Prediction in Dynamic Environments for Autonomous On-Road Driving

Abstract

We have developed PRIDE (Prediction In Dynamic Environments), a hierarchical multi-resolutional framework for moving object prediction that incorporates multiple prediction algorithms into a single, unifying framework. PRIDE incorpo-rates a long-term (LT) prediction approach based on situation recognition and a short-term (ST) prediction approach based on vehicle models. These two approaches provide for the prediction of the future location of moving objects at various levels of resolution at the frequency and level of abstraction necessary for planners at different levels within the hierarchy using sensory data. In this paper, we demonstrate the ability to use the results of the short-term prediction algorithms to strengthen/weaken the estimates of the long-term prediction algorithms via experimental results. We provide experimental results in an autonomous on-road driving scenario using AutoSim1, a high-fidelity simulation tool that models details about road networks, including individual lanes, lane markings, intersections, legal intersection traversability, etc.