State of the art sensors within a single autonomous vehicle
(AV) produce video and lidar data at more than 30
GB/hour. Not surprisingly, even small AV research teams
easily accumulate tens of terabytes of sensor data from multiple
trips and multiple vehicles. AV practitioners would
like to extract information about specific locations, or specific
situations for further study, but are often unable to.
Queries over AV sensor data are different from generic analytics
or spatial queries because they demand reasoning
about fields of view as well as heavy computation to extract
features from scenes. In this demo we present Vroom, a system
for ad-hoc queries over AV sensor databases. Vroom
combines domain specific properties of AV datasets with selective
indexing and multi-query optimization to rise to the
challenges posed by AV sensor data.