Drivers are dangerous

McClintock and Apache wreck photo: Mark Beach/Fresco News

last week’s Uber-involved crash should highlight the current state of affairs: Drivers are dangerous. Tempe Police cited a “human” driver for making a bad left at McClintock and Apache causing the crash with the Uber/Volvo SUV to rollover on its side.  The Uber was being autonomously driven at the time, according to the company.

There are over 100,000 reported wrecks per year in Arizona involving motor vehicle drivers only, with hundreds dead and thousands injured; a driver is at fault in each and every one of those incidents. 

 

Uber’s self-driving cars back on the road after Friday crash in Tempe

…On Friday, a driver in Tempe was cited for making an illegal left turn and hitting one of Uber’s test cars, which was in self-driving mode

There was a pretty detailed article written after the police report was issued with some interesting tidbits, Police Report on Uber Self-Driving Car Crash Tells Complex Tale; although I’m not sure how complex it is, it’s a very common crash. The facts of the crash are crystal clear, so we know, e.g. that an eye witness who told police “It was the other driver’s (referring to the uber) fault for trying to beat the light and hitting the gas so hard… The other person (referring to the uber) just wanted to beat the light and kept going.” we know is factually incorrect, there was no “gunning” and the light was yellow. the uber did not “try” to beat the light, it did beat the light. That’s how signals work.

The driver turning left turned in front of the uber vehicle which was legally proceeding through the intersection; left turners must wait.

This incident is sometimes referred to as Uber Tempe 2017; to differentiate it from “Uber Tempe 2018”, the pedestrian fatality linked below.

“moving screen”

It’s been suggested that the uber could (or should) be programmed to anticipate the left turner was going to violate the Ubers right-of-way; in other words the Uber could have slammed on the brakes to stop before entering the intersection, even though the light had just turned yellow; which would of course cause bitching an moaning from drivers behind the Uber who would inevitably rear-end it. Though some, like @safeselfdrive invariably gild the lilly by claiming the other traffic was already stopped; I didn’t get that from the reports, what seemed to me was the traffic in 1 and 2 was ahead of the uber, which was in lane 3, creating a “moving (visual) screen” from the perspective of the left turner… see illustration; The bicyclist can’t be seen by the left-turning driver because of the green vehicle (that’s why bicyclists shouldn’t be riding on the far right).

Most at Fault

In Arizona, every reported MV crash assigns fault, “Most-at-fault”, to the operator of unit #1 :

Traffic Unit #1 is the vehicle, pedestrian, pedalcycle that caused the collision or was most at fault. — Arizona Crash Form Manual , 2017 update, p.20 

All crashes, by definition involve at least one MV; and as noted above the vast majority only involve a motor vehicle (a “single vehicle” crash) or motor vehicles. A small percentage involve in addition to a MV, a bicycle or a pedestrian (both about only 2% each).

In case you are wondering about MV-bike, or MV-ped crash fault-rates, they run about 50:50. For the period 2009-2015 the rate for bicyclists was 51% bicyclist-at-fault vs. 49% motorist-at-fault. There is some indication that bicyclists are over-faulted, that is to say faulted when there is no violation; but that would just be quibbling.

Motorists have higher fault rates in MV-ped crashes, 60:40 for the same period mentioned above.

There are about 110,000 total reported wrecks per year lately in AZ, fewer than 2% of those involve a bicyclist, likewise fewer than 2% involve a pedestrian.

Tempe Fatality

3/18/2018 An autonomously operating Uber struck and killed a pedestrian

see uber-involved-pedestrian-fatality

 

One thought on “Drivers are dangerous”

  1. Most-at-fault query, peds and bicyclists; see this query to compute by jurisdiction:

    SELECT count(1) FROM 2012_incident i WHERE EXISTS (SELECT 1 FROM 2012_unit u WHERE u.IncidentID=i.IncidentID AND u.eUnitType IN ("PEDALCYCLIST") AND u.UnitNumber=1 ) ;
    SELECT count(1) FROM 2012_incident i WHERE EXISTS (SELECT 1 FROM 2012_unit u WHERE u.IncidentID=i.IncidentID AND u.eUnitType IN ("PEDALCYCLIST") ) ;
    
    SELECT count(1) FROM 2012_incident i WHERE EXISTS (SELECT 1 FROM 2012_unit u WHERE u.IncidentID=i.IncidentID AND u.eUnitType IN ("PEDESTRIAN") AND u.UnitNumber=1 ) ; SELECT count(1) FROM 2012_incident i WHERE EXISTS (SELECT 1 FROM 2012_unit u WHERE u.IncidentID=i.IncidentID AND u.eUnitType IN ("PEDESTRIAN") ) ;

Leave a Reply

Your email address will not be published. Required fields are marked *