Autonomous Vehicles, LEAF and others...

My Nissan Leaf Forum

Help Support My Nissan Leaf Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Is this infographic the first time Nissan (or any credible manufacturer) has stated full door-to-door autonomy (presumably by utilizing a remote human back-up driver) will be delivered by 2020?

Nissan’s future generation ProPILOT technology – available for real-world use from 2020
https://newsroom.nissan-global.com/releases/release-88c03969de1fb342eeb0a427220132ce-infographic-nissans-future-generation-propilot-technology-enables-the-vehicle-to-operate-autonomously-on-urban-roads
 
Followup to earlier post, via IEVS:
Safety Probe Coming For Vegas Self-Driving Shuttle Accident – video
https://insideevs.com/safety-probe-coming-for-vegas-self-driving-shuttle-wreck/

The recent accident involving an autonomous NAVYA Arma electric shuttle in Las Vegas just moments after its official launch, has now became the subject of U.S. National Transportation Safety Board investigation.

The cause of the collision is known – a semi truck backed into the stopped shuttle in an intersection. Fault was found because of the human error of the truck diver, who was then cited and a received ticket, however the accident appeared to still have been easily avoidable – had the shuttle taken appropriate and expected actions to avoid the incident. . . .

“There have been other crashes involving self-driving vehicles but this was the first involving a self-driving vehicle operating in public service, O‘Neil said. Four NTSB investigators were expected to arrive in Las Vegas on Friday.” – via Reuters

Thankfully, the accident was not a serious one, as the Navya ARMA shuttle’s front end sustained only minor damage, and resumed service the next day.

However, the accident could have potentially be avoided if shuttle would have used its horn to alert the truck driver (as would be expected of a human driver) or simple reverse a few feet (the street was clear behind the vehicle at the time). These are areas were autonomous vehicles could be upgraded. . . .
 
Via ABG:
Volvo pushes self-driving Drive Me program back by four years
Safety questions push back introduction.
https://www.autoblog.com/2017/12/15/volvo-drive-me-program-delay/

Volvo is pumping the brakes on its Drive Me autonomous vehicle program, delaying by four years plans to deploy a fleet of 100 self-driving cars in Sweden and acknowledging its technology isn't ready. Volvo says the move will help it better understand how people interact with driverless technologies and improve its first Level 4 autonomous car when it becomes available by 2021.

Volvo four years ago said it would have 100 self-driving cars on the streets of Gothenburg by the end of 2017 in conjunction with the Swedish government to test out the technology and its effects on traffic and the economy. It now says it'll have up to 100 people involved in the Drive Me program over the next four years to funnel customer feedback to the company's engineers.

"On the journey, some of the questions that we thought were really difficult to answer have been answered much faster than we expected. And in some areas, we are finding that there were more issues to dig into and solve than we expected," Marcus Rothoff, the company's autonomous driving program director, told Automotive News Europe. He added the automaker was reluctant to pick a "sensor set," since sensor technology is improving much faster than Volvo predicted in 2013, and electrical architecture has also proven to be a "huge challenge." Volvo also wants its customers to be more confident with Level 4 self-driving technology, which CEO Håkan Samuelsson has said could add close to $10,000 to the price of a vehicle. . . .

"No technology will ever be introduced if there is any question over its safety," Volvo said in a release.
Good on them.
 
Via ABG:
Autonomous Chevy Bolt collides with motorcycle in San Francisco
GM's Cruise Automation not at fault, police say.
https://www.autoblog.com/2017/12/22/autonomous-chevy-bolt-motorcycle-crash/

. . . According to the report filed by Cruise, the Bolt was operating in autonomous mode in heavy traffic at the time of the incident. It was in the center of three lanes on the one-way street, when the car saw an opportunity to switch to the left lane, and began the maneuver. The gap closed, though, and the Bolt returned to the center lane, just as a lane-splitting motorcycle on the right moved into the center lane. Cruise says its Bolt was traveling at 12 miles per hour, while the motorcycle was doing 17 mph.

The Honda S90 motorcycle "glanced" the passenger side of the Bolt (leaving a "long scuff"), wobbled, then tumbled over. The motorcyclist was able to walk away from the incident, moving the bike to the side of the road, but complained of shoulder pain. The police were called, and the motorcyclist was taken to receive medical care. Police determined the motorcyclist to be at fault "for attempting to overtake and pass another vehicle on the right under conditions that did not permit that movement in safety," according to Cruise's report.

In California, lane-splitting by motorcycles is neither permitted nor prohibited under law. The DMV no longer offers lane-splitting guidelines, but suggests a few safety tips for motorcyclists, including watching speed, assuming other drivers don't see them and avoiding blind spots.
 
