Subaru taps Xilinx for key chip in driver-assistance system

19th August 2020

"The system will also be capable of hands-free driving during traffic jams at speeds below about 50 kilometers per hour."

Xilinx Inc and Subaru Corp on Wednesday said the Japanese automaker will use one of the Silicon Valley company’s chips to power a new driver-assistance system that will let drivers go hands-free during traffic jams, among other new features.

The companies said Subaru’s forthcoming Levorg, a midsized hatchback vehicle, will use one of Xilinx’s “Zynq” chips in a new version of its “EyeSight” driver-assistance system.

The system uses two cameras to get a stereo field of view of the road and combines that data with input from radar sensors mounted around the vehicle rather than using more expensive sensors such as lidar, which uses laser light to measure the car’s distance away from objects.

Xilinx’s chip will provide the processing power for the camera images, the companies said. The system will have functions such as slowing the car down before a curve on highways based on an estimation of the curve.

The system will also be capable of hands-free driving during traffic jams at speeds below about 50 kilometers per hour, or about 30 miles per hour, as an add-on option. The feature will only be available in Japan on the Levorg and requires the vehicle to be loaded with high-resolution mapping data.

The driver would still be required to keep his or her eyes forward on the road, and sensors inside the cabin will detect whether the driver is paying attention.

Subaru also plans to update the system over time with new features, which played into the decision to choose Xilinx’s chips, officials said. The chips are what are known as programmable chips, so some of how they work can be changed as new features are added.

Also Read:

NEW UPCOMING ELECTRIC CARS IN NEPAL 2020 | PRICE, SPECS

AMPERE ELECTRIC OPENS DEALERSHIP IN NEPAL

GM BETS ON ELECTRIC CADILLACS AND MICRO-VANS TO REVERSE CHINA SLIDE

TATA MOTORS ROLLS OUT 1000TH NEXON EV IN INDIA

Yatri bike's launch date and price

Review of NIU N-series

Genius electric bikes available in Nepal

Advantages and disadvantages of Electric Vehicles

10% VAT REFUND ON ELECTRIC PAYMENT IN NEPAL

Compiled by :

Team Reviews

Elon Musk Tweets Tesla Will Achieve Full Self-Driving Capabilities

19th August 2020

"Elon Musk had earlier claimed that he had a major program under work at Tesla about which he didn't share details."

Tesla's CEO, Elon Musk recently announced on Twitter to share a variety of updates on Tesla Autopilot and Full Self Driving (FSD). The Dojo Neural Net training centre. Since 2019 April Tesla has been working on a "DOJO", a supercomputer to train neural networks (NN). Elon Musk had earlier claimed that he had a major program under work at Tesla about which he didn't share details.

The mighty super training computer is now called Dojo. The aim behind Dojo is to take in vast amounts of data and train at a video level and do unsupervised training of massive amount of video data with the help of the Dojo program.

The dojo will be enabling faster Neural Network learning. The dojo is used to increase training speed and enable self-supervised learning. But then what is the issue with the existing Autopilot that the new update will be enhancing.

Tesla is developing a NN training computer called Dojo to process truly vast amounts of video data. It’s a beast! Please consider joining our AI or computer/chip teams if this sounds interesting.
— Elon Musk (@elonmusk) August 14, 2020

When / what do you think will be the next big public release that shows a massive leap in Autopilot FSD? I feel like when Navigate on Autopilot came out that was a huge leap, will there be another one like that soon?
— Everyday Astronaut (@Erdayastronaut) August 14, 2020

Our main engineering locations are Palo Alto, Austin & Seattle, but working remotely is acceptable for exceptional candidates
— Elon Musk (@elonmusk) August 15, 2020

Tesla will be releasing an improved Autopilot which will be a limited release in 2 months. Tesla had announced that they were doing a rewrite of the program. As we all know the current Autopilot/FSD is great. In fact, Musk claimed that he has been running an "alpha" version of the update on his personal vehicle.

Tesla has 10 billion km of real-world driving data. This brings to our common curiosities whether the regulatory authorities will find Tesla's autopilot safer for robotaxis to ply on roads.