LiDAR; Mapping Out the Future of Autonomous Vehicles

•••
Blog
September 2018


If you’ve seen autonomous cars driving around cities, you might have wondered what the bulky box mounted on the roof does, or what it is. Well, wonder no longer – christened ‘LiDAR’, this boxy piece of hardware wields a crucial role in the race to rollout autonomous cars to the public. An acronym for ‘Light Detection And Ranging’, LiDAR uses pulsed laser beams to enable self-driving cars to ‘see’ obstacles in their path, and react accordingly.

Below, we’ve outlined a brief overview of LiDAR’s origins, how it works and how, arguably, it’s the key to the future success of the autonomous vehicle.
 
The History of LiDAR
LiDAR emerged in the 1960s, just after the advent of the laser, entering into public consciousness during the 1971 Apollo 15 mission, when astronauts mapped the surface of the moon. Certainly, it was widely used in the field of archaeology to map out large areas of land, providing benefits to both archaeology and agriculture. But it wasn’t until the 2000’s that LiDAR found a niche in the automotive industry, when it was used in the 2005 DARPA Grand Challenge – a driverless robot car race, albeit in a rudimentary, 2D form.
 
How does LiDAR work?
LiIDAR’s sensors continually fire off beams of laser light, measuring how long it takes for the light to return to the sensor and accurately calculating the distance to the object. By firing millions of beams of light per second, LiDAR can create measurements which build a 3D visualisation of its surroundings – a map.

LiDAR enables a self-driving car to see a 360-degree view of its surroundings, at all times. It has incredibly accurate perception of depth, with an accuracy of +2cm of objects in relation to the vehicle. This kind of predictability is exactly what technology like self-driving cars requires, and has been a significant contributing factor in the progress in autonomous vehicle technology over the last 5 years.
 
LiDAR today
Modern LiDAR enables cars to differentiate between a person on a bike, or a pedestrian, and even at what speed and which direction they are going in. The combination of accurate navigation, predictability and high-resolution object-tracking has meant that LiDAR is the key sensor in self-driving cars today, and it’s hard to see that domination changing.

However, despite its impressive capabilities, LiDAR isn’t without its flaws. It can be affected by adverse weathers such as rain, snow and especially fog. It’s also incredibly expensive – average costs are £50,000 per car – which makes universal accessibility an issue.
 
The Future of LiDAR
For years, the industry leader in LIDAR has been Velodyne, but the rapid advance of research, investment and development in self-driving vehicles has prompted other firms to start exploring LiDAR too. Competition between manufacturers could help to decrease costs, with aims to bring prices down to approximately $1,000 a unit.

Competition is also encouraging further advancement in LiDAR technology, especially a focus on creating higher resolution output and increased tracking ranges, to provide better object recognition and tracking.