Researchers have reported on a solar storm event that produced a GPS glitch serious enough to crash self-driving cars. The finding adds a critical new dimension to the growing concern about autonomous vehicle safety: space weather. While much of the public discussion around self-driving car failures focuses on software bugs, sensor limitations, or unpredictable pedestrians, the Earth’s magnetic environment is now being recognized as a hidden vulnerability. A single, moderate geomagnetic storm can distort the timing and location data provided by GPS, and for vehicles that rely entirely on satellite navigation to know their position, the result can be catastrophic.

What is a solar storm?

A solar storm, also called a geomagnetic storm, begins with an eruption on the Sun’s surface. Coronal mass ejections (CMEs) hurl billions of tons of plasma and magnetic field into space. When these charged particles collide with Earth’s magnetic field, they can disturb the ionosphere, the part of the upper atmosphere from about 60 kilometers to about 1000 kilometers altitude. Scientists measure the severity of these events using the Kp index and the Dst index, which quantify how much Earth’s magnetic field is disturbed. Strong solar storms are more common near the peak of the Sun’s eleven-year activity cycle. During such periods, multiple CMEs can strike Earth in quick succession, causing prolonged disturbances in the ionosphere.

The most famous solar storm in history is the Carrington Event of 1859, which caused auroras visible at the equator and set fire to telegraph systems. Modern society is far more dependent on radio signals, power grids, and satellite navigation than the telegraph. A similar storm today could seriously disrupt communications and navigation systems across the globe. While a Carrington-level event is rare, recent research examines storms of more moderate intensity that nonetheless have measurable and dangerous effects on GPS accuracy.

How solar storms disrupt GPS

GPS satellites transmit signals that travel through the ionosphere to receivers on Earth. These signals contain precise time stamps that allow the receiver to calculate distance to each satellite. By triangulating signals from multiple satellites, a GPS receiver determines its position with an accuracy of a few meters under normal conditions. But when the ionosphere is disturbed by a solar storm, the propagation speed of the satellite signal changes. This creates an unexpected delay, which translates directly into a positioning error known as an ionospheric delay.

During severe space weather events, the ionosphere can become highly turbulent, causing rapid fluctuations in signal amplitude and phase. These disruptions are called scintillation, and they can even cause a GPS receiver to lose lock entirely. Even without scintillation, the simple delay errors can push a vehicle’s position off by tens of meters. At highway speeds, an error of just ten meters could place a self-driving car in the wrong lane, or prevent it from stopping before an intersection. The report from the researchers emphasizes that such errors are not rare in regions near the magnetic poles or in the equatorial anomaly zone, where ionospheric disturbances are most intense.

Self-driving cars depend on GPS

Autonomous vehicles rely on a suite of sensors, including cameras, radar, lidar, and ultrasonic sensors. However, GPS remains a fundamental component for determining global position and for route planning. A self-driving car uses GPS to know which street it is on, where it is relative to its destination, and what maneuvers it should perform next. High-definition maps are georeferenced to GPS coordinates, meaning that if the GPS position is wrong, the car may believe it is in a completely different location. When that happens, the vehicle’s path planning and decision-making algorithms are applied to the wrong context, which can lead to sudden braking, unnecessary lane changes, or a collision with another vehicle or a stationary object.

Most production systems implement some degree of sensor fusion: they compare GPS data with the observed environment and dead-reckoning from wheel speeds and steering angles. This can mitigate against occasional GPS dropouts. But the research report suggests that a cleverly distorted GPS signal can contaminate the entire perception pipeline. If the car’s internal mapping module receives a shifted position that is consistent with the road geometry, the vehicle may not detect the error. The sensor fusion algorithm might then trust the false GPS position and override the visual system, leading to a crash. The report documents multiple instances during a specific solar storm where autonomous vehicle test platforms deviated from their intended trajectories by more than twenty meters without issuing any warning alerts.

From positioning error to potential crash

The researchers analyzed the sequence of events during the solar storm and traced the GPS glitch to a rapid variation in the ionospheric total electron content (TEC). They found that the TEC gradients were steep enough to introduce a ranging error of over 30 meters on some satellite links. In a dual-frequency GPS receiver, such an error would normally be mitigated by the two-frequency ionospheric correction. However, many mass-market vehicle navigation systems use single-frequency receivers to reduce cost and power consumption. These receivers cannot apply the ionospheric correction directly, making them especially vulnerable. According to the report, even dual-frequency receivers can experience problems if the ionospheric gradients are rapid, because the differential correction may be invalid for the actual signal path.

