COSEM: Collaborative Semantic Map Matching Framework for Autonomous Robots
- 14 April 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 69 (4) , 3843-3853
Abstract
Relative localization is a fundamental requirement for the coordination of multiple robots. To date, existing research in relative localization mainly depends on the extraction of low-level geometry features such as planes, lines, and points, which may fail in challenging cases when the initial error is large and the overlapping area is low. In this article, a novel approach named collaborative semantic map matching (COSEM) is proposed to estimate the relative transformation between robots. COSEM jointly performs multimodal information fusion, semantic data association, and optimization in a unified framework. First, each robot applies a multimodal information fusion model to generate local semantic maps. Since the correspondences between local maps are latent variables, a flexible semantic data association strategy is proposed using expectation-maximization. Instead of assigning hard geometry data association, semantic association and geometry association are jointly estimated. Then, the minimization of the expected cost results in a rigid transformation matrix between two semantic maps. Evaluations on semantic KITTI benchmarks and real-world experiments show the improved accuracy, convergence, and robustness.Keywords
Funding Information
- National Natural Science Foundation of China (62003039)
- Beijing Institute of Technology Research Fund Program for Young Scholars
- Young Elite Scientists Sponsorship Program
- CAST (2020QNRC001)
This publication has 41 references indexed in Scilit:
- Multiple-Robot Simultaneous Localization and Mapping: A ReviewJournal of Field Robotics, 2015
- Initialization techniques for 3D SLAM: A survey on rotation estimation and its use in pose graph optimization2015
- Robust and Efficient Multirobot 3-D Mapping Merging With Octree-Based Occupancy GridsIEEE Systems Journal, 2015
- Active Persistent Localization of a Three-Dimensional Moving Target Under Set-Membership Uncertainty Description Through Cooperation of Multiple Mobile RobotsIEEE Transactions on Industrial Electronics, 2015
- LOAM: Lidar Odometry and Mapping in Real-time2014
- Group Mapping: A Topological Approach to Map Merging for Multiple RobotsIEEE Robotics & Automation Magazine, 2014
- Normal distributions transform Monte-Carlo localization (NDT-MCL)2013
- OctoMap: an efficient probabilistic 3D mapping framework based on octreesAutonomous Robots, 2013
- Expectation-Maximization Algorithm2012
- A method for registration of 3-D shapesIEEE Transactions on Pattern Analysis and Machine Intelligence, 1992