GRA said:
Via ABG:
Autonomous Chevy Bolt collides with motorcycle in San Francisco
GM's Cruise Automation not at fault, police say.
https://www.autoblog.com/2017/12/22/autonomous-chevy-bolt-motorcycle-crash/

. . . According to the report filed by Cruise, the Bolt was operating in autonomous mode in heavy traffic at the time of the incident. It was in the center of three lanes on the one-way street, when the car saw an opportunity to switch to the left lane, and began the maneuver. The gap closed, though, and the Bolt returned to the center lane, just as a lane-splitting motorcycle on the right moved into the center lane. Cruise says its Bolt was traveling at 12 miles per hour, while the motorcycle was doing 17 mph.

The Honda S90 motorcycle "glanced" the passenger side of the Bolt (leaving a "long scuff"), wobbled, then tumbled over. The motorcyclist was able to walk away from the incident, moving the bike to the side of the road, but complained of shoulder pain. The police were called, and the motorcyclist was taken to receive medical care. Police determined the motorcyclist to be at fault "for attempting to overtake and pass another vehicle on the right under conditions that did not permit that movement in safety," according to Cruise's report.

In California, lane-splitting by motorcycles is neither permitted nor prohibited under law. The DMV no longer offers lane-splitting guidelines, but suggests a few safety tips for motorcyclists, including watching speed, assuming other drivers don't see them and avoiding blind spots.


I've seen 4 of these driving together in the SF area lately.
 
GRA said:
I'd say that's pretty much still dem/val. I'm talking about the point when the public can buy cars with L4 capability.
Waymo says the public will be able to do this with their ride sharing service next year, in Chandler, Az. Might be 2025 before it is cheap, ride share or car purchase, and available across much of the U.S.
 
Too bad PRO-PILOT assist is not yet available fro the comparisons below:

We compared 3 self-driving systems in cars — and the winner is clear

We've tried Tesla Autopilot, Cadillac Super Cruise, and Mercedes Drive Pilot.

Each has its own pros and cons....


And the winner is ... Cadillac Super Cruise!

I stand by my verdict from my original Autopilot-Super Cruise head-to-head.

Super Cruise does one thing quite well. Outside that one thing — hands-free highway driving — I had zero confidence in it, but then again, Cadillac didn't design it to promote any confidence at all off the highway.

I've added another layering of thinking to my views on these autonomous technologies. My question is, "Would I pay for it?"

I can get Super Cruise for under $100,000 and trust it to handle some major highways, so I would seriously considering parting with the cash. With Autopilot, given that the hardware on new vehicles is already in place, I might wait until Tesla improves the system before ponying up the dough.

And with Drive Pilot, although it's the underachiever that overachieves, it's not so much better than good old cruise control that I'd feel obligated to shoulder the considerable extra expense.

So it's Super Cruise for the win!
http://www.businessinsider.com/best-self-driving-systems-in-cars-compared-2017-12/#and-the-winner-is-cadillac-super-cruise-7

Listen here, for a discussion also affirming GM's superiority:

http://www.autonocast.com/blog/2017/12/26/38-supercruise-mazda-and-china
 
What automakers will be offering OTA updates to their driving tech besides Tesla? I see it odd to compare what could be finished products on one vehicle to one that is evolving and offers constant updates and enhancements for free with the ability to upgrade to full autonomous. Will all automakers be offing this value proposition? Do they offer the ability to leverage unused and included hardware to enhance their present capabilities?
 
Via GCC:
New Navigant Leaderboard puts GM and Waymo at the head of the autonomous driving technology pack, Tesla last
http://www.greencarcongress.com/2018/01/20180119-navigant.html

. . .The Navigant AD Leaderboard assesses which participants are best-equipped to be the Leaders in developing complete automated driving stacks—including perception systems, processing, and control software—and services platforms. It quantifies the current relative position of each of these companies as the latest features are being developed and commercialized to help improve safety. . . .

In this latest report, General Motors (GM) and Waymo are effectively in a dead heat on the technology front; both companies are on the verge of having production ready AVs suitable for mobility services.

While Waymo has made progress on the business front with partnerships with Avis, AutoNation, and Lyft, GM still has the edge with the ability to manufacture vehicles and provide services through its dealer network, its captive finance arm, and the Maven mobility service division. . . .

Of note, Tesla ranked last in this Leaderboard, tied with Apple in the Challengers group. Navigant observed that while Tesla has been aggressive in promoting Autopilot, since ending its relationship with Mobileye the company has struggled to reach the same level of functionality with its in-house developed system.

Navigant’s evaluation criteria for the report include:

Vision
Go-to-Market Strategy
Partners
Production Strategy
Technology
Sales, Marketing, and Distribution
Product Capability
Product Quality and Reliability
Product Portfolio
Staying Power
There's an X-Y chart comparing all the companies, with the X-axis showing Strategy and the Y-axis Execution.
 
