allowed for controlled measurements. We tested at what distance from office buildings the Neo M8P system would lose RTK assistance.

The results were disappointing: even at a distance of 8–10 meters from a building wall; the signal faded, and RTK corrections were unavailable. The conclusion was clear: a single-frequency system is not suitable for navigation in environments with terrain obstacles.

We decided to test the dual-frequency Ublox c099-f9p-1 kit with the ZED-F9P receiver, which operates simultaneously on the L1 and L5 frequencies. The receiver processes signals from two independent paths, which allows for the correction of, among other things, tropospheric errors-one of the main sources of inaccuracy in standard GNSS systems.

The results were qualitatively different. Under the same test conditions, when the receiver was moved to within 1 meter of a building wall, the RTK system operated stably, without signal loss or position drift. The readings accurately reflected the actual change in location in real time.

Curious if a GPS system with RTK corrections could solve your problem?

Conclusions and Next Steps

The tests clearly demonstrated the advantage of dual-frequency solutions in environments with limited sky visibility and high levels of signal reflection. Despite its lower cost, the single-frequency system is unable to provide the required stability in dense urban or port environments.

The results of the tests in an urban environment provide a basis for further work. The next step is to test the dual-frequency system directly in a container port—the target deployment environment—where signal propagation conditions are even more challenging.

Want to build a similar solution?

Do you have an unusual idea or a technical challenge that cannot be solved with an off-the-shelf product? Get in touch with us. We like projects that do not fit into standard frameworks.

Want to build a similar solution?

Do you have an unusual idea or a technical challenge that cannot be solved with an off-the-shelf product? Get in touch with us. We like projects that do not fit into standard frameworks.

detectingwatercontamination

Aqua Insights by Aqua Alarm AS – Smart Water Contamination Detection System

Read more
termowizja-krowy-pastwisko

Smart Livestock Hall Cooling with Edge AI and Thermovision

Read more
Interaktywna papuga ai rapidlab case study

Interactive AI Parrot

Read more