Human Factors and DVI Development Task

Experiment 1 – Auditory Warnings

• Objectives
  • Characterize sound environments of vehicles
  • Identify which acoustic properties of sounds are associated with urgency, annoyance, and notice ability
  • Consider that multiple sounds may be required for different warnings
    • How do sounds work together? What is the potential for confusion? How easily are they learned?
  • Can some sound modifications enhance localization of sounds?
  • Are reaction times to warnings affected by pulse characteristics?

Constraints on Auditory Warnings

• Sounds must be audible in the vehicle
• Must sound “appropriate” in context
  • Urgency
  • Annoyance
  • Past experience can influence sound
• A warning sound must be understood quickly
  • Rare events may be a problem
  • Multiple warnings must be distinguishable from each other and learnable

Study of rated annoyance and urgency of sounds

• Subjects listened to sounds that varied systematically in acoustic attributes
  • Sound rated on urgency, annoyance & notice ability
• All factors were highly correlated—most urgent = most annoying
• The most influential sound characteristics on urgency were:
  • Pitch contour (melodic): decreased rated urgency
  • Pulse number (3, 5, 7): increased rated urgency
  • Pulse speed: increased rated urgency
  • Onset time: short onset increased rated urgency

Perceived urgency is increased with

• Higher frequency
• Number of pulses
• Flat pitch contour
• Abrupt onsets

Use of different warning sounds for each warning condition

• How will groups of sounds work together?
  • Urgency is not the only criteria
    • If everything is urgent, sounds may not be distinguishable
  • Learn ability (trials to acquisition criteria)
  • Confusability (Errors)
  • Response efficiency (RT)

Evaluation of Sound Suites

• Constructed 3 suites containing 4 sounds:
  • FCW, LCM, CSW, and LDM
  • Suite A used auditory icons
  • Suite B used modified auditory icons
  • Suite C used abstract sounds based on urgency measures

Conclusions

• Warning sounds that are similar to auditory icons can be:
  • Learned more quickly than abstract sounds
  • Are especially helpful for older drivers
  • Result in shorter choice reaction times
• But…even modest alterations of an auditory icon can result in significant performance differences

Localization enhancements

• Warning sounds were enhanced to improve sound localization
  • Broadband noise
  • Expansion of spatial direction
• Does this improve directional accuracy?
• Does this speed reaction time?
• No evidence that enhancements improve performance

Experiment Design: FCW, LDW Example Scenarios

• FCW triggered when LV (L) suddenly decelerates, Adjacent vehicle (A) block subject (s) from changing lanes
• Construction barrels force subject to change lanes, LDW triggered when subject changes lane with Adjacent vehicle in blind spot

Experiment 3 Shared Warnings

• Which is best
  • 1 master warning
  • a warning for each response class (lateral-steer), longitudinal (steer & brake)
  • a warning for each hazard
  • some combination?
• FCW - Accelerator Release & Brake Onset
• CSW - Accelerator Release & Brake Onset
• CSW - Mean Reduction in Speed
• FCW - Mean Reduction in Speed
• LDW - Distance from Lane Center
• Post-drive Survey, Which Warning Suite is Preferred?

Experiment 4 Design (Warning Delay-Accuracy Tradeoff)

• FCW deceleration
• LDW Lateral Position

Experiment 5 Warning Co-Occurrence

When 2 warnings occur at the "same" time, how should they be presented? Simultaneous, Preempt, Sequential

• Scheduling "simultaneous warnings was difficult (& uncommon)
• No single rule can be recommended
• For some conditions, but not all, simultaneous warnings led the greatest reductions

Influence on Light Vehicle DVI

• Tones (Experiment 1)
  • Guided selection of suite of warning tones that were used in the jury drives
• Number of warnings (Experiment 3)
  • Preference data ruled out four warnings
• Simultaneous warnings (Experiment 5)
  • Guided the length of the warnings

Influence on Heavy Truck DVI

• Threshold for changes to the DVI was high
  • Vorad system subject to several previous evaluations, and years of use
• Tones (Experiment 1)
  • Guided selection of suite of warning tones that were used in the jury drives
• Number of warnings (Experiment 3)
  • Preference data led to dual-hybrid approach
• Simultaneous warnings (Experiment 5)
  • Guided the length of the warnings