Will we see a real self driving car in our life time?

joeyd999

Joined Jun 6, 2011
6,516
The Tesla’s automated vehicle control system was not designed to, and could not, identify the truck crossing the Tesla’s path or recognize the impending crash.
Could not because it wasn't designed to? Or because recognition was impossible regardless?

If the latter, this is a major problem. Humans are unpredictable to other humans, more so to machines.

Liability costs will prevent mass adoption of autonomous vehicles. I am sure of this.
 

nsaspook

Joined Aug 27, 2009
16,533
Could not because it wasn't designed to? Or because recognition was impossible regardless?

If the latter, this is a major problem. Humans are unpredictable to other humans, more so to machines.

Liability costs will prevent mass adoption of autonomous vehicles. I am sure of this.
It's was not designed to.

The primary on road object recognition system was via Visual camera. When the white truck turned in front of the Tesla in the sunny background the truck side failed to provide the needed contrast low to the ground (where it looks for other cars normally) for detection. The Radar sensor was programmed to trigger up much higher to detect overpasses instead of on road hazards to reduce false positives.

http://fortune.com/2016/09/11/tesla-autopilot-shift/
The shift to radar comes in the wake of the first-ever fatality of an operator using the Autopilot system. The circumstances of the May crash, as well as some similar but nonfatal incidents, suggested that the camera-based Autopilot system may have problems detecting overhanging objects. Final findings of an NTSB investigation of that crash have not been released.
...
While making radar the primary sensor for Autopilot decision-making could help prevent similar incidents, the company says it presents its own challenges. While radar is good at seeing through environmental obstructions like smoke and rain, it is not as good as a camera at detecting people, and has serious problems seeing wood or plastic objects.
 

GopherT

Joined Nov 23, 2012
8,009
Good news for a Tech Companies, the House passed a bill last week that voids a proposed law in Illinois (and proposed lanes in other states) that only allows OEM automakers to approve (install) self-driving technology into a vehicle.

This Illinois bill was the result of the North American Automobile Manufacturers Lobby. Trying to use laws to inhibit innovation under the guise of consumer safety.



The House last week passed a bipartisan bill addressing key concerns of automakers and tech companies, while also taking on safety questions.

The House bill instructs Chao to require “safety assessment certifications” that demonstrate driverless cars “are likely to . . . function as intended and contain fail safe features.” That would have to be done within two years.

The bill also prevents states from regulating “the design, construction, or performance” of automated vehicles, saying that power rests in Washington. Driverless developers have been largely unified in opposing what they call a “patchwork” of regulations that they say would stymie the industry and undercut interstate travel.
https://www.washingtonpost.com/loca...1076f6d6152_story.html?utm_term=.40c1ad869f29


In Massachusetts, Maryland, Georgia, Tennessee, and Illinois, lawmakers have proposed legislation that contains provisions that would limit the deployment of self-driving fleets to automaker-led or automaker-affiliated projects. At least three other states have mulled similar proposals, but resulting bills have not contained the restrictive provisions.
http://blog.caranddriver.com/propos...s-self-driving-cars-and-sideline-competitors/
 

Glenn Holland

Joined Dec 26, 2014
703
All the excitement over self driving cars is nothing but a big hoax to get government $$$ and boost the stock of the companies like Google/Waymo, Uber, Lyft, etc.

As a former automatic train control (ATC) tech for San Francisco's rail transit system, I can confidently say that the concept of "autonomous" self driving vehicles is a crock of Do Do.

Furthermore, if these cars will not have a steering wheel and there's no way for a human to take control of their movement, what happens when the automatic control system malfunctions and it goes into fail safe mode thereby stopping the car in the middle of the freeway?

When our ATC goes into fail safe mode and stops the train, the operator must switch to manual and operate on sight at a maximum of 20 MPH or even less on blind curves. That backs up traffic for a couple of miles and delays service.

However, with 1000s of cars on a major freeway during commute hour traffic, only a very, very small number of failures will bring the entire transportation network to a permanent standstill.
 

GopherT

Joined Nov 23, 2012
8,009
All the excitement over self driving cars is nothing but a big hoax to get government $$$ and boost the stock of the companies like Google/Waymo, Uber, Lyft, etc.

As a former automatic train control (ATC) tech for San Francisco's rail transit system, I can confidently say that the concept of "autonomous" self driving vehicles is a crock of Do Do.

Furthermore, if these cars will not have a steering wheel and there's no way for a human to take control of their movement, what happens when the automatic control system malfunctions and it goes into fail safe mode thereby stopping the car in the middle of the freeway?

When our ATC goes into fail safe mode and stops the train, the operator must switch to manual and operate on sight at a maximum of 20 MPH or even less on blind curves. That backs up traffic for a couple of miles and delays service.

