[ad_1]
U.S. government regulators They are opening an investigation into Tesla’s Autopilot system after cars crashed into parked emergency vehicles that used this feature.
National Highway Transportation Safety Administration he announced research on Monday, and includes 765,000 Tesla sold in the U.S., a large portion of the company’s sales in the country. The agency says the probe will cover 11 accidents from 2018; the accident left 17 injured and one dead.
NHTSA is studying Tesla’s entire line, including models S, X, 3, and Y, from 2014 to 2021. match the traffic speed.
In every 11 accidents Teslas has been hit first response vehicles parked and marked with flashing lights, flares, illuminated arrow boards, or road cones.
The research will include the full scope of the Autopilot system, how to control and enforce driver attention and commitment on how the system detects and responds to objects and events on or near the road.
Tesla has examined the way Autopilot checks the driver’s attention while activating the system. In the Assessment of Advanced Driver Assistance Systems (ADAS), Autopilot received intermediate marks New European Self-Assessment Program. The system was prevented by the relative inability of drivers to connect with the road.
Like many other ADAS systems, Autopilot requires the driver to keep their hands on the wheel, although these systems can easily be fooled by throwing weight on one of the steering wheel spokes. Recent research by the author Car and Driver they found that it took 25 to 40 seconds for the vehicle to flash a warning when drivers pulled their hands off the wheel, depending on the model. If the drivers did not respond, the car would be driven for another 30 seconds before starting to brake. With highway speed, this system can operate without compromising the driver up to a mile.
In the wake of the accident in California in January 2018, the National Transportation Safety Commission criticized the way Tesla was trying to hold on to drivers. In that regardless, Which is also part of the NHTSA probe, ended up in the back of a 2014 Model S fire truck on the back of the Culver City Interstate 405 freeway. The Tesla driver engaged the autopilot and was following another vehicle in the HOV lane when the lead vehicle changed lanes to avoid the parked fire truck. The autopilot did not deflect or brake, and the driver, who he was eating bagels, did not take control of the vehicle. According to the accident report, the Tesla fire truck hit 31 mph.
The NTSB said the driver’s lack of attention was likely the cause of the accident “due to attention and excessive reliability of the vehicle’s advanced driver assistance system; Tesla Autopilot design that allows the driver to disassemble from the task;
Tesla recently began to change the way Autopilot works, leaving the radar sensor In models 3 and Y in favor of additional cameras. (Models S and X will maintain radar in the near future.) As shown by accidents that are part of the NHTSA probe, radar data does not guarantee that ADAS systems will detect road obstacles properly, although additional sensors generally helped the systems get a full picture of the scene. Because radar and lidar the data are basically a set of measurements that help determine how far the vehicle is from an object. Although ADAS systems can obtain the same information from camera images, they require more complicated computations than radar or lidar. It is not clear in NHTSA research new models with only a Tesla camera.
It is also unclear whether the probe will affect a feature called Tesla’s Full Auto-Drive, the beta versions of which have been released to a group of drivers. Videos of the running system are showing that there is a lot of work to be done and that it needs the driver’s attention at all times.
[ad_2]
Source link
