Methods and Systems for Adaptive Triggering of Data Collection

摘要

A computing system may receive a map of features in an environment. The computing system may identify one or more regions of the map for data collection. The computing system may receive sensor data from a plurality of devices. The sensor data may be associated with one or more periods of time when the sensor data was collected by the plurality of devices. The computing system may determine a likelihood of one or more devices being within a portion of the environment that corresponds to the one or more regions of the map for data collection during a future period of time. The computing device may provide a request for given sensor data from the one or more devices based on the likelihood. The given sensor data may be associated with the future period of time.

主权项

1. A method comprising:
receiving, by a computing system that includes one or more processors, a map of physical features in a physical environment; identifying one or more regions of the map for data collection; receiving sensor data from a plurality of devices, wherein the sensor data is associated with one or more periods of time when the sensor data was collected by the plurality of devices; determining, based on the sensor data, a likelihood of one or more devices of the plurality of devices being within a portion of the physical environment that corresponds to the one or more regions of the map for data collection during a future period of time; and providing, based on the likelihood being greater than a threshold likelihood, a request for given sensor data from the one or more devices, wherein the given sensor data is associated with the future period of time, and wherein the given sensor data pertains to the physical features in the physical environment.