Traversability estimation in off-road environments requires a robust perception system. Recently, approaches to learning a traversability estimation from past vehicle experiences in a self-supervised manner are arising as they can greatly reduce human labeling costs and labeling errors. Nonetheless, the learning setting from self-supervised traversability estimation suffers from congenital uncertainties that appear according to the scarcity of negative information. Negative data are rarely harvested as the system can be severely damaged while logging the data. To mitigate the uncertainty, we introduce a method to incorporate unlabeled data in order to leverage the uncertainty. First, we design a learning architecture that inputs query and support data. Second, unlabeled data are assigned based on the proximity in the metric space. Third, a new metric for uncertainty measures is introduced. We evaluated our approach on our own dataset, `Dtrail', which is composed of a wide variety of negative data.
翻译:越野环境中的可变性估计要求有一个强大的认知系统。 最近,以自我监督的方式从以往车辆经验中学习可穿行性估计的方法正在形成,因为它们可以大大减少人类标签成本和标签错误。然而,自监督的可穿行性估计的学习环境仍然有先天性的不确定性,这种不确定性视消极信息的匮乏而出现。由于系统在记录数据时可能受到严重破坏,因此很少收集负面数据。为了减轻不确定性,我们采用了一种方法,将未贴标签的数据纳入其中,以利用不确定性。首先,我们设计了一个输入查询和支持数据的学习结构。第二,根据公制空间的近距离,指定了未贴标签的数据。第三,采用了新的不确定性计量标准。我们评估了我们自己数据集“缺陷”的方法,该数据集由各种负面数据组成。