Using the data from the storm, the researchers simulated the response of a self-driving car’s navigation stack. In the simulation, the car was traveling on a divided highway at a speed of 110 kilometers per hour. When the false position shifted the vehicle into the left lane, the perception system recognized the lane line markings, but the planning module did not command a corrective action because the GPS data suggested the car was still centered in the original lane. This mismatch between perception and GPS caused the car to drift further left, eventually crossing into oncoming traffic. The virtual incident ended with a head-on collision in the simulator. The researchers note that this is a plausible scenario, not merely a theoretical one, because the actual GPS errors observed during the storm were nearly identical to those used in the simulation.

The report also highlights that the solar storm was not considered extreme by space weather standards. It was a moderate storm with a Kp index of 7, which occurs several times every solar cycle. Such storms can easily go unnoticed by the general public, yet they can generate GPS errors that are severe enough to break normal vehicle safety assumptions. The researchers argue that self-driving car developers should treat space weather as a non-negligible threat and design their systems with redundant positioning methods that do not exclusively rely on satellite signals.

Mitigation and future directions

Improving the resilience of autonomous vehicles to solar storms requires a multi-layered approach. One option is to equip vehicles with dual-frequency and multi-constellation GPS receivers, which are significantly more robust to ionospheric disturbances. Another is to use ground-based augmentation systems, such as differential GPS or real-time kinematic positioning, which correct for ionospheric delays based on base stations with known positions. However, these corrections are only effective over relatively short distances and are not always available in rural areas.

In addition, vehicle software can be updated to detect anomalous GPS inconsistencies. If the reported position jumps by more than a few meters within a second, or if the number of visible satellites suddenly drops, the system could enter a safe mode: slow down, pull over, or switch to a completely map-free navigation strategy using dead reckoning and perception alone. The researchers suggest that autonomous vehicle developers should study real-time space weather data from organizations such as the NOAA Space Weather Prediction Center, which issues forecasts and alerts for geomagnetic storms. By integrating these alerts into the vehicle’s risk management system, the car could automatically increase following distance and reduce speed when a solar storm is expected to hit.

Another area of development is the use of inertial measurement units (IMUs) combined with high-precision odometry. Modern IMUs can maintain position accuracy to within a few centimeters over short distances, and they are unaffected by solar storms. By integrating IMU data with GPS more tightly, the vehicle can bridge longer periods of GPS loss or corruption. This is already a common practice in terrestrial robotics and aviation. But low-cost IMUs used in today’s production cars are not reliable enough to sustain navigational accuracy for more than a few seconds without GPS updates. The high cost of tactical-grade IMUs remains a barrier to mass deployment.

Car manufacturers and technology companies are also exploring vehicle-to-vehicle (V2V) communication, which could allow neighboring vehicles to share precise positions derived from sources other than GPS. A vehicle that still has a valid GPS fix could broadcast a correction factor to nearby vehicles that are suffering from ionospheric distortion. Similarly, infrastructure such as smart traffic signals could act as fixed reference points, transmitting their known locations as beacons for passing vehicles. These approaches would make the entire fleet more resilient, but they require widespread deployment and robust authentication to prevent malicious spoofing.

The recent report adds a valuable dataset to a growing body of evidence that space weather is a real and present danger to modern transportation systems. From aviation to maritime navigation, GPS has long been recognized as a critical infrastructure that can be degraded by solar activity. Now, with autonomous vehicles on the horizon, the risks extend to every road user. The researchers recommend that regulatory bodies develop standards for space weather resilience in autonomous vehicle certification. They also propose creating an empirical risk map that shows regions where ionospheric disturbances are most likely to create dangerous GPS errors, so that automated drivers can be forewarned when traveling through those areas.

As the Sun continues its cycle of activity, the frequency and intensity of geomagnetic storms will fluctuate. Future storms may be stronger than the one studied in this report, and even moderate events at the wrong time could have serious consequences. For now, the research serves as a wake-up call: self-driving cars must be designed to understand the sky above them, not just the road below.

Source:TechRadar News