Abstract
Simultaneous Localization And Mapping (SLAM) algorithm allows the robot to map the environment while locating it in the space. SLAM algorithm is the more efficient and more accredited system by autonomous vehicle navigation and robotic application in current research. However, it has not yet adopted a complete end-to-end hardware implementation. Our work is aimed at hardware/software optimization of a time-consuming and expensive functional block of a SLAM-based autonomous navigation application, precisely monocular ORB-SLAM. Thus, the proposed optimization is achieved through the implementation of a heterogeneous FPGA-based embedded architecture. A comparative study is then conducted with parallel architectures such as automotive embedded architecture TX1 and a high-end machine. This work shows that heterogeneous FPGA-based embedded architectures are appealing embedded platforms for the heavy computational algorithm, achieving a 16% improvement of the original, whereas high-end machines achieve 45% improvement.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Thrun S (2007) Simultaneous localization and mapping. Springer, Berlin, Heidelberg
National instrument, introduction to fpga technology. https://www.ni.com/en-lb/innovations/white-papers/08/fpga-fundamentals.html
Cong J, Liu B, Neuendorffer S, Noguera J, Vissers K, Zhang Z (2011) High-level synthesis for FPGAs: From prototyping to deployment. IEEE Trans Comput-Aided Des Integr Circuits Syst 30(4):473–491. https://doi.org/10.1109/TCAD.2011.2110592
Xu JY (2008) OpenCL – the open standard for parallel programming of heterogeneous systems
Li H, Yu D, Kumar A, Tu YC (2014) Performance modeling in CUDA streams—a means for high-throughput data processing. In 2014 IEEE international conference on big data (big data), pp 301–310
Mur-Artal R, Montiel JMM, Tardos JD (2015) ORB-SLAM: a versatile and accurate monocular SLAM system. IEEE Trans Robot 31(5):1147–1163
Campos C, Elvira R, RodrÃguez JJG, Montiel JM, Tardós JD (2020) ORB-SLAM3: an accurate open-source library for visual, visual-inertial and multi-map SLAM. arXiv preprint arXiv:2007.11898
Scaramuzza D, Fraundorfer F (2011) Visual odometry [tutorial]. IEEE Robot Autom Mag 18(4):80–92
Nguyen D-D, Elouardi A, Florez SAR, Bouaziz S (2018) HOOFR SLAM system: an embedded vision SLAM algorithm and its hardware-software mapping-based intelligent vehicles applications. IEEE Trans Intell Trans Syst 20(11):4103–4118
Mur-Artal R, Tardos JD (2017) ORB-SLAM2: An open-source SLAM system for monocular, stereo, and rgb-d cameras. IEEE Transac Robotics 33(5):1255–1262. https://doi.org/10.1109/TRO.2017.2705103
Li J, Deng G, Zhang W, Zhang C, Wang F, Liu Y (2020) Realization of CUDA-based real-time multi-camera visual SLAM in embedded systems. J Real-Time Image Process 17(3):713–727. https://doi.org/10.1007/s11554-019-00924-4
Aldegheri S, Bombieri N, Bloisi DD, et al (2019) Data flow orb-slam for real-time performance on embedded gpu boards. In 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp 5370–5375. IEEE
Bourque D (2017) Cuda-accelerated orb-slam for uavs
Abouzahir M, Elouardi A, Latif R, Bouaziz S, Tajer A (2018) Embedding SLAM algorithms: has it come of age? Robot Auton Syst 100:14–26
Rosten E, Porter R, Drummond T (2010) Faster and better: A machine learning approach to corner detection. IEEE Trans Pattern Anal Mach Intell 32(1):105–119. https://doi.org/10.1109/TPAMI.2008.275
Shi X, Cao L, Wang D, Liu L, You G, Liu S, Wang C (2018) HERO: accelerating autonomous robotic tasks with FPGA. In 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, pp 7766–7772
Kohlbrecher S, von Stryk O, Meyer J, Klingauf U (2011) A flexible and scalable SLAM system with full 3D motion estimation. In 2011 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), pp 155–160
Triggs B, McLauchlan PF, Hartley RI, Fitzgibbon AW (2000) Bundle adjustment a modern synthesis. In Vision Algorithms: Theory and Practice. New York, NY, USA, pp 298–372. Springer
TUM1 dataset, https://vision.in.tum.de/data/datasets/rgbd-dataset/download
Hartley R, Zisserman A (2003) Multiple view geometry in computer vision. 2nd ed., Cambridge, U.K.: Cambridge Univ. Press
Altera sdk for OpenCL, Best Practices Guide. https://www.intel.com/content/dam/www/programmable/us/en/pdfs/literature/hb/opencl-sdk/aocl-best-practices-guide.pdf
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Mamri, A., Abouzahir, M., Ramzi, M., Sbihi, M. (2021). High-Level Synthesis Implementation of Monocular SLAM on Low-Cost Parallel Platforms. In: Motahhir, S., Bossoufi, B. (eds) Digital Technologies and Applications. ICDTA 2021. Lecture Notes in Networks and Systems, vol 211. Springer, Cham. https://doi.org/10.1007/978-3-030-73882-2_37
Download citation
DOI: https://doi.org/10.1007/978-3-030-73882-2_37
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-73881-5
Online ISBN: 978-3-030-73882-2
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)