Why LiDAR Goes Blind in Fog While Automotive Radar Sees Straight Through It

Optical engineers argue that light-based sensors provide millimeter-perfect three-dimensional maps, making them the only viable path to genuine vehicle autonomy. Radar specialists counter that radio waves are the only spectrum that does not go blind the moment the weather turns, rendering lasers a fair-weather luxury that cannot be trusted in a storm.

This debate over how a car should perceive its surroundings defines the modern automotive industry. A buyer selecting a driver-assistance package today is essentially choosing between these two competing philosophies of machine vision, often without realizing the profound technical differences beneath the marketing terminology.[2]

The disagreement is not about software algorithms or processing power, but about the fundamental physics of electromagnetic waves. When a vehicle encounters a dense bank of morning mist on a coastal highway, the survival of its automated systems depends entirely on how those waves interact with suspended water.[1]

The Geometry of a Cloud

Fog is not a gas, but a dense suspension of microscopic liquid water spheres hanging in the air. In a typical coastal or advective fog, these individual droplets measure between one and ten micrometers in radius, creating a highly reflective environment for certain types of energy.[1]

To a human eye, this suspension looks like an opaque white wall because the droplets scatter visible sunlight in every direction. The exact same scattering phenomenon determines whether a vehicle's advanced sensors can penetrate the mist or if they will be blinded by their own reflections.

The deciding factor is the mathematical ratio between the size of the water droplet and the length of the electromagnetic wave trying to pass through it. When the wave and the obstacle are roughly the same size, a specific physical interaction takes over and blocks the signal.[2]

Why Lasers Hit a Wall

Most automotive LiDAR systems fire rapid pulses of near-infrared light at a wavelength of 905 nanometers, which is just under one micrometer. This means the laser's physical wavelength is almost exactly the same size as the fog droplets it encounters on the road.[1]

When a wave strikes a particle of its own size, it triggers a phenomenon known as Mie scattering. The water droplet acts like a tiny, highly efficient mirror, bouncing the light pulse backward and sideways rather than letting it pass through the atmospheric moisture.[2]

"In dense advective fog, Mie scattering can attenuate a 905-nanometer LiDAR return signal by up to 98 percent within just 30 meters," notes the October 2026 arXiv physics review on autonomous sensor degradation.[1]

The sensor fires its laser, but instead of hitting the car ahead, the light bounces off the fog itself. The vehicle's computer receives a blinding wall of static, forcing the system to disengage immediately and hand control back to the human driver.

Some manufacturers are shifting to 1550-nanometer lasers, which are slightly larger and offer marginally better penetration. However, even these longer optical waves remain close enough to the droplet size that heavy fog still degrades their effective range, leaving the vehicle vulnerable in severe weather.

The Radio Wave Advantage

Automotive radar operates on an entirely different segment of the electromagnetic spectrum. Modern systems transmit radio waves at a frequency of 77 gigahertz, which translates to a physical wavelength of 3.9 millimeters, making them fundamentally immune to the optical scattering effect.

A 3.9-millimeter radio wave is nearly one thousand times larger than a five-micrometer fog droplet. Because the obstacle is vastly smaller than the wave itself, Mie scattering cannot occur, and the physical interaction shifts into an entirely different set of rules.[2]

Instead, the interaction falls into the Rayleigh regime, where the tiny water droplets are practically invisible to the massive radio waves. The radar pulse rolls right over the suspended moisture without being deflected, absorbed, or degraded in any meaningful way.

"At 77 gigahertz, atmospheric attenuation from fog or heavy rain is less than 0.1 decibels per kilometer, allowing the sensor to maintain its full 250-meter detection range regardless of visibility," Bosch Mobility engineers explain in their 2023 technical specification.

This is why a vehicle driving into a whiteout can still track the speed and distance of the truck ahead. The radar system provides a continuous, unbroken data stream even when the driver and the optical cameras can see nothing but a wall of gray.

Fusing the Spectrum

If radar is completely immune to weather, a buyer might wonder why automakers bother with expensive laser systems at all. The trade-off lies in resolution, as those massive 3.9-millimeter waves cannot resolve fine details or identify the exact shape of an object.

Radar can tell a computer that a large metallic object is 50 meters ahead, but it struggles to distinguish a parked car from a metal bridge joint. LiDAR, with its microscopic wavelengths, paints a precise three-dimensional portrait of a pedestrian's exact posture and direction.

"The most resilient autonomous architectures do not choose between the two, but fuse them," researchers at the MIT Center for Transportation & Logistics wrote in a March 2026 evaluation of sensor reliability.

In clear weather, the vehicle relies heavily on LiDAR for high-resolution mapping and precise object classification. When the weather turns and Mie scattering blinds the lasers, the computer seamlessly shifts its trust to the lower-resolution but weather-proof radar returns to maintain a safe following distance.

Understanding this physics explains why a modern driver-assistance suite behaves the way it does. It is not a single infallible brain, but a chorus of different sensors, each exploiting a different physical law to keep the vehicle moving safely through a changing environment.[2]

The Showroom Reality

For a consumer cross-shopping modern electric vehicles, this physics lesson translates directly into daily usability. A car equipped only with optical cameras and LiDAR will inevitably demand manual intervention during heavy precipitation, chiming loudly to alert the driver that it has lost its vision.[2]

Conversely, a vehicle utilizing a robust sensor fusion suite that includes 77-gigahertz radar will maintain its adaptive cruise control and lane-keeping functions much deeper into a storm. The radar continues to measure the exact closing speed of the traffic ahead, preventing sudden phantom braking.

This is why regulatory bodies are increasingly scrutinizing systems that rely on a single segment of the electromagnetic spectrum. As automakers push toward higher levels of autonomy, proving that a vehicle can handle a sudden coastal fog bank is becoming a mandatory safety benchmark.

The next time a thick morning mist rolls across the highway, the invisible physics of Mie scattering and the Rayleigh regime will dictate which cars pull over and which continue safely on their way. The safest vehicles will be those that know exactly when to stop trusting their eyes.[2]