Spatial Uncertainty Sampling for End-to-End Control
Abstract
End-to-end trained neural networks (NNs) are a compelling approach to autonomous vehicle control because of their ability to learn complex tasks without manual engineering of rule-based decisions. However, challenging road conditions, ambiguous navigation situations, and safety considerations require reliable uncertainty estimation for the eventual adoption of full-scale autonomous vehicles. Bayesian deep learning approaches provide a way to estimate uncertainty by approximating the posterior distribution of weights given a set of training data. Dropout training in deep NNs approximates Bayesian inference in a deep Gaussian process and can thus be used to estimate model uncertainty. In this paper, we propose a Bayesian NN for end-to-end control that estimates uncertainty by exploiting feature map correlation during training. This approach achieves improved model fits, as well as tighter uncertainty estimates, than traditional element-wise dropout. We evaluate our algorithms on a challenging dataset collected over many different road types, times of day, and weather conditions, and demonstrate how uncertainties can be used in conjunction with a human controller in a parallel autonomous setting.
Cite
@article{arxiv.1805.04829,
title = {Spatial Uncertainty Sampling for End-to-End Control},
author = {Alexander Amini and Ava Soleimany and Sertac Karaman and Daniela Rus},
journal= {arXiv preprint arXiv:1805.04829},
year = {2021}
}
Comments
Originally published in Neural Information Processing Systems (NIPS) Workshop on Bayesian Deep Learning 2017