Saturday, January 10, 2015

CMU wins FHWA contract for Advanced, Low-Cost Snowplow Visual Guidance System


ASP features include a main computer, human-machine interface,
forward radar sensors, and magnetometers.
Advanced, Low-Cost Snowplow Visual Guidance System

: DTFH6114R00035

: Award

: January 8, 2015

: DTFH6115C00004

: $1,499,669

: Carnegie Mellon University
: Added: May 01, 2014 1:42 pm

Synopsis - Advanced, Low-Cost Snowplow Visual Guidance System

The Federal Highway Administration (FHWA) has a requirement to conduct highly specialized research and develop a prototype low-cost, high-accuracy, visual guidance system for snowplow operators. The conditions for intended use are low and zero visibility environments such as encountered during blizzards, rural nighttime operations, super-cooled fog, and other high visual impairment situations. The effort will include gathering/synthesizing data, distilling this data to parse out technology applicable to the project goals, and designing and assembling an operational prototype unit that will be ready for further validation, enhancement and field use. To successfully complete this project, expertise will be required in diverse fields, including:

  • digital imaging, 
  • digital multidimensional mapping,
  • global positioning systems (GPS), 
  • radar/LiDAR/forward-looking infrared-based (FLIR) sensing, and 
  •  related emerging/existing technologies. 
Experience will also be needed in advanced digital imaging and displays, including

  • augmented reality, 
  • virtual reality, 
  • synthetic vision, and 
  • advanced user displays such as  heads-up displays (HUDs), 
  • visors and other wearable technology.

Related Links





TRANSPORTATION HUMAN FACTORS, 2(3), 247-264
Aaron Steinfeld (now at CMU) and Han-Shue Ian
Partners for Advanced Traffic and Highways (PATH}
University of California, Berkeley
A system to aid snowplow operators during adverse, low-visibility conditions was developed using a quick, low-cost iterative design process The system provided display-independent visual information on lane edges, late ml position, and potential forward collisions Short practice curves were seen on a prototype lateral assist system with drivers achieving a steady state within 3 trials. The subsequent
operational system resulted in positive driver feedback and quick learning periods Drivers had positive impressions regarding ease of use, appeal, and safety Behavior data showed that the shott leaming curves seen with the prototype could also be associated with the operational system. Using the framework of this snowplow problem, an iterative design process is also described that is robust to variations in available time and resources Application specific and process design lessons are
also discussed.

No comments: