Gaussian Lane Keeping: A Robust Prediction Baseline

TL;DR

This paper introduces Gaussian Lane Keeping (GLK), a robust, interpretable baseline prediction method for autonomous vehicles that outperforms neural network models on the CitySim dataset, aiding development and deployment.

Contribution

The paper presents GLK, a novel, robust baseline prediction model for autonomous driving that is interpretable, requires less training, and outperforms neural network approaches.

Findings

GLK outperforms neural network-based predictions on CitySim dataset.

GLK provides a robust and interpretable baseline for autonomous vehicle prediction tasks.

Extensions to the GLK model improve its accuracy and applicability.

Abstract

Predicting agents' behavior for vehicles and pedestrians is challenging due to a myriad of factors including the uncertainty attached to different intentions, inter-agent interactions, traffic (environment) rules, individual inclinations, and agent dynamics. Consequently, a plethora of neural network-driven prediction models have been introduced in the literature to encompass these intricacies to accurately predict the agent behavior. Nevertheless, many of these approaches falter when confronted with scenarios beyond their training datasets, and lack interpretability, raising concerns about their suitability for real-world applications such as autonomous driving. Moreover, these models frequently demand additional training, substantial computational resources, or specific input features necessitating extensive implementation endeavors. In response, we propose Gaussian Lane Keeping (GLK), a robust prediction method for autonomous vehicles that can provide a solid baseline for comparison when developing new algorithms and a sanity check for real-world deployment. We provide several extensions to the GLK model, evaluate it on the CitySim dataset, and show that it outperforms the neural-network based predictions.