Thursday, March 17, 2016

Caterpillar Self-Driving Trucks

AutoSens Conference to be held on Sept. 20-22, 2016 in Brussels, Belgium, publishes an interview with Bibhrajit Halder, a Software Technical Specialist in the ADAS and Self-Driving team, Faraday Future automotive startup. The interview touches Caterpillar work on self-driving vision-based mining trucks:

Caterpillar self-driving track weights 390 tonnes
(860,000 lb) fully loaded


  1. This obviously goes beyond the usual scope of imaging, but can someone provide a pointer to a decent analysis of economics of this technology with respect to other ways to haul stuff around a mine site? What problem is this solving, and at what cost? You have expensive vehicles, so the cost of driver is not that big, relatively, and gets offset by new need for specialists to program the path and stuff. In a mine setting, how does this compare to a fixed track haul, be it a railway, or a guided track for wheeled vehicles? As far as I know, paths in this context tend to be fairly fixed, and more or less circular - pick up your load, take it to processing or piling point, go back for more. Why do you need awareness, vision, radars, and all this other fancy stuff? You could do this with a conveyor belt.

    1. See CAT's answer here:

  2. This explanation is from a supplier of "Driver Monitoring System", lot of the mines where this type of heavy equipment is used are open pit type. These large trucks haul ores all day long. This can be very monotonous for drivers, who are going around in circular paths. They tend to become dizzy and are prone to get into accidents, i.e. one truck hitting the other. Some of the regulations require the mine to be shut down when an accident happens for investigation, this means loss of millions of dollars for the mine operators. There is an economic incentive to have these driver monitoring systems with imaging and processing capabilities in place to avoid the accidents in the first place.


All comments are moderated to avoid spam.