Via GCC:
udelv makes first public road test delivery from autonomous electric last-mile delivery vehicle
http://www.greencarcongress.com/2018/01/20180131-udelv.html

udelv, a Burlingame, Calif. company, unveiled its autonomous, electric, last-mile delivery vehicle and made the first public road test deliveries from Draeger’s Market in San Mateo to two nearby customers. The deliver route was a 2.5-mile loop with traffic lights, lane changes, unsignalized left turns and two delivery stops. In compliance with existing California regulations, the vehicle was supervised by a safety driver and in test mode.

The distinctive orange customized vehicle is built on a fully electric powertrain and features 18 secure cargo compartments with automatic doors using a cloud-based proprietary technology that is shared between the vehicle, customers and merchants. In its current configuration, the vehicle can drive for up to 60 miles (97 km) per cycle and can load up to 700 pounds (317.5 kg) of cargo.

A dedicated application is available on iOS to track and potentially reschedule deliveries, with an Android version to be released soon.

To complement its autonomous driving technology and ensure reliability of the service, udelv also created an ultra-low latency teleoperations system to monitor and control the vehicles remotely and allow for overrides and human-assisted guidance in unique situations. . . .

The company, with several technology patents pending, is planning to test dozens of udelv vehicles on the roads in a few states within a short timeframe. . . .
 
Via ABG:
Waymo orders thousands of Pacifica vans for self-driving ride-hailing
Commercial service is scheduled to begin in Phoenix this year and expand from there.
https://www.autoblog.com/2018/01/30/waymo-thousands-pacifica-vans-self-driving/

. . . Fiat Chrysler provided Waymo with 100 Pacifica minivans refitted for self-driving testing in 2016, then 500 in 2017. . . .

The companies said the automaker would start delivering "thousands" of minivans in late 2018. Waymo is due to begin offering a ride-hailing service to the public in Phoenix later this year.

"The additional Pacifica Hybrid minivans will be used to support Waymo as it expands its service to more cities across the United States," the companies said. . . .

Last week, Waymo said it began testing self-driving vehicles in Atlanta, bringing to 25 the total number of U.S. cities in which it is testing. .
. .
 
Thanks to https://teslamotorsclub.com/tmc/threads/autonomous-car-progress.99413/page-5#post-2541464I found out that CA DMV https://www.dmv.ca.gov/portal/dmv/detail/vr/autonomous/disengagement_report_2017 are finally up.

Tesla's answer is amusing to me vs. Waymo. I'll have to spend some time to look at the others. https://www.theverge.com/2018/1/31/16956902/california-dmv-self-driving-car-disengagement-2017 is a summary story.

As a refresher, 2016 CA DMV disengagement Reports are at https://www.dmv.ca.gov/portal/dmv/detail/vr/autonomous/disengagement_report_2016.

https://www.engadget.com/2018/02/01/waymo-2017-report-card/ says
Waymo's vehicles drove 2 million miles in self-driving mode across 25 cities in 2017, putting its total autonomous miles to 4 million. It accelerated its testing to prepare for its ride-hailing fleet's launch this year, allowing it to "gather as much data as possible in order to improve [its] technology." According to its annual report submitted to the government of California, Waymo drove 352,545 of those miles in The Golden State from December 2016 to November 2017...
 
https://www.autoblog.com/2018/02/01/waymo-self-driving-human-intervention lists some stats from the 2017 CA DMV disengagement reports I pointed to earlier.

Unrelated:
We were in an accident during an automated driving tech demo
https://techcrunch.com/2018/02/01/we-were-in-an-accident-during-an-automated-driving-tech-demo/

Also, the Navigant chart from much earlier is a bit goofy. How the heck could they know anything about Apple's program to put it on the chart? They haven't released any info.
 
GM To Go Large With Self-Driving Cars In U.S. In 2019:
InsideEVs said:
According to GM, robo-taxis could generate “several hundred thousands of dollars” of revenue over its lifetime, compared to $30,000 on average that GM’s gets from the initial sale.

In 2016, GM acquired Cruise Automation and engaged it to develop fleet of autonomous Chevrolet Bolt EVs.

One of the main goals for the future is lowering the costs of autonomous vehicles. GM decided to equip the Bolt EVs with all the stuff for self-driving from in-house components. Chief Executive Mary Barra said:
GM Chief Executive Mary Barra said:
We are the only company that has this under one roof.
A key element in GM’s self-driving technology, Lidar, is expected to cost $300 at some point in the future (huge decrease from $20,000 today). To make it reality, GM recently acquired Lidar company Strobe Inc. that promised 99% cost reduction.
FWIW, LiDAR is the technology that is claimed to be needed on the Teslas to keep them from driving into parked firetrucks at full speed:

6-750x421.png
 
Back
Top