A Hierarchical Temporal Memory Based End-to-End Autonomous Driving System

Luc Le Mero, M. Dianati, Graham Lee
{"title":"A Hierarchical Temporal Memory Based End-to-End Autonomous Driving System","authors":"Luc Le Mero, M. Dianati, Graham Lee","doi":"10.1115/1.4064989","DOIUrl":null,"url":null,"abstract":"\n Achieving human-level driving performance in complex environments remains a major challenge in the field of Deep Learning (DL) based end-to-end Autonomous Driving Systems (ADS). In ADS, generalization to rare edge cases poses a serious safety concern with DL based models. The leading solution to this problem is scaling; the construction of larger models and datasets. However, limitations in the computational power available to autonomous vehicles, coupled with the under-representation of safety-critical edge cases in large autonomous driving datasets raise questions over the suitability of scaling for ADS. In this work, we investigate the performance of an alternate, computationally less demanding, Machine Learning (ML) algorithm, Hierarchical Temporal Memory (HTM). Existing HTM models use rudimentary encoding schemes that have thus far limited their application to simple inputs. Motivated by this shortcoming, we first propose a bespoke CNN based encoding scheme suited to the input data used in ADS. We then integrate this encoding scheme into a novel DL-HTM end-to-end ADS. The proposed DL-HTM based end-to-end ADS is trained and evaluated against a conventional DL end-to-end ADS based on the widely used AlexNet model from the literature. Our evaluation results show that the proposed DL-HTM model achieves comparable performance with far fewer trainable parameters than the conventional DL based end-to-end ADS. Results also indicate that the proposed model demonstrates a superior capacity for learning under-represented classes, i.e. edge cases, in the dataset.","PeriodicalId":164923,"journal":{"name":"Journal of Autonomous Vehicles and Systems","volume":"18 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Autonomous Vehicles and Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1115/1.4064989","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Achieving human-level driving performance in complex environments remains a major challenge in the field of Deep Learning (DL) based end-to-end Autonomous Driving Systems (ADS). In ADS, generalization to rare edge cases poses a serious safety concern with DL based models. The leading solution to this problem is scaling; the construction of larger models and datasets. However, limitations in the computational power available to autonomous vehicles, coupled with the under-representation of safety-critical edge cases in large autonomous driving datasets raise questions over the suitability of scaling for ADS. In this work, we investigate the performance of an alternate, computationally less demanding, Machine Learning (ML) algorithm, Hierarchical Temporal Memory (HTM). Existing HTM models use rudimentary encoding schemes that have thus far limited their application to simple inputs. Motivated by this shortcoming, we first propose a bespoke CNN based encoding scheme suited to the input data used in ADS. We then integrate this encoding scheme into a novel DL-HTM end-to-end ADS. The proposed DL-HTM based end-to-end ADS is trained and evaluated against a conventional DL end-to-end ADS based on the widely used AlexNet model from the literature. Our evaluation results show that the proposed DL-HTM model achieves comparable performance with far fewer trainable parameters than the conventional DL based end-to-end ADS. Results also indicate that the proposed model demonstrates a superior capacity for learning under-represented classes, i.e. edge cases, in the dataset.
基于分层时态记忆的端到端自动驾驶系统
在复杂环境中实现人类水平的驾驶性能,仍然是基于深度学习(DL)的端到端自动驾驶系统(ADS)领域的一大挑战。在自动驾驶系统中,基于深度学习的模型对罕见边缘情况的泛化是一个严重的安全问题。这一问题的主要解决方案是扩展,即构建更大的模型和数据集。然而,由于自动驾驶车辆的计算能力有限,再加上大型自动驾驶数据集中的安全关键边缘案例代表性不足,人们对自动驾驶辅助系统的扩展是否合适提出了质疑。在这项工作中,我们研究了另一种计算要求较低的机器学习(ML)算法--分层时态记忆(HTM)的性能。现有的 HTM 模型使用的是初级编码方案,迄今为止,其应用仅限于简单输入。鉴于这一缺陷,我们首先提出了一种基于 CNN 的定制编码方案,适合 ADS 中使用的输入数据。然后,我们将这一编码方案集成到新颖的 DL-HTM 端到端 ADS 中。我们将基于 DL-HTM 的端到端 ADS 与基于文献中广泛使用的 AlexNet 模型的传统 DL 端到端 ADS 进行了对比训练和评估。我们的评估结果表明,与传统的基于 DL 的端到端 ADS 相比,所提出的 DL-HTM 模型只需较少的可训练参数就能达到相当的性能。结果还表明,所提出的模型在学习数据集中代表性不足的类别(即边缘案例)方面表现出卓越的能力。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信