Challenges in Front of LiDAR Technology

Challenges in Front of LiDAR Technology

  • Technology
  • August 6, 2022
  • No Comment
  • 12
  • 7 minutes read


What is LiDAR?

Light detection and ranging, better known as LiDAR, is a technology used to detect and range objects in a space. A LiDAR system creates a three-dimensional model of any environment using reflection lasers to measure the distance of objects. In this way, it is very similar to radar technology, the only difference being the use of lasers instead of radio waves.

LiDAR is used in various applications where accurate detection or ranging of objects is required. It can have a resolution of a few centimeters at a distance of 100m which is significantly better than the several meters of radars. LiDAR’s accuracy makes it the preferred choice for altimetry, contour mapping, scanning for AR experiences like the new iPhone, and various other applications.

Today, the main application of LiDAR is in vehicles for ADAS and autonomous driving functions. The race to create a low-cost LiDAR system that provides safe autonomous driving capabilities is happening as you read this. However, the technology has some issues to deal with and competing technology to overcome before it emerges as the winner. Let’s look at the main challenges facing LiDAR.

1. The Range

LiDAR manufacturers claim the technology has a range of 100m and even 200m in some cases. These statements can be misleading, as scope 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 a LiDAR is moving along a road. A dark object at 100m may not be detected in its entirety due to reflectivity and the LiDAR may be unable to create an accurate 3D map from point clouds of reflected laser beams. The same applies to the case when a bright object is too close to the vehicle and a dark object is further away. These cases call into question the claimed ranges of LiDAR devices.

The range issue must be verified by testing in real conditions. The range question is less about specific situations and more about the limitations of LiDAR in various cases. Manufacturers and researchers need to find a general solution for this problem to ensure the accuracy of the system.

2. Security concerns in Edge cases

As mentioned above, the issue of LiDAR accuracy under certain conditions can be significant if it affects safety. In conditions such as fog, rain, snow and bright sun behind a white object, autonomous vehicles of all types face detection issues. This can be dangerous and even fatal in the worst case scenario.

Weather conditions can obstruct LiDAR laser beams to cause similar problems. Fog and rain are known to limit the use of LiDAR due to limited penetration and reflection of laser beams under these conditions. Whether it is the weather or some object carried by the wind, the environment mapped by LiDAR becomes erroneous and the information can be misleading.

The inability to differentiate between a weather phenomenon or everyday objects and a vehicle on the road, could be a deal breaker for the autonomous car industry. However, this problem is already being worked on with high power lasers and better algorithms that can use the available data under these conditions to get the best results.

3. The cost

Another major problem with LiDAR is its higher cost. Although costs have dropped rapidly over the years, a LiDAR system is still significantly more expensive than the alternative camera vision system. LiDAR still costs about $500 each, while eight cameras in a Tesla cost less than $100. In a competitive market with low margins, it can make a big difference.

The cost of a LiDAR will continue to come down based on what we’ve seen over the years. In 2015, a LiDAR unit used to cost $75,000. Although cost reduction slows after a certain point, with its greater accuracy, LiDAR could soon enter a competitive range against cameras.

4. Reliability

Common LiDAR devices are electromechanical systems with multiple moving parts. These systems tend to be less reliable and may see more errors and breakdowns. Add to that the working conditions of the vehicles where they go through dirt, water, vibrations and all kinds of real world conditions and you have an important system that may not last very long before it fails.

Creating reliable LiDAR is possible by reducing moving parts. This being an engineering problem, it can be solved with better designs. Some solid-state LiDAR systems have been created that could also become the ultimate solution to this long-term problem.

LiDAR is a promising technology for autonomous vehicles. With the resources invested in research and development by car and laser manufacturers, it has great potential to find solutions to all challenges. LiDAR accuracy can make self-driving cars safer and bring the future closer to all self-driving tech enthusiasts. If you’re one of those, keep an eye on the LIDAR space as it’s only going to get better.

The Daily Californian editorial staff and editorial staff were not involved in the production of this ad. For advertising and sponsorship opportunities or more information about paid content, please contact [email protected]





Source link

Related post

JavaScript-React Developer – IT-Online

JavaScript-React Developer – IT-Online

Platform for developers / Application integration Understanding of BI tools will be a plus Understanding the integration between different technologies Coordination…
JavaScript-React Developer – Gauteng Pretoria

JavaScript-React Developer – Gauteng Pretoria

Platform for developers / Application integration Understanding of BI tools will be a plus Understanding the integration between different technologies Coordination…
How to access Netflix games on iPhone and iPad

How to access Netflix games on iPhone and iPad

While Netflix has made a name for itself as a streaming service that offers movies and TV shows, including an extensive…

Leave a Reply

Your email address will not be published.