Tesla CEO Elon Musk Calls Lidar ‘Fool’s Errand,’ Other Experts Agree


Tesla presented its latest and best characteristics of autonomy of vehicles in the recent Day of Investors of Autonomy. The brain of the system is the Tesla Full Self-Driving (FSD) computer, which is now included in all the Teslas that are produced. The hardware package of the autopilot includes 8 cameras, 12 ultrasonic sensors, radar, GPS, an inertial measurement unit and sensors that measure the angle of the steering wheel and the accelerator pedal.

  • This article comes to us courtesy of EVANNEX (which also manufactures Tesla aftermarket accessories). Written by Charles Morris. The opinions expressed in these articles are not necessarily those of InsideEVs.

Above: Tesla's camera and Waymo's lidar (Images: Tesla / Waymo)

A feature that does not have is LIDAR, a technology that uses pulsed laser light to measure the distance to a target, and that is preferred by most other companies that work in autonomous driving (Uber, Waymo, Cruise). In his presentation of Autonomy Investor Day, the ever iconoclastic Elon Musk dismissed the lidar in terms reminiscent of those he has used to describe hydrogen fuel cells. Lidar is "a foolish task," he said. It's "costly" and "unnecessary," and "anyone who trusts lidar is doomed, it's like having a lot of expensive appendages, like, an appendix is ​​bad, well, now you have a lot of them, it's ridiculous, you'll see." .

As TechCrunch reports, LIDAR advocates cite their ability to see through adverse conditions such as rain, fog or dust better than cameras. The disadvantages are that it is expensive and it consumes a lot of energy, and that is why the Tesla system depends on the cameras. Andrej Karpathy, senior director of AI, explained that visual recognition offers a better picture of the real world. He said that lidar systems have trouble distinguishing things like the difference between a plastic bag and a rubber tire. "In that sense, lidar is really a shortcut," Karpathy said. "It avoids the fundamental problem, the important problem of visual recognition, which is necessary for autonomy. It gives a false sense of progress and, ultimately, it's a crutch. "

Along with the cameras, Tesla relies on the vast neural network of real-world driving information recorded by thousands of Tesla vehicles equipped with autopilot on the road. Using various artificial intelligence techniques, Tesla is teaching his system to recognize and react to the wide variety of situations that can be found in nature.

"Everyone is training on the net all the time," Musk said. "Whether the autopilot is on or off, the network is being trained. Every mile that is driven for the car that is [equipped with Autopilot Hardware version 2 or above] is to train the network. "

& # 39;

Above: Anthony Levandowski, former Google / Waymo engineer, now backs off in lidar and says: "Elon is right" (Youtube: TechCrunch)

Some autonomy experts agree with Musk. As Gizmodo reports, Cornell researchers argue that cameras can rival lidar if mounted correctly. Lidar systems are designed to provide a 3D image of the surroundings of a vehicle and the road ahead, so they are often mounted on top of a vehicle to get the best view. However, in a document to be presented at the next Conference on Computer Vision and Pattern Recognition, the Cornell team explains that a pair of inexpensive cameras mounted behind the windshield of a vehicle can produce stereoscopic images that can be converted to data. 3D almost as accurate as the one generated by lidar, at a fraction of the cost.

Mashable reports that the British startup Wayve is a member of the camp that does not need to LIDAR. The company says it does not need multiple data sources, only a GPS system, cameras and a powerful computer are all that is needed to teach cars to drive, as well as humans.

Other players do not agree. A Chinese autonomous driving firm called AutoX implemented a camera-based auto-driving system in 2017, but has now added lidar sensors to its vehicles for redundancy and additional input.

Musk is not totally opposed to lidar: SpaceX uses it in some applications. However, for the autonomy of the vehicles, he believes that the set of Tesla sensors and the amount of real driving data will be sufficient.


Written by: Charles Morris; Source: TechCrunch, Gizmodo, Mashable

  • Editor's note: EVANNEX, which also sells aftermarket equipment for Teslas, has allowed us to share part of its content with our readers for free. Our thanks to EVANNEX. Take a look at the site here.
Source link