Re: False Positives override has no place in any automated vehicle architecture.....
Author: BOB2
Date: 05-09-2018 - 06:46

Multiple "positive" by comparison and confirmation, by multiple inputs, visual, radar/laser ranging, roadside signals from DSRC transponders (on things like traffic lights), and eventually direct "car to car" communications, using transponders, was the ITS architecture we were discussing back in 2008 for automated vehicles, the last time I met with the heads of five state DOT's at ITS America in Palm Springs before I retired.

Like with PTC (in a closed system with trains) you want to have multiple confirmations on critical safety inputs, so you can actually "compare and verify" your location, speed, signal indications, and the like. And, by comparing multiple "confirming" inputs, this automated vehicle would have known this was not a "false positive" but a person with a bicycle, which would have been confirmed easily with current radar or laser technology.

So instead of just relying on "one" input, you would have "multiple" inputs, and if they show different things, and fail to confirm each others indications, then the most "restrictive" indication (in this case the cars radar detection system, had they not disabled it) would have "confirmed" the visual indication was not a "false positive" in the visual detection system, and stopped the vehicle safely in time to avoid a collision.

Ultimately, with transponders in traffic signals, or on things like curve speed warnings, you would also confirm speed restrictions and signal indications to the automated vehicle. And, with electronic "license" plates, using a transponder, each vehicle would also send out a signal, to the other vehicles around it, conveying information on its location, heading, and speed to the other vehicles in its immediate vicinity.

If you are relying on only one input, and only visual, at that, and it is required to be "programmed" to reject otherwise restricting indications, in order not to make the ride less "herky jerky" for the riders, then you would have an inherently unsafe system architecture design, in my "best professional" opinion, based on my ITS architecture design training and experience in developing ITS user needs requirements.

Why I may even need to go see my old FELA "employers", and see if there is also "expert witness" money to be made on these automated vehicle accident lawsuits......?

Safety of the first importance, obedience to the rules (or in this case the laws of physics and of civil liability) is required........

As I noted, this will be another very interesting NTSB report.

I think that reference to losers might have more to do with the racist post, than mine.



Subject Written By Date/Time (PST)
  Running over pedestrians synonymouse 05-08-2018 - 00:53
  Re: Trains "NOT" Running over pedestrians- BOB2 05-08-2018 - 07:22
  Re: Trains "NOT" Running over pedestrians- synonymouse 05-08-2018 - 22:41
  Re: Trains "NOT" Running over pedestrians- <..> 05-08-2018 - 22:44
  Re: Running over pedestrians (Vehicles and Trains) Taxpayer 05-08-2018 - 07:32
  Re: Running over...... Just Sayin' 05-08-2018 - 08:01
  Re: Running over...... Bon Mot 05-08-2018 - 21:24
  Re: False Positives override has no place in any automated vehicle architecture..... BOB2 05-09-2018 - 06:46


Go to: Message ListSearch
Subject: 
Your Name: 
Spam prevention:
Please, enter the code that you see below in the input field. This is for blocking bots that try to post this form automatically.
  *******   **     **  **    **  **     **  ********  
 **     **  **     **  **   **   **     **  **     ** 
 **         **     **  **  **    **     **  **     ** 
 ********   **     **  *****     **     **  ********  
 **     **   **   **   **  **    **     **  **        
 **     **    ** **    **   **   **     **  **        
  *******      ***     **    **   *******   **        
This message board is maintained by:Altamont Press
You can send us an email at altamontpress1@gmail.com