The electric car startup Lucid doesn't like the phrase "Tesla killer", but the comparison is difficult to avoid. The company raised $ 1 billion from Saudi Arabia two years ago and is working on Lucid Air, a high-end, battery-powered sedan reminiscent of Tesla's Model S. Lucid is due to officially launch the car in September and start selling next year.
One area in which Lucid wants to stand out from its more established electrical competitor is the advanced driver assistance system (ADAS) called DreamDrive. Elon Musk has ambitious goals for Tesla's autopilot technology, but the company has made efforts to achieve them. One possible factor: Musk has ruled out the use of Lidar, a sensor commonly used by companies trying to develop fully driverless vehicles.
"Anyone who relies on Lidar is doomed to fail," Musk said at an event last year to demonstrate Tesla's advances in self-driving technology. Musk believes cameras and radar will be enough to achieve full autonomy, and that lidar is a "crutch" that prevents companies from making more fundamental breakthroughs.
Lucid disagrees. This week, the company announced that its first vehicle will feature a high-range, high-range lidar sensor. The company hopes to make DreamDrive an industry leader.
How Lidar can help DreamDrive skip the autopilot
Andrei Stanescu / Getty Images
When Lucid launches the Air next year, it will have adaptive cruise control and lane centering features that Autopilot has had for several years. First of all, the Air does not have some of the advanced features of Autopilot, such as: B. automatic lane change.
However, DreamDrive might be superior to autopilot in other ways. The Air will have a camera with a view of the driver. When driver assistance is active, the car monitors the driver's eyes to ensure that they are facing the road. If the driver looks away too long, the car issues a warning.
Eugene Lee, Lucid’s ADAS chief (and unrelated to the author), compares this feature to GM’s Super Cruise, which is widely regarded as the industry’s leading ADAS.
Lucid also has five radars that offer 360-degree coverage. That should make Lucid's cross traffic alarm function more powerful. Tesla cars have only one forward-facing radar.
And Lucids Lidar could help DreamDrive skip autopilot on the basic task of not encountering things. Two Tesla drivers have died after the autopilot did not slow down for semitrailers that crossed the road. A third died when his car crashed into a concrete lane divider. Several other Tesla owners have suffered non-fatal accidents when autopilot steered them into parked fire trucks or police cars in the lane.
To be fair, this problem is not only seen at Tesla. Radar-based ADAS systems typically ignore stationary objects because the radar lacks the resolution to distinguish a stationary car in the lane from a car parked next to the road or a metal sign hanging above it. Image recognition with cameras depends on machine learning algorithms, which can be confused in unusual situations.
In contrast, Lidar is extremely reliable in detecting the position of large, stationary objects such as a tractor or a concrete barrier. Lidar works equally well in all lighting conditions and can recognize unusual objects just as reliably as ordinary objects. For example, Lucid may be able to produce an ADAS that is less likely to hit parked cars than autopilot or other ADAS products on the market today.
An ambitious roadmap
Enlarge /. Eugene Lee, director of ADAS at Lucid.
Lucid will not be the first car manufacturer to launch a lidar sensor – this award goes to Audi, which delivered a lidar-capable A7 a few years ago. However, the lidar from Audi was comparatively primitive – it only had four lines of vertical resolution, while the lidar from Lucid will have more than 100 lines. Lucids Lidar has a field of view of 120 degrees horizontally and 25 degrees vertically, which means that he has a wide view of the street in front of him – but not on the sides or rear. According to Lucid, the lidar will have a range of 150 meters for objects with a reflectivity of 10 percent, which makes it roughly comparable to the lidar units used for fully self-driving prototypes.
Lucid will not be the only car company to sell high-quality lidar sensors in the next few years. Volvo announced in May that it would deliver high-end lidar from Luminar in 2022.
Cars may not need such powerful lidars for basic ADAS functionality, but the sensor could give these companies more headroom. Lucid hopes that a software update will enable more complex functions in the next few years.
The automatic lane change should be delivered about a year after the start with the first major software upgrade. The Air's 360-degree radar coverage could help DreamDrive do this more safely than competitors like Tesla, which only offer forward-facing radar.
The update also allows drivers to take their hands off the steering wheel in certain circumstances, relying on eye-tracking technology to ensure that the driver's attention stays on the road. A powerful lidar sensor offers additional safety margin and detects objects that may be overlooked by the cameras and the radar.
A year later, Lucid aims to deliver a system that does not require active driver monitoring. The vehicle should be intelligent enough to recognize when it is in a difficult situation and to alert the driver to take it over. Volvo is working on a similar highway drive system.
I'll be honest: that makes me nervous. If a driver has not paid attention to the road, it takes a few seconds to fully orient himself. It is easy to imagine that a driver would be surprised by a sudden emergency.
But I would like to prove the opposite. And building such a system will certainly be easier for a car with a powerful lidar sensor than using cameras and radar alone.
Lucid also deserves realistic expectations for its ADAS technology. For several years, Musk has been claiming that Tesla vehicles will soon have such driverless capabilities that customers could earn extra money by shipping their Teslas as driverless taxis. I am skeptical that today's Tesla vehicles will ever have this ability. Experts have told me that Tesla vehicles lack both the sensors and redundant control systems that are required for fully driverless operation to be safe.
Lucid's roadmap seems more realistic. The company is trying to build a great ADAS system, but does not claim that cars that customers buy next year can be operated completely driverless a few years later.