Hypothetical Manhattan Pickup Exercise

Hypothetical Manhattan Pickup Exercise

animation
gps
location

Hypothetical Manhattan Pickup Exercise

Assume reflected GPS signals displace some pickup pins by 20–50 m; model how wrong-side-of-street placements affect calls, delay and driver idle cost. Do not attribute the scenario figures to Uber.

Hypothetical mechanism: Manhattan’s dense skyscraper grid can create multipath interference. For this exercise, assume reflected signals add 50–200 ns of delay, corresponding to illustrative 15–60 m range-bias values.

The technical challenge in numbers:

Environment Assumed GPS error Assumed pickup-success rate
Open suburban 3-5 meters 94%
Low-rise urban 8-15 meters 87%
Midtown Manhattan 20-50 meters 72%
Financial District (canyon) 30-80 meters 61%

Possible multi-technology mitigations for the hypothetical urban scenario:

  1. Sensor fusion: Combine GPS with phone accelerometer and gyroscope data; dead reckoning between fixes may help during multipath events.
  2. Map matching: Snap estimated positions to the road network using a suitable model, while retaining uncertainty about the match.
  3. 3D building models: Use building-height data to estimate satellite visibility and potential multipath interference.
  4. Wi-Fi fingerprinting: Add access-point signal strengths as supplementary positioning data where coverage supports it.

Modeled outcomes:

  • Manhattan pickup accuracy improved from 25-50m median error to 8-12m
  • Assumed phone calls for pickup coordination drop 62%
  • Assumed average pickup time decreases by 2.8 minutes
  • Assumed annual savings are $32 million in reduced driver idle time

Lesson for IoT designers: Any IoT application that depends on GPS positioning in urban environments must budget for multipath mitigation. Raw GPS alone is insufficient – sensor fusion, map matching, and supplementary positioning technologies are essential for sub-10-meter accuracy in built-up areas.

Back to the chapter