However, with 1000s of cars on a major freeway during commute hour traffic, only a very, very small number of failures will bring the entire transportation network to a permanent standstill.

My son just took a graduate course in control systems with real-time system corrections. The class actually covered the BART ATC system as an "as-is" and teams had to Break down the problem and build up a solution. The professors hypothesis was that any one of the teams should be able to solve the issue for well under $1 billion.

https://www.bart.gov/about/projects/traincontrol
 

Glenn Holland

Joined Dec 26, 2014
703
My son just took a graduate course in control systems with real-time system corrections. The class actually covered the BART ATC system as an "as-is" and teams had to Break down the problem and build up a solution. The professors hypothesis was that any one of the teams should be able to solve the issue for well under $1 billion.

https://www.bart.gov/about/projects/traincontrol
I read the article about BART's plans to upgrade to a so called "Communication Based Train Control" (CBTC) system for $915 Million. By the way, CBTC is actually something called "Positive Train Control" or "Moving Block" which measures the actual position of the train to a few feet rather than relying on an occupancy on moderate to long sections of track.

From my experience with S.F. Muni Railway's CBTC system (installed in the late 90s and early 2000) I can competently say the actual cost will be much higher - possibly around 4 to 5 $Billion for a system that has several 100 miles of track. Furthermore, from my experience with Muni, CBTC is not a reliable technology for mass transit. In fact, BART had a contract with General Electric to build and test a prototype of their CBTC on a limited section of track and it failed most of the tests resulting in a lawsuit. To this day, Muni is still having major problems with its CBTC and trying to maintain and repair it is costing the agency big $$$.

If Google/Waymo, Uber, Lyft want to get in computerizing transportation, they should start with modernizing the rail mass transit industry. Monitoring and controlling vehicles on a fixed guideway is a lot simpler than self driving cars on streets and freeways.
 
Last edited:

Glenn Holland

Joined Dec 26, 2014
703
From the traffic reports during the morning and evening commute, I frequently hear a report about a "Motorcycle down in the center lane of the freeway". That's why there's a box on the DMV license form to check if you want your organs donated in case you're involved in a fatal wreck.
 

GopherT

Joined Nov 23, 2012
8,009
From the traffic reports during the morning and evening commute, I frequently hear a report about a "Motorcycle down in the center lane of the freeway". That's why there's a box on the DMV license form to check if you want your organs donated in case you're involved in a fatal wreck.
Those guys just don’t realize you can steer with your elbows as you slide across lanes of the highway.
 

GopherT

Joined Nov 23, 2012
8,009
.. and when will there be self-driving motorcycles? :eek:
Yamaha is working in it.

They’ve already built a robot to ride it for you. There is just no room for the rider.



Come to think of it, Yamaha has been a leader of self actuating devices. They’ve been making self-playing pianos for years.

 

Glenn Holland

Joined Dec 26, 2014
703
Yamaha is working in it.

They’ve already built a robot to ride it for you. There is just no room for the rider.



Come to think of it, Yamaha has been a leader of self actuating devices. They’ve been making self-playing pianos for years.

Self playing pianos aren't going down the freeway at 55 MPH plus. Furthermore, pianos spend all their time parked and there's no way they could ever get in an accident with another piano - unless there's an earthquake and the floor becomes severely tilted.
 

joeyd999

Joined Jun 6, 2011
6,516
Self playing pianos aren't going down the freeway at 55 MPH plus. Furthermore, pianos spend all their time parked and there's no way they could ever get in an accident with another piano - unless there's an earthquake and the floor becomes severely tilted.
I Googled "player piano accidents". Seems no two have ever collided.

But I did find this.

I'm gonna go jump in the pool, now.
 

Glenn Holland

Joined Dec 26, 2014
703
I Googled "player piano accidents". Seems no two have ever collided.

But I did find this.

I'm gonna go jump in the pool, now.
My mother and step father were piano and organ instructors and most piano performances are solo and there's only one piano on stage at a time.

That eliminates any possibility of a collision between two pianos, however there's always the possibility of a piano hitting a pedestrian. :p
 

GopherT

Joined Nov 23, 2012
8,009
Self playing pianos aren't going down the freeway at 55 MPH plus. Furthermore, pianos spend all their time parked and there's no way they could ever get in an accident with another piano - unless there's an earthquake and the floor becomes severely tilted.
No, but they can come out of tune, that is even more tragic than an accident.

Here is a self-tuning piano system... already 5-year-old technology.
https://www.popsci.com/technology/article/2012-06/2012-invention-awards-self-tuning-piano#page-8
 
Top