Statistics

• Given a route for object A and a starting point for object B, create a route that collides B into A
• Characterize the connectivity of a user's phone/app.
  • Reporting frequency
  • Reporting accuracy (correlate with location?)
  • Accuracy score over a time duration
  • OpenSignal database to predict when a route will lead to signal loss

Location Stream

[ l1, l2, l3, .. ]

stats block { 
  last_reporting_spacing: 3 minutes
  length_of_last_reporting_spacing_steady: 5 hours
  reporting_variability_history: [] 0.0-1.0 (0 steady, 1 unpredictable)
}

a change in steady state comes from a halving or doubling (base2 change)

queries

Given a start time (usually now)

• how far back do points go given a min time gap of x min and a min distance accuracy gap of y meters.
• what is the time gap with the lowest variance over the last x minutes
• cluster the points by timestamp, then how long is the cluster that includes now