It is one of the most exciting races in the automotive industry right now, and it is running at breakneck speed. The prize? Nothing less than the future leadership in sensor technology – and the pressure to offer competitive prices. Manufacturers of vehicle components are working feverishly to develop more advanced assistance systems in order to gain a competitive advantage in the global market for vehicle automation. Sensors are found in almost every part of a modern car, including the engine, cabin, bumpers, wing mirrors, and roof. They provide safety, performance, and comfort.
One thing is clear: the higher the level of automation in cars, the more reliable the sensors in tomorrow’s vehicles will have to be. Manufacturers need to offer their customers comprehensive assistance systems and, where possible, use hardware that will be in demand in the long run. Which poses the question: Which technology will become the gold standard?
For the moment, LiDAR (Light Detection and Ranging) remains a defining technology due to its superior performance. LiDAR uses laser pulses to create highly accurate 3D maps of a car’s surroundings, giving it the ability to see exactly where it is. This makes LiDAR indispensable for navigation and collision avoidance.
LiDAR technology dates back to the time of the US conquest of space. In 1961, the Hughes Aircraft Company developed the first LiDAR prototype. Ten years later, the US space agency NASA used the new technology to map the moon. To do this, laser pulses were transmitted and reflected off the surface, measuring the time between the emitted and reflected light pulses, thus determining the distance. Since then, LiDAR has been integrated into numerous applications, including unmanned aerial vehicles, robotics and, of course, self-driving cars.
Will it be niche?
“At the moment you can find many driver assistance systems using LiDAR – especially in the Chinese market. They can be found not only in luxury cars, but also in mid-range cars, and the trend is clearly continuing,” says Tobias Hofmeier, Global Product Marketing Manager for LiDAR & In-Cabin Sensing at AMS-Osram.
However, Hofmeier explains that in the manufacturing of semi-automated vehicles, especially at L2 and L2+ automation levels, LiDAR-systems are facing some cost issues. This prompted some manufacturers such as Luminar and Innoviz to launch systems that are significantly cheaper and smaller than previous generations. The People’s Republic of China has even brought forth some car models that do not use LiDAR at all, including the AITO M7 Pro from Seres and Huawei, the DeepAL S7 Qiankun from Changan Automobile, and the Mona 03 from Xpeng. These vehicles work with “Pure Vision”, which utilises cameras and algorithms to provide autonomous driving functions.
It looks like Elon Musk might have been right to call LiDAR a ‘crutch’. He has long relied on camera-based perception systems and neural networks to analyse and interpret image data, and badmouthes LiDAR as a cost factor that unnecessarily drives up the price of cars.
This is because, while LiDAR sensors are suitable for highly automated vehicles, laser-based sensors encounter optical systems problems when the weather is bad, the light is particularly bright, or the environment is dusty. It makes the sensors go blind.
Matthias Feulner, Senior Director ADAS at NXP, claims that radar-based systems are a much better solution. “Despite their capabilities, LiDAR sensors have been capped by their high cost, even in high volume production. Radar-based, high-resolution sensors can bridge this gap and drive their use in volume applications. In fully autonomous vehicles, cameras, radar, and LiDAR are still used in combination due to the need for redundancy. At the same time, the number of use cases where LiDAR is absolutely necessary will shrink due to improved radar sensors – until LiDAR ultimately becomes a niche product in the future,” says Feulner.
LiDAR gathering dust on the shelf? That seems like a presumptuous statement. But Chinese trade journals also report that the wind has changed. The automotive newspaper “Zhonggao Qiche Bao”, for example, writes that numerous companies in the People’s Republic, such as Baidu, Huawei, Li Auto, Nio, and Sense Time, are switching to solutions that are completely independent of LiDAR.
Growth opportunities
Obviously, this race is far from over. Just a few days ago, the Hesai Group, which specialises in LiDAR technology, announced an imminent revolution in the LiDAR market for electric vehicles – as it plans to offer its sensors at half the price next year.
“We are moving to a stage where millions of cars sold annually are equipped with lidar,” says CEO David Li. The announced rapid price drop could indeed have a massive impact. Ultimately, whether LiDAR remains the leading technology depends primarily on the cost. Companies need to find ways to reduce costs significantly. They also need to efficiently integrate LiDAR systems into vehicle platforms without this driving up the price. Finally, manufacturers must keep an eye on potential alternative markets. In addition to autonomous vehicles, LiDAR could be increasingly used in areas such as smart cities, logistics, and drone technology.
LiDAR remains a key technology for the time being, especially for highly autonomous vehicles. However, cameras and radar are gaining ground, particularly in price-sensitive markets. While LiDAR impresses with its precision, the factors cost, scalability, and regulatory requirements will ultimately determine which technology dominates the future. Manufacturers need to be flexible and develop hybrid systems that combine the strengths of each technology to win the future.
