Headlights may or may not be necessary for future automobiles.
Headlights may or may not be necessary for future automobiles.
( Source: gemeinfrei / Pexels)

AUTONOMOUS DRIVING Are headlights necessary for future automobiles?

Author / Editor: Seth Lambert / Isabell Page

Because autonomous vehicles are equipped with radar, sonar, LiDAR, and even (in some cases) cameras that can see in the dark, the question arises whether traditional headlights will still be necessary for future cars.

Automobiles with at least some level of autonomous functionality are becoming commonplace, even if they can’t drive us where we want to go or let us take our attention off the road (yet). Since these vehicles are equipped with a variety of cameras and sensors, it may be reasonable to ask: do they still need headlights to drive in the autonomous modes they offer?

Before we can make a definitive determination regarding this question, a more rigorous examination of these vehicles’ sensing technologies is in order.


Radar uses radio waves to bounce off of objects to measure how far away they are. A transceiver/sensor both transmits the waves and receives them and is able to measure the distance to objects based on the time it takes for the reflected radio waves to come back to the sensor. Radar certainly doesn’t need light to operate, and indeed, it’s technology that’s proven especially helpful to multiple modes of transportation in the dark and/or weather conditions that result in poor visibility.

Prototype autonomous vehicles (AVs) have been equipped with as many as eight radar sensors so that these vehicles can perceive objects, obstacles, and oncoming hazards in 360 degrees at distances up to 300 meters.


Sonar (sometimes referred to as ultrasound because of the high frequencies that are used) is another type of waveform that’s employed similarly to radar in AVs. Prototype AVs have utilized as many as ten sonar/ultrasonic transceiver/sensors for additional environmental perception. But sonar can be prone to returning false information by bouncing, and sound waves move significantly slower than light waves and/or radio waves; this makes them less accurate for perception when a vehicle is in motion.


LiDAR, which stands for “light detection and ranging,” is a technology that uses lasers in the same way that radar uses radio waves and sonar uses sound (LiDAR is more accurate than both). Typically, LiDAR laser beams are not visible, and often, the lasers are spun on motors and emitted in 360 degrees from a central source. However, LiDAR technology tends to be much more expensive than that of radar and sonar, so prototype AVs that have been equipped with multiple LiDAR transceiver/sensors have tended to be used for commercial purposes, rather than as consumer automobiles (at least so far).

The costs of LiDAR technology are falling as sensors can now be created using not much more than individual solid-state computer chips, but even solid-state LiDAR is not a panacea for comprehensive AV perception. Research has shown that LiDar is less functional in precipitative weather than it is in clear conditions.


Lastly, digital cameras - either of the normal variety or those that can operate in infrared light - are used by many AVs as redundant technology for detecting objects, obstacles, and oncoming hazards, besides radar, sonar, and LiDAR sensors. In the case of cameras that operate in infrared light, that light needs to be transmitted from the vehicle in question, so even in a case of a car without headlights, another (invisible) type of light would be projected from the vehicle.

It should be understood that all of the above sensing methods have advantages and disadvantages; a camera cannot "see through" solid objects, creating blind spots when obstacles stand in the way of other objects or hazards, while certain types of materials may be at least semi-permeable by sonar or radar waves or even LiDAR laser beams, allowing AVs to potentially “see through” certain objects, hazards, or even other vehicles for better perception than cameras (or human eyes) could offer.

Will headlights still be necessary?

But in the end, no matter how good all of the above technologies are, the question of whether headlights may be necessary on a vehicle may come down to whether it will be operated by a human driver.

AVs that function at Society of Automotive Engineers (SAE) Level 3 (“conditional automation”) only drive themselves for limited distances (generally exclusively on highways), so that when drivers take back control of their vehicles, they would again need headlights in any conditions of poor visibility. Even at SAE Level 4 (“high automation”), the deciding factor would still come down to how often a human would need to drive the vehicle.

At SAE Level 5 (“full automation”), AVs will drive themselves from the starting point of a journey to its destination, but then there’s the question of other vehicles. Will these function at SAE Level 5 as well? If they have human drivers, some kind of light emitting from all vehicles may still be necessary to prevent accidents and collisions from occurring. There are also bicyclists, motorcyclists, and pedestrians to consider. In each of these cases, a vehicle’s headlights serve a double purpose - they allow a human driver to navigate at night and/or in inclement weather conditions, but they also give an advance warning of a vehicle’s approach to those people (and/or animals) outside the vehicle. Therefore, until all cars on the road are exclusively SAE Level 5 AVs and adequate safety systems (perhaps powered by intelligent transportation systems [ITS] technologies) are in place for bicyclists, motorcyclists, and pedestrians, most cars will likely still need to have and use headlights on a regular basis.