Should your robot car kill you to save others?

[M Paquette]

Regarding Asimov's Three Laws of Robotics:

1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2. A robot must obey the orders given to it by human beings, except where such orders would conflict with the First Law.
3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

Just bear in mind that he wrote these not as an ideal for all robotics systems to implement, but as an interesting basis to write some really nifty 'locked room' mysteries, the 'how-dun-it' stories of his Robot series that played with ambiguities.

For fun, take a look at Sally.

 

[Texas Proud]

The Blue Ridge Parkway has a LOT of cliffs, and only some guardrails, mostly on curves. In the 13 mile stretch I drive a lot, I remember 3 or 4 cars that have gone over the edge for one reason or another-- could be a flat tire, or started by an animal darting out in the road, or just inattentiveness. I think all came to a stop 50-200 feet down, hung up in trees and such, and no fatalities. I also have a friend that lives out in Telluride and says there's a place where come the spring thaw they often find a car or two that went over the cliff, and they discover what happened to that person that went missing over the winter.

Ahhhh, yes... I remember driving that stretch of road when it was snow covered.... I was sliding around those corners ...

When I got to the other side of the mountain and went to the ranger station... I was told 'sorry, the road is closed ahead due to bad weather'.... he kinda looked at me funny when I told him I just came from that direction...

 

[6miths]

Regarding Asimov's Three Laws of Robotics:

1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2. A robot must obey the orders given to it by human beings, except where such orders would conflict with the First Law.
3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

Just bear in mind that he wrote these not as an ideal for all robotics systems to implement, but as an interesting basis to write some really nifty 'locked room' mysteries, the 'how-dun-it' stories of his Robot series that played with ambiguities.

For fun, take a look at Sally.

Sally is one of the few Asimov stories in which the robot cars do actually get to break the three laws. The laws themselves only evolved over time and weren't mentioned in early Asimov robot stories. And much later still came the Zeroth Law: 'A robot must not harm humanity or, through inaction, allow humanity to come to harm.'

 

[FUEGO]

We're still driving SUV's in the future? What kind of dystopia is this? I was promised a flying car in the future!

 

[martyp]

It is not an engineering problem. It is a classic problem in moral philosophy called the "Trolley Car Problem". It was first proposed in 1967 according to this Wikipedia entry:

Trolley problem - Wikipedia, the free encyclopedia

The choice of what to do is a moral problem. The engineering part of the problem is trivial.

However, this thread does illustrate that ethics should be a part of an engineer's education.

 

[urn2bfree]

How about using some of this amazing technology to fix the ridiculously inefficient traffic lights in this country?! How much wasted fuel (and time) would be saved by moving traffic in a more demand sensitive way than the current and antiquated relatively fixed cycling of lights?

 

[ERD50]

How about using some of this amazing technology to fix the ridiculously inefficient traffic lights in this country?! How much wasted fuel (and time) would be saved by moving traffic in a more demand sensitive way than the current and antiquated relatively fixed cycling of lights?

There you go! I should maybe add this to the pet peeves thread. But now we have red light cameras in so many areas, why can't they use those to move traffic more efficiently?

So often, I sit there at a red light, with zero cross traffic for a minute, and then just as someone is approaching my light turns green, and now they have to stop. We both waste time/gas idling, and waste gas and brakes with the stop/go.

And at busy intersections around here, they could let the left turn green arrow on for as long (or short) as it takes to clear that left turn queue. I think those short lights add to accidents, people try to push the yellow into the red or wait another whole cycle.

Far lower risks than autonomous vehicles, as long as the computer can't set two greens, but that should be easy to actually physically lock out with some old fashioned hard-wired relay logic.

I think it would be a good first step.

-ERD50

 

[Fermion]

Why did the robot car not detect low air pressure in the tire or issue a warning on tread depth/age before letting the car operate? The radar scan should have detected any road debris before it punctured the tire.

Defective robot in my opinion.

 

[springnr]

How about using some of this amazing technology to fix the ridiculously inefficient traffic lights in this country?! How much wasted fuel (and time) would be saved by moving traffic in a more demand sensitive way than the current and antiquated relatively fixed cycling of lights?

Chamber of Commerce in charge of traffic light sequencing?

 

[martyp]

How about using some of this amazing technology to fix the ridiculously inefficient traffic lights in this country?! How much wasted fuel (and time) would be saved by moving traffic in a more demand sensitive way than the current and antiquated relatively fixed cycling of lights?

There is a very good book called "Traffic" by Tom Vanderbilt. It won't have a specific answer to this question but it will provide a glimpse of how difficult it is to engineer traffic.

 

[Midpack]

There is a very good book called "Traffic" by Tom Vanderbilt. It won't have a specific answer to this question but it will provide a glimpse of how difficult it is to engineer traffic.

How hard can it be? Yipes!

 

[urn2bfree]

Let's not let the perfect be the enemy of the good. We don't have to solve every traffic problem, but synchronized and demand sensitive traffic lights would certainly be low hanging fruit.

 

[NW-Bound]

How hard can it be? Yipes!

The above has the following caption: "The longest traffic jam in the world recorded in China. Its length is 260 kilometers." Getting caught in one would drive one insane.

 

[FUEGO]

How about using some of this amazing technology to fix the ridiculously inefficient traffic lights in this country?! How much wasted fuel (and time) would be saved by moving traffic in a more demand sensitive way than the current and antiquated relatively fixed cycling of lights?

I used to do exactly this in my career at one point. Most signals at busy intersections are "actuated" and sense (normally using inductive loops that detect metal) when cars are present. Some can tell how many cars are present and how closely they are spaced (the "density") so they give just enough green time to that direction to get most of the cars through without skimping on green time for other directions (essentially stochastic/predictive algorithm).

Big cities and major arterial streets commonly have "traffic signal systems" that move traffic in a coordinated fashion from signal to signal. If you have ever driven on a street where you seem to hit almost every light on green, even with busy side street traffic, then you're probably traveling on a corridor with a well-timed traffic signal system. For example, I can drive north or south all the way through my mid-sized city's downtown (Raleigh NC) and only hit 1 red light at any time of day on the streets with coordinated signals.

Unfortunately it costs tens of millions of dollars to install the fiberoptics and other signal technology and controls for a major coordinated signal system for a mid size city. In terms of bang for your buck, it's a great way to spend money. Unfortunately politicians have a hard time bragging about making traffic 10% more efficient all over their city by building a coordinated signal system (for a couple of $10MM) when they can build an awesome new bridge or freeway interchange or widen to 4 lanes 5-10 miles of a needed local street at the same price tag. The ribbon cutting ceremony for new pavement or bridges looks pretty sweet in the newspaper. Installing hundreds or thousands of miles of new (hidden) fiber makes it hard to have a ribbon cutting ceremony.

Poor maintenance of traffic signal detector loops and failure to recalibrate programming in signals or entire signal systems of multiple signals can all lead to poor traffic operations on roads. Maintenance spending on transportation infrastructure is disgustingly low. They can build all kinds of stuff, but again, no politician gets kudos for maintaining the hell out of infrastructure. If you can siphon money away from maintenance to fund your other pet projects, and defer maintenance until the next guy's in office, that's a good way to make yourself look good, and stick your successor with the problems.

Working with our state's top transportation decision makers and politicians was frustrating. Sausage was made, and it's best not to look too closely at the process.

 

[Walt34]

Working with our state's top transportation decision makers and politicians was frustrating. Sausage was made, and it's best not to look too closely at the process.

Suspicions confirmed, thank you for the inside story.