WU Linfeng,WANG Lutao.Graph-based SLAM Embedded Processing Technology[J].Journal of Chengdu University of Information Technology,2019,(02):118-122.[doi:10.16836/j.cnki.jcuit.2019.02.003]
图优化SLAM的嵌入式处理技术
- Title:
- Graph-based SLAM Embedded Processing Technology
- 文章编号:
- 2096-1618(2019)02-0118-05
- Keywords:
- simultaneous localization and mapping; graph optimization; GPU parallel computing; embedded systems
- 分类号:
- TP391
- 文献标志码:
- A
- 摘要:
- 基于图优化的同时定位与地图构建(simultaneous localization and mapping,SLAM),可以让移动机器人在未知环境下构建周围的地图同时确定机器人在地图中的位置。复杂环境下其精确度和计算效率均优于基于滤波的SLAM,现已成为主流的方法。图优化SLAM用图来表示和解决SLAM问题,对计算资源要求很高,这极大限制了其在嵌入式系统上实时应用。对此提出了一种基于NVIDIA TX2 的图优化SLAM处理技术,根据其架构特点优化SLAM算法,加速处理,实现图优化的实时处理。实验证明,通过高性能嵌入式处理硬件与算法优化的有机结合,可有效提升图优化SLAM在嵌入式系统中的处理性能。
- Abstract:
- Graph-based simultaneous localization and mapping allows the mobile robot to construct the surrounding map in an unknown environment while determining the robot's position in the map. The accuracy and computational efficiency of Graph-based SLAM are better than those of the filter-based SLAM in complex environment, which has become the mainstream method. Graph-based SLAM uses the graph to represent and solve the SLAM problem, which requires high computational resources and greatly limits its application in real-time on embedded systems. This paper proposes an optimized SLAM processing technology based on NVIDIA TX2 and optimizes the SLAM algorithm according to its architecture characteristics to speed up processing and achieve real-time processing of graph optimization. The experiments have proved that the organic combination of high-performance embedded processing hardware and algorithm optimization can effectively improve the processing performance of Graph-based SLAM in embedded systems.
参考文献/References:
[1] Durrant-Whyte H,Bailey T.Simultaneous Localization and Mapping:Part I[J].IEEE Robotics & Amp Amp Automation Magazine,2017,13(2):99-110.
[2] Grisetti G,Kummerle R,Stachniss C,et al.A Tutorial on Graph-Based SLAM[J].IEEE Intelligent Transportation Systems Magazine,2011,2(4):31-43.
[3] Carlevaris-Bianco N,Eustice R M.Generic factor-based node marginalization and edge sparsification for pose-graph SLAM[C].IEEE International Conference on Robotics and Automation.IEEE,2013:5748-5755.
[4] Klein G,Murray D.Parallel Tracking and Mapping for Small AR Workspaces[C].IEEE and ACM International Symposium on Mixed and Augmented Reality.IEEE Computer Society,2007:1-10.
[5] Lee S.Embedded Visual SLAM:Applications for Low-Cost Consumer Robots[J].IEEE Robotics & Automation Magazine,2013,20(4):83-95.
[6] Konolige K,Grisetti G,Kümmerle R,et al.Efficient Sparse Pose Adjustment for 2D mapping[C].Ieee/rsj International Conference on Intelligent Robots and Systems.IEEE,2010:22-29.
[7] Khosoussi K,Sukhatme G S,Huang S,et al.Designing Sparse Reliable Pose-Graph SLAM:A Graph-Theoretic Approach[J].2016.
[8] Kretzschmar H,Stachniss C.Information-theoretic compression of pose graphs for laser-based SLAM[M].Sage Publications,Inc.2012.
[9] Zhao L,Huang S,Dissanayake G.Linear SLAM:A linear solution to the feature-based and pose graph SLAM based on submap joining[C].Ieee/rsj International Conference on Intelligent Robots and Systems.IEEE,2013:24-30.
[10] Whelan T,Leutenegger S,Moreno R S,et al.ElasticFusion:Dense SLAM Without A Pose Graph[C].Robotics:Science and Systems.2015.
[11] Wu C,Agarwal S,Curless B,et al.Multicore bundle adjustment[J].2016:3057-3064.
[12] Rodriguez-Losada D,Segundo P S,Hernando M,et al.GPU-Mapping: Robotic Map Building with Graphical Multiprocessors[J].IEEE Robotics & Automation Magazine,2013,20(2):40-51.
[13] Fraundorfer F,Scaramuzza D.Visual Odometry:Part II:Matching,Robustness,Optimization,and Applications[J].IEEE Robotics & Automation Magazine,2012,19(2):78-90.
[14] Paz L M,Jensfelt P,Tardos J D,et al.EKF SLAM updates in O(n)with Divide and Conquer SLAM[C].IEEE International Conference on Robotics and Automation.IEEE,2007:1657-1663.
[15] Huang S,Dissanayake G.Convergence and Consistency Analysis for Extended Kalman Filter Based SLAM[J].IEEE Transactions on Robotics,2007,23(5):1036-1049.
[16] Polok L,Ila V,Solony M,et al.Incremental Block Cholesky Factorization for Nonlinear Least Squares in Robotics[C].Robotics:Science and Systems.2013.
[17] Sibley G,Matthies L,Sukhatme G.Sliding window filter with application to planetary landing[J].Journal of Field Robotics,2010,27(5):587-608.
[18] Leutenegger S,Furgale P,Rabaud V,et al.Keyframe-Based Visual-Inertial SLAM using Nonlinear Optimization[C].Robotics:Science and Systems.2013:789-795.
[19] Bell N.Implementing sparse matrix-vector multiplication on throughput-oriented processors[C].Conference on High PERFORMANCE Computing Networking,Storage and Analysis.ACM,2009:18.
[20] Dine A,Elouardi A,Vincke B,et al.Graph-based SLAM embedded implementation on low-cost architectures:A practical approach[C].IEEE International Conference on Robotics and Automation.IEEE,2015:4612-4619.
[21] Otterness N,Yang M,Rust S,et al.An Evaluation of the NVIDIA TX1 for Supporting Real-Time Computer-Vision Workloads[C].Real-Time and Embedded Technology and Applications Symposium.IEEE,2017:353-364.
[22] Kümmerle R,Grisetti G,Strasdat H,et al.G2O:A general framework for graph optimization[C].IEEE International Conference on Robotics and Automation.IEEE,2011:3607-3613.
相似文献/References:
[1]纪嘉文,杨明欣.一种基于多传感融合的室内建图和定位算法[J].成都信息工程大学学报,2018,(04):400.[doi:10.16836/j.cnki.jcuit.2018.04.009]
JI Jia-wen,YANG Ming-xin.A Indoor Mapping and Localization Algorithm based on Multi-sensor Fusion[J].Journal of Chengdu University of Information Technology,2018,(02):400.[doi:10.16836/j.cnki.jcuit.2018.04.009]
备注/Memo
收稿日期:2018-06-28 基金项目:四川省科技厅资助项目(2018GZ0184)