State Estimation and Response to Active School Vehicles in a Self-Driving System

This document discloses system, method, and computer program product embodiments for estimating the state of a school transportation vehicle. For example, the method includes capturing sensor data that includes information about an environment of an autonomous vehicle. The method further includes detecting a school transportation vehicle in the sensor data, receiving a list of indicators associated with a set of candidate states for the school transportation vehicle, and analyzing data from one or more sources to determine indicator values. The method further includes using the indicator values to compute a probability mass function that includes a likelihood that the school transportation vehicle is in each of the candidate states. The method further includes imposing one or more goals on a motion control system of the autonomous vehicle based on the probability mass function and causing the autonomous vehicle to operate according to the one or more goals.