Dynamic Environment Prediction in Urban Scenes using Recurrent Representation Learning

TL;DR

This paper introduces a neural network-based framework that predicts future occupancy states in urban environments for autonomous driving, enhancing trajectory planning by modeling dynamic agents' behavior using recurrent representation learning.

Contribution

It presents a novel recurrent neural network architecture leveraging occupancy grid data to improve environment prediction accuracy in urban scenes.

Findings

Higher accuracy than baseline methods

Effective prediction of dynamic agents' future states

Validated on KITTI dataset

Abstract

A key challenge for autonomous driving is safe trajectory planning in cluttered, urban environments with dynamic obstacles, such as pedestrians, bicyclists, and other vehicles. A reliable prediction of the future environment, including the behavior of dynamic agents, would allow planning algorithms to proactively generate a trajectory in response to a rapidly changing environment. We present a novel framework that predicts the future occupancy state of the local environment surrounding an autonomous agent by learning a motion model from occupancy grid data using a neural network. We take advantage of the temporal structure of the grid data by utilizing a convolutional long-short term memory network in the form of the PredNet architecture. This method is validated on the KITTI dataset and demonstrates higher accuracy and better predictive power than baseline methods.