Is The Future of LiDAR in Self-Driving Vehicles?
Updated: Mar 28, 2022
LiDAR, also known as light detection and ranging, is a popular method for determining ranges using lasers which target objects. LiDAR is also used for creating 3D visual representations of an area by differing the wavelength of light.
LiDAR technology has been around for over 50 years now, when it was first introduced in the 1960s it was mainly used for terrain mapping. However, since then LiDAR has evolved into something much bigger with its applications spreading into various areas. To name a few, self-driving vehicles, archaeology, AR games, 3D scanning, robotics, tsunami modelling and much more. With all of these different applications of LiDAR, the demand for it is at an all-time high and will continue to grow as the technology improves and new applications are found for LiDAR tech.
So the question that should be asked is as to where LiDAR will go from here, whether it will continue to be implemented with its current applications or broadened into other industries. One of these recent industries implementing LiDAR is self-driving vehicles. There has been quite a debate going on whether LiDAR will be the future of self-driving vehicles or will it get dumped by everyone, according to Elon Musk, once Tesla perfect their self-driving cars. At the moment LiDAR is being used by multiple companies for developing self-driving vehicles but it is expensive and hasn’t been perfected yet. Essentially what LiDAR does is that it sends out lasers in every possible direction and return data on how far the nearest surface away is.
At the Tesla Autonomy Day for Investors on April 22, 2019, Andrej Karpathy, Director of AI at Tesla, discussed how visual recognition is necessary for autonomy which is something that LiDAR tech does not possess. Tesla use their computer vision system which uses visual recognition software to detect objects on the road which allows the AI to make changes based on the perceived objects detected. Elon Musk has clearly shown his dislike for LiDAR in self-driving as he believes it is “expensive and unnecessary” and once you solve the vision aspect of their AI system, it becomes worthless. However, since then LiDAR devices have become a lot more affordable and is expected to drop in price even further by 2025.
Velodyne CEO Anand Gopalan has stated before that Musk’s view on LiDAR technology is 5, 6 years out of date as it has improved since. Gopalan brought up a good argument that is it not wise for people to rely on just one component for your safety. As the cost of LiDAR technology decreases, LiDAR tech can be combined with cameras to provide the safest possible experience in self-driving autonomy. After all the most important factor to consider in developing self-driving autonomy is the safety of the passengers and if combining both LiDAR and cameras achieves this, then LiDAR will still be needed in the future of self-driving autonomous vehicles.