arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2203.02179cs.LG
收藏

电动汽车驾驶员行为的被动学习与主动学习

Passive and Active Learning of Driver Behavior from Electric Vehicles

University of Gothenburg(哥德堡大学);Chalmers University of Technology(查尔姆斯理工大学);Volvo Car Corporation(沃尔沃汽车集团)

Federica Comuni, Christopher Mészáros, Niklas Åkerblom, Morteza Haghir Chehreghani

AI总结:

本研究针对电动汽车驾驶员行为建模,比较非递归与递归模型在能耗预测中的表现,并探索多种主动学习方法以减少标注成本,发现主动采样可优于随机采样。

摘要:

对驾驶员行为进行建模在汽车行业中具有多项优势,包括预测电动汽车的能耗。研究表明,在某些驾驶场景下,激进驾驶比温和驾驶多消耗高达30%的能量。机器学习方法被广泛用于驾驶员行为分类,然而,这可能会带来一些挑战,例如长时间窗口上的序列建模,以及由于标注成本高昂而导致的标记数据缺乏。为解决第一个挑战,即驾驶员行为的被动学习,我们研究了非递归架构,如自注意力模型和结合联合递归图(JRP)的卷积神经网络,并将其与递归模型进行比较。我们发现自注意力模型取得了良好的性能,而JRP并未表现出任何显著改进。然而,在我们研究中使用的5秒和10秒窗口长度下,没有一种非递归模型优于递归模型。为解决第二个挑战,我们研究了多种具有不同信息量度量的主动学习方法。我们评估了不确定性采样,以及更先进的方法,如查询委员会和主动深度丢弃。我们的实验表明,一些主动采样技术可以优于随机采样,从而减少标注所需的工作量。

英文摘要:

Modeling driver behavior provides several advantages in the automotive industry, including prediction of electric vehicle energy consumption. Studies have shown that aggressive driving can consume up to 30% more energy than moderate driving, in certain driving scenarios. Machine learning methods are widely used for driver behavior classification, which, however, may yield some challenges such as sequence modeling on long time windows and lack of labeled data due to expensive annotation. To address the first challenge, passive learning of driver behavior, we investigate non-recurrent architectures such as self-attention models and convolutional neural networks with joint recurrence plots (JRP), and compare them with recurrent models. We find that self-attention models yield good performance, while JRP does not exhibit any significant improvement. However, with the window lengths of 5 and 10 seconds used in our study, none of the non-recurrent models outperform the recurrent models. To address the second challenge, we investigate several active learning methods with different informativeness measures. We evaluate uncertainty sampling, as well as more advanced methods, such as query by committee and active deep dropout. Our experiments demonstrate that some active sampling techniques can outperform random sampling, and therefore decrease the effort needed for annotation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