Scientists have discovered a option to make GPS remarkably accurate in city “urban canyons,” where navigation has long gone wrong.
Most of us trust that the location shown by our GPS is accurate.
But anyone who has tried navigating an unfamiliar city knows that is not always the case. You may be walking steadily down the same sidewalk, yet your map app makes it look as though you are zigzagging across the street or jumping from one location to another.
“Cities are brutal for satellite navigation,” explained Ardeshir Mohamadi.
Mohamadi is a doctoral fellow at the Norwegian University of Science and Technology (NTNU). His research focuses on improving the accuracy of low-cost GPS receivers (like the one in your mobile phone or your fitness watch) without relying on expensive correction services.
Greater GPS accuracy is becoming increasingly important, especially for vehicles that are designed to drive themselves, also known as autonomous or self-driving vehicles.
Why Cities Confuse GPS Signals
To tackle this challenge, Mohamadi and his colleagues at NTNU have created a new positioning system aimed at helping autonomous vehicles navigate safely through urban environments.
“In cities, glass and concrete make satellite signals bounce back and forth. Tall buildings block the view, and what works perfectly on an open motorway is not so good when you enter a built-up area,” said Mohamadi.
When GPS signals bounce off buildings, they take longer to reach the receiver. Because GPS calculates location based on how long signals take to travel from satellites, those delays can result in incorrect distance measurements and inaccurate positioning.
Researchers often refer to these dense city environments as ‘urban canyons’. The term describes streets lined with tall buildings that resemble a deep canyon. Signals reaching a person or a self-driving vehicle may have reflected off multiple surfaces before arriving at the receiver.
“For autonomous vehicles, this makes the difference between confident, safe behaviour and hesitant, unreliable driving. That is why we developed SmartNav, a type of positioning technology designed for ‘urban canyons’,” explained Mohamadi.
SmartNav Targets Centimeter-Level Accuracy
Signal reflections are only part of the problem. Even when GPS signals arrive correctly, they often lack the precision needed for advanced navigation systems.
To address this, the NTNU team combined several technologies that help refine and correct positioning data. The result is a software system that can be integrated into autonomous vehicle navigation platforms.
Part of the solution also relies on a recently introduced Google service. Before looking at that, it helps to understand how GPS works.
GPS – the Global Positioning System – consists of a network of satellites orbiting Earth. These satellites continuously transmit radio signals that can be detected by GPS receivers. By receiving signals from at least four satellites, a receiver can calculate its location.
Each signal contains information about the satellite’s position and the exact time the message was sent. In simple terms, it is similar to receiving a text message from a satellite containing its location and timestamp.
Using Radio Waves Instead of GPS Codes
The information encoded in those signals can become unreliable when reflections occur in urban environments. One approach explored by the NTNU researchers was to ignore the code itself and instead analyze the radio wave carrying it.
A key measurement is whether the wave is moving upward or downward when it arrives at the receiver. This characteristic is known as the carrier phase.
“Using only the carrier phase can provide very high accuracy, but it takes time, which is not very practical when the receiver is moving,” said Mohamadi.
The challenge is that the receiver must remain stationary while enough measurements are collected. Achieving the desired accuracy can take several minutes rather than a fraction of a second.
Another option is to improve GPS data using correction services. One widely used method relies on base stations known as RTK (Real Time Kinetics), which provide correction information that improves positioning accuracy.
RTK performs well when users are close to one of these stations. However, the system can be costly and is typically aimed at professional applications.
A different solution is PPP-RTK (Precise Point Positioning – Real-Time Kinematic), which combines highly accurate corrections with satellite data. The European Galileo navigation system now provides these corrections free of charge through its satellite broadcasts.
But the researchers had another tool available as well.
How Google’s 3D Maps Help GPS
While the NTNU team was developing its technology, Google introduced a new service for Android users.
Imagine planning a trip to London. You open Google Maps, search for your hotel, and zoom in to examine the surrounding streets, nearby buildings, and the hotel’s exterior.
Google has built detailed 3D models of buildings in nearly 4,000 cities worldwide. These models allow the company to estimate how satellite signals are likely to bounce and reflect through urban areas. The goal is to reduce positioning errors, including situations where a map incorrectly shows you walking on the opposite side of the street.
“They combine data from sensors, Wi-Fi, mobile networks and 3D building models to produce smooth position estimates that can withstand errors caused by reflections,” Mohamadi said.
Reliable GPS for Self-Driving Cars and Beyond
The NTNU researchers combined Google’s capabilities, satellite correction systems, and their own navigation algorithms into a single solution.
When SmartNav was tested on the streets of Trondheim, it achieved accuracy better than ten centimeters 90% of the time.
According to the researchers, this level of performance provides the kind of reliable positioning needed for navigating complex city environments.
The use of PPP-RTK could also make the technology widely available because it is relatively affordable compared with traditional correction services.
“PPP-RTK reduces the need for dense networks of local base stations and expensive subscriptions, enabling cheap, large-scale implementation on mass-market receivers,” concluded Mohamadi.
Reference: “Phase-Only positioning in urban environments: assessing its potential for mass-market GNSS receivers” by Ardeshir Mohamadi, Hossein Nahavandchi and Amir Khodabandeh, 25 July 2025, Journal of Spatial Science.
DOI: 10.1080/14498596.2025.2536567
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.