Recurrent-OctoMap: Learning State-based Map Refinement for Long-Term Semantic Mapping with 3D-Lidar Data

Sun, Li and Yan, Zhi and Zaganidis, Anestis and Zhao, Cheng and Duckett, Tom (0018) Recurrent-OctoMap: Learning State-based Map Refinement for Long-Term Semantic Mapping with 3D-Lidar Data. IEEE Robotics and Automation Letters . ISSN 2377-3774

Documents
Recurrent-OctoMap: Learning State-based Map Refinement for Long-Term Semantic Mapping with 3D-Lidar Data

Request a copy
[img] PDF
recurrent-octomap-learning-final.pdf - Whole Document
Restricted to Repository staff only

2MB
Item Type:Article
Item Status:Live Archive

Abstract

This paper presents a novel semantic mapping approach, Recurrent-OctoMap, learned from long-term 3D
Lidar data. Most existing semantic mapping approaches focus on improving semantic understanding of single frames, rather than 3D refinement of semantic maps (i.e. fusing semantic observations). The most widely-used approach for 3D semantic map refinement is a Bayes update, which fuses the consecutive predictive probabilities following a Markov-Chain model. Instead, we propose a learning approach to fuse the semantic features, rather than simply fusing predictions from a classifier. In our approach, we represent and maintain our 3D map as an OctoMap, and model each cell as a recurrent neural network (RNN), to obtain a Recurrent-OctoMap. In this case, the semantic mapping process can be formulated as a sequenceto-sequence encoding-decoding problem. Moreover, in order to extend the duration of observations in our Recurrent-OctoMap, we developed a robust 3D localization and mapping system for successively mapping a dynamic environment using more than two weeks of data, and the system can be trained and deployed with arbitrary memory length. We validate our approach on the ETH long-term 3D Lidar dataset [1]. The experimental results show that our proposed approach outperforms the conventional “Bayes update” approach.

Keywords:semantic mapping, long-term mapping, lidar perception, 3D semantic map refinement
Subjects:H Engineering > H671 Robotics
Divisions:College of Science > School of Computer Science
ID Code:32558
Deposited On:05 Jul 2018 15:01

Repository Staff Only: item control page