To bring fully autonomous cars to market, automakers must equip vehicles with sensing technologies to create a virtual 3D map that can plan a path through our real world of roads, bridges, tunnels and more. Lidar technology provides the high-resolution, three-dimensional information about the surrounding environment needed to make fully-autonomous driving a reality.
Lidar technology commonly uses a pulsed laser to measure the distance to an object. The system bounces laser pulses off objects at an enormously high rate – millions of laser pulses each second – and measures how long the light from the laser takes to reflect from the object to a light detector. Based on the time of flight, the distance to the object is calculated by the lidar device in real time, and the millions of data points are used to generate a 3D point cloud, which is a complex “map” of the surroundings.
Current state-of-the-art lidar systems are largely based on sensors using one of two lidar wavelengths: 905 nm (nanometers) and 1550 nm. Utilizing either lidar wavelength presents engineers with tradeoffs for consideration. Let’s look at three considerations around lidar wavelengths: laser safety, water absorption, and technology components.
All commercially-sold lidar products must achieve eye-safety certification via compliance with the U.S. Food & Drug Administration (FDA) eye-safety performance standard, which conforms with the International Electrotechnical Commission (“IEC”) 60825 standard. If sensors are designed to meet eye safety standards, both wavelengths can be used safely.
All of Velodyne Lidar’s sensors operate at 905 nm, which is the level of the safest types of lasers (Class I). Any single laser beam from a sensor that may sweep across an eye would only be in contact for a fraction of a second, with an average power less than common laser pointers. Also, since each laser points at a different angle, multiple lasers striking the eye simultaneously would not cause harm because distinct areas of the retina are illuminated. The combination of low power and rapid rotation maintains eye safety.
1550 nm systems use a wavelength that is allowed to run more power compared to 905 nm. However, under certain conditions, the 1550 nm wavelength of light can still cause corneal damage and potential damage to the eye lens.
Another factor when evaluating lidar wavelength is water absorption. This is an important issue for automotive lidar systems given the adverse weather conditions cars can encounter on the road. Pulsed time-of-flight sensors operating at the 1550 nm wavelength experience substantially reduced capabilities in rain, fog and snow conditions, thereby degrading returning image data used to create point clouds.
An academic paper originally published in Opto-Electronics Review noted that 1550 nm waves had much higher water absorption – two orders of magnitude higher. The study found 905 nm systems to be a “much more weather−proof solution.” Under most realistic settings, as discussed in LaserFocusWorld, loss of laser light at 905 nm is less because water absorption is stronger at 1550 nm than at 905 nm.
In part to offset this degradation, 1550 nm systems send out more laser light to achieve performance comparable to 905 nm systems. As a result, 1550 nm systems typically consume more electrical power. This could create the need for larger systems which must be stored in a vehicle (most likely in the trunk, due to size). This makes it harder to scale from a lab-instrument-grade technology to large commercial volumes to support automakers.
Additionally, the lower the power consumption of sensors on the car, the more energy that an electric-powered autonomous car’s battery can devote to driving functions. So, sensors with high power requirements will be less attractive to use in electric vehicles. Moreover, high power consumption might sometimes lead to excess heat generation that requires active cooling systems, which add to cost and complexity.
A third important consideration when assessing lidar wavelength is technology components. Currently, key components for 1550 nm systems are either custom-made or available only through non-standard supply chains. Despite significant advancements driven by the telecom industry over recent decades, 1550 nm lasers and detectors “aren’t cheap because they require more exotic materials to manufacture.”
On the other hand, 905 nm lidar systems can be built using widely available, lower-cost CMOS technology, as noted in Electronic Specifier. Given their more widespread deployment, 905 nm lidar systems have a well-established ecosystem that provides standard, off-the-shelf components.
As engineers assess lidar technologies, it is important to examine foundational architecture characteristics such as wavelength. By looking at wavelength through the lenses of safety, water absorption and component supply chains, companies can determine the best fit for their engineering, development and production needs.