Robot Dreams
It's the 21st century all right.
The South African National Defence Force "is probing whether a software glitch led to an antiaircraft cannon malfunction that killed nine soldiers and seriously injured 14 others during a shooting exercise on Friday."
"It appears as though the gun, which is computerised, jammed before there was some sort of explosion, and then it opened fire uncontrollably, killing and injuring the soldiers. ... a female artillery officer risked her life... in a desperate bid " to save members of her battery from the gun."
But the brave, as yet unnamed officer was unable to stop the wildly swinging computerised Swiss/German Oerlikon 35mm MK5 anti-aircraft twin-barrelled gun. It sprayed hundreds of high-explosive 0,5kg 35mm cannon shells around the five-gun firing position. By the time the gun had emptied its twin 250-round auto-loader magazines, nine soldiers were dead and 11 injured.
And here's a little snippet of non-sequitur. What, no error log?
Other reports have suggested a computer error might have been to blame. Defence pundit Helmoed-Römer Heitman told the Weekend Argus that if “the cause lay in computer error, the reason for the tragedy might never be found."
Fortunately the media interviewed somebody who had his own human memory "error log", albeit of different systems.
Electronics engineer and defence company CEO Richard Young says he can't believe the incident was purely a mechanical fault. He says his company, C2I2, in the mid 1990s, was involved in two air defence artillery upgrade programmes, dubbed Projects Catchy and Dart.
During the shooting trials at Armscor's Alkantpan shooting range, “I personally saw a gun go out of control several times,” Young says. “They made a temporary rig consisting of two steel poles on each side of the weapon, with a rope in between to keep the weapon from swinging. The weapon eventually knocked the pol[e]s down.”
Automated weapons systems aren't foolproof. "On Sunday July 3, 1988, towards the end of the Iran Iraq War, the aircraft flying IR655 was shot down by the U.S. Navy guided missile cruiser USS Vincennes between Bandar Abbas and Dubai, killing all 290 passengers and crew aboard, including 38 non-Iranians and 66 children. The Vincennes was inside Iranian territorial waters at the time." [a reader notes that the Vincennes incident was really human erro.]
posted by wretchardthecat at 10/18/2007 04:29:00 PM
16 Comments:
the Vincennes was human error
Coast Guard Vet Practicing Gun Control.
My favorite scene from the Robocop movies was when the "next generation robotic security robot, ED-209" malfunctions and shoots up the corporate boardroom.
View it here...
http://youtube.com/watch?v=o0kWgcIlWn0
At the time of the shootdown, the Vincennes had been in active combat with units of the Iranian Navy all morning, as part of U.S. efforts to prevent the Iranians from interdicting oil tankers in the Gulf.
That a country would think it Okay to have a commercial aircraft overfly a warship which its military had been shooting at for the past several hours - well, that says it all.
The mishap spurred a USN interest in conducting exercises and weapons systems evaluations under conditions similar to actual combat. While the AGIS system had been thoroughly tested they came to realize that such tests invariably involved ONLY the system under test and not associated ships and weapons systems that logically could be operating nearby.
Commercial air has a very unique radar signature, the EW's in CiC screwed up.
Whatever happened to the poor schlub that got cashiered for it?
The Captain got the Legion of Merit, but his minivan was later bombed in mysterious circumstances.
Here's a Powerpoint presentation describing the background of the Vincennes incident, including the sinking of the USS Stark, the tanker war, the heightened tension.
Even with a man in the loop, the parameters of his decision are provided by an analysis and interpretation of data which comes from sensors. Here's another interesting link which describes how the lights flickered in the Vincennes' CIC every time the guns fired, how documents and stuff flew around when the ship made sharp turns; how the crew, by listening in on the voice circuit, partly disabled it; about human interface flaws that required a reset of some switches on the IFF or it would give a wrong reading. Events and stress conspired to create "electronic and verbal confusion".
All under the gun of time constraint:
Take too long to decide, and another kind of unwanted event can happen to you and your shipmates.
So much for the First Law of Robotics.
We all know that flying airliners into things causes no damage, right?
By the way, the USS Stark was not sunk. Although hit by two Exocet missiles fired from an Iraqi Mirage fighter, that brave crew lierally wired their ship back together, and with the aid of an emergency repair ship, brought the Stark back to Jacksonville, FL under its own power.
