Category Archives: Scoop.it

Ford uses NVIDIA Deep Learning for lane position

after a thorough evaluation of the results that they could proudly claim, “we are able to estimate the lane position in 99% of the cases with less than five pixel error”

Read the full article at: safecarnews.com

Lane positioning and precise object location (in this case striped on the road) is one of the biggest challenges toward autonomous driving. Triple redundancy needs all the sources of information possible.

Intel Disrupted: Why large companies find it difficult to innovate, and what they can do about it

It’s becoming clear that extracting intelligence from that big data requires machine learning architectures which are better implemented with non x86 chips from companies like NVidia. It’s possible that by the end of this decade history might repeat itself in Intel’s most profitable segment.

Read the full article at: www.linkedin.com

Intel must think and execute like a startup, otherwise the failure in mobile will repeat as a loss at the enterprise – but not without a fight. More is to gain by offense than defense, however.

Car Wars: The Battle For Automotive Tech

it will likely be the end of the decade before we really know how the car wars play themselves out. In the meantime, it’s going to be one of the most epic battlegrounds for both new world and old world businesses the market has seen in quite some time

Read the full article at: seekingalpha.com

What is missing from this excellent summary of the autonomous driving players is the critical Functional Safety element which will prevent any new player from joining the party until they have decades of proof points. No lawyer wants to take chances while technology is claiming to save lives, not lose more.

BMW, Intel and Mobileye team up for autonomous cars by 2021

The goal of the collaboration is to develop future-proofed solutions that enable the drivers to not only take their hands off the steering wheel, but reach the so called “eyes off” (level 3) and ultimately the “mind off” (level 4) level transforming the driver’s in-car time into leisure or work time. This level of autonomy would enable the vehicle, on a technical level, to achieve the final stage of traveling “driver off” (level 5) without a human driver inside

Read the full article at: safecarnews.com

Clearly we expect some caveats to these claims, and 2021 is only about 12-18 months away from planning start so testing and validation is likely the biggest part of the project. No mention of a communications partner, but the project claims to be “open” and “standards”. We’ll see.

Ericsson establishes 7Gbps connection to moving vehicle using 5G prototype | ZDNet

The trial was undertaken at Ericsson’s Stockholm, Sweden, headquarters on June 10, utilising 5G radio access prototype equipment using a 5G radio interface named New Radio (NR) alongside massive multiple-input multiple-output (MIMO). Massive MIMO combines a high number of antenna elements with beamforming for higher throughput, increased coverage, improved capacity, and more efficient usage of the mobile network

Read the full article at: www.zdnet.com

7Gbps might be enough to keep a high definition map streamed to a vehicle up to date enough to avoid a crash I might think. How many cars at a time can 5G deliver to? Tech is not the limiting factor, it will be economics and scalability that determine if autonomous cars will work on real roads.