The Value of Inferring the Internal State of Traffic Participants for Autonomous Freeway Driving

摘要

Safe interaction with human drivers is one of the primary challenges forautonomous vehicles. In order to plan driving maneuvers effectively, thevehicle's control system must infer and predict how humans will behave based ontheir latent internal state (e.g., intentions and aggressiveness). Thisresearch uses a simple model for human behavior with unknown parameters thatmake up the internal states of the traffic participants and presents a methodfor quantifying the value of estimating these states and planning with theiruncertainty explicitly modeled. An upper performance bound is established by anomniscient Monte Carlo Tree Search (MCTS) planner that has perfect knowledge ofthe internal states. A baseline lower bound is established by planning withMCTS assuming that all drivers have the same internal state. MCTS variants arethen used to solve a partially observable Markov decision process (POMDP) thatmodels the internal state uncertainty to determine whether inferring theinternal state offers an advantage over the baseline. Applying this method to afreeway lane changing scenario reveals that there is a significant performancegap between the upper bound and baseline. POMDP planning techniques come closeto closing this gap, especially when important hidden model parameters arecorrelated with measurable parameters.