The Stark was repaired and went back to sea. I used to see it listed as support for Space Shuttle missions in the 1990's. It was decomissioned and then scrapped last year.
RWE,
Thanks for the correction on the Stark. I've been getting sloppy these past few days.
I can testify that the Stark survived. Back in 1993 it was part of the carrier task force that my ship, USS Worden, CG 18, also belonged to.
Stark was struck on May 17, 1987, by two Exocet antiship missiles fired from an Iraqi Mirage F1 fighter during the Iran-Iraq War. The fighter had taken off from Shaibah at 20:00 and had flown south into the Persian Gulf.
Shortly after being routinely challenged by voice on the international air distress frequency by the frigate around 21:07 (local time), the fighter fired two Exocet missiles. The frigate did not detect the missiles with radar and warning was given by the lookout only moments before the missiles struck.
The first penetrated the port-side hull; it failed to detonate, but spewed flaming rocket fuel in its path. The second entered at almost the same point, and left a 3-by-4-meter gash—then exploded in crew quarters. Thirty-seven sailors were killed and twenty-one were injured.
No weapons were fired in defense of Stark. The Phalanx CIWS remained in standby mode, Mark 36 SRBOC countermeasures were not armed, and the attacking Exocet missiles and Mirage aircraft were in a blindspot of the defensive STIR (Separate Target Illumination Radar) fire control system, preventing usage of the ship's Standard missile defenses. The ship failed to maneuver to bring its weapons batteries to bear prior to the first missile impact.
Afire and listing, the frigate was brought under control by its crew during the night. The ship made its way to Bahrain where, after temporary repairs by the tender USS Acadia (AD-42) to make her seaworthy, she returned to her home port of Mayport, Florida, under her own power. The ship was eventually repaired at Ingalls Shipbuilding in Mississippi for $142 million.
http://en.wikipedia.org/wiki/USS_Stark_(FFG-31)
Conflicting reports have surfaced as to the fate of the Iraqi pilot.
The Stark was equipped with the Navy's Vulcan Phalanx Close-In Weapon System, a 3000 round-per-minute automatic gattling gun which fires 50 caliber depleted uranium rounds for close-in anti-air defense. The Stark's system remained in standby mode during the attack.
BTW: My ship, USS Worden, was struck by two anti-radiation missiles during the Vietnam War.
They were fired by a US Air Force jet.
http://en.wikipedia.org/wiki/USS_Worden_%28CG-18%29
This was a "friendly-fire" incident, rather than an example of inter-service rivalry.
:^)
There is a quick read, Set Phasers On Stun, that chronicles various fatal mishaps. The first mishap, from which the book's title comes, describes an X-ray machine procedure. A man undergoing cancer radiation therapy was exposed to a much higher dose of radiation than prescribed. Due to the computer controller being in a variable state not anticipated, the operator, out of line of sight of the patient, only saw that the dose was not administered; she repeated the process multiple times. The man later joked that they forgot to set phasers on stun. He died four months later of acute radiation poisoning.
IMHO... the captain of the Vincennes should have been cashiered.
The people who certified the Vincennes crew to go into a combat zone should have been shot.
And the people who trained the people who did that should be burnt at the stake.
There was a camera crew on the bridge of the Vincennes during the surface action. I've seen better disciplined, more professional kindergarten classes.
This was a Charlie Foxtrot of the first magnitude.
There were also problems with the HCI, and that did not help. On its own it could have led to tragedy. But with the bunch of clowns in the CiC, no system would have stopped them committing canine intercourse sooner or later.
The Stark on the other hand shows the USN at its finest. And some very, very brave and thoroughly professional people. The contrast with the bozos on the Vincennes couldn't be more clear.
Oh yes, I design naval combat systems. If you think that what I said about the Vincennes captain and crew was scathing, well, my opinion of the people of MY PROFESSION at OCZ, some of whom I think I've worked with, would strip the phosphor off monitors, and cause LEDs to fuse.
I still think about work I did 15 years ago, praying every night "Please God May I Not Have Screwed Up". Anyone who doesn't take things that seriously shouldn't be in the business. Good people depend on us.
Post a Comment
<< Home