What is LiDAR?
Light detection and range, more popularly known as LiDAR, is a technology used to detect and position objects in a space. A LiDAR system creates a three-dimensional model of any environment using reflection laser to measure the distance of objects. In this way it is very similar to radar technology, the only difference is the use of lasers instead of radio waves.
LiDAR is used in various applications where accurate object detection or range is required. It can have a resolution of a few centimeters at a distance of 100m, which is significantly better than several meters of radar. The accuracy of LiDAR makes it the preferred choice in altimetry, contour mapping, scanning for AR experiences like in the new iPhone, and various other applications.
Today, the main application of LiDAR is in vehicles for ADAS and autonomous driving functionality. The race to create a low-cost LiDAR system that provides safe autonomous driving capabilities is unfolding as you read this. However, the technology has some problems to tackle and a competitive technology to beat before being a winner. Let’s take a look at the main challenges facing LiDAR.
1. The range
LiDAR manufacturers claim that the technology has a range of 100 meters and in some cases even 200 meters. These claims could be misleading as the range can be defined in different ways. A LiDAR system may not be as accurate at detecting objects at a greater distance in real life situations, even if it can detect a presence.
For example, suppose a self-driving car with LiDAR is moving on the road. A dark object at 100m may not be detected in its entirety due to reflectivity and the LiDAR may not be able to create an accurate 3D map from the reflected laser beam point clouds. The same goes for the case where a bright object is too close to the vehicle and a dark object is further away. Such cases question the claimed ranges of LiDAR devices.
The problem of autonomy must be verified through tests in real life conditions. The range question is about less specific situations and more about LiDAR limits in various cases. Manufacturers and researchers need to find a general solution to this problem to ensure system accuracy.
2. Security issues in extreme cases
As mentioned above, the problem of LiDAR accuracy under certain conditions can be important if it affects safety. In conditions such as fog, rain, and snow bright sun behind a white object, autonomous vehicles of all kinds face detection problems. This can be dangerous and even fatal at worst.
Weather conditions can obstruct LiDAR’s laser beams causing similar problems. Fog and rain are known to limit the use of LiDAR due to the limited penetration and reflection of laser beams under such conditions. Whether it is the weather or some object carried around by the wind, the surroundings mapped by LiDAR become incorrect and the information can be misleading.
The inability to distinguish between a weather phenomenon or everyday objects and a vehicle on the road can be a problem for the self-driving automotive industry. However, this problem is already being worked out using high-powered lasers and better algorithms that can use the data available under such conditions for best results.
3. The cost
Another big problem with LiDAR is its higher cost. Although costs have fallen rapidly over the years, a LiDAR system is still significantly more expensive than the alternative camera vision system. LiDAR still costs around $ 500 for while eight cameras on a Tesla cost less than $ 100. In a competitive market with low margins, it can make a huge difference.
The cost of a LiDAR will continue to decline based on what we have seen over the years. In 2015 alone, a LiDAR unit cost $ 75,000. While cost reduction becomes slower after a certain point, with its higher accuracy LiDAR could soon enter a competitive range compared to cameras.
Common LiDAR devices are electromechanical systems with multiple moving parts. Such systems tend to be less reliable and may see more failures and failures. Add to that the working conditions of vehicles where they go through dirt, water, vibration and all kinds of real world conditions and you have an important system that may not last long before failure.
The creation of reliable LiDAR is possible by reducing the moving parts. Being an engineering problem, it can be solved with better designs. Some solid-state LiDAR systems have been created that could also become the final solution to this long-term problem.
LiDAR is a promising technology for autonomous vehicles. With the resources invested in research and development by automobile and laser manufacturers, it has great potential to find solutions for all challenges. The accuracy of LiDAR can make self-driving cars safer and bring the future closer to all fans of autonomous technology. If you are one of those, keep an eye on the LIDAR space because it will only get better.
The editorial and editorial staff of the Daily Californian were not involved in the production of this advertisement. For advertising and sponsorship opportunities or more information on paid content, please contact [email protected]