An FPGA Based Energy Efficient DS-SLAM Accelerator for Mobile Robots in Dynamic Environment
Source abstract
Abstract
The Simultaneous Localization and Mapping (SLAM) algorithm is a hotspot in robot application research with the ability to help mobile robots solve the most fundamental problems of “localization” and “mapping”. The visual semantic SLAM algorithm fused with semantic information enables robots to understand the surrounding environment better, thus dealing with complexity and variability of real application scenarios. DS-SLAM (Semantic SLAM towards Dynamic Environment), one of the representative works in visual semantic SLAM, enhances the robustness in the dynamic scene through semantic information. However, the introduction of deep learning increases the complexity of the system, which makes it a considerable challenge to achieve the real-time semantic SLAM system on the low-power embedded platform. In this paper, we realized the high energy-efficiency DS-SLAM algorithm on the Field Programmable Gate Array (FPGA) based heterogeneous platform through the optimization co-design of software and hardware with the help of OpenCL (Open Computing Language) development flow. Compared with Intel i7 CPU on the TUM dataset, our accelerator achieves up to 13× frame rate improvement, and up to 18× energy efficiency improvement, without significant loss in accuracy.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Robotics and Sensor-Based Localization · Advanced Image and Video Retrieval Techniques · Robotic Path Planning Algorithms
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing 100044, China
School of Information Science and Technology, Baotou Teachers’ College, Baotou 014030, China
Department of Electronic Engineering and BNRist, Tsinghua University, Beijing 100084, China
Intel Labs China, Beijing 100090, China
Department of Precision Instrument, Tsinghua University, Beijing 100084, China
Department of Mechanical Engineering, Tsinghua University, Beijing 100084, China
References (25)
Show source-provided references
1. Song, S., Choi, D., Hur, J., Lee, M., Park, Y.J., and Kim, J. (2009, January 18–21). Development and application of Mobile Robot system for Marking Process in LNGC cargo tanks. Proceedings of the 2009 ICCAS-SICE, Fukuoka, Japan.
2. Saitoh, M., Takahashi, Y., Sankaranarayanan, A., Ohmachi, H., and Marukawa, K. (2002, January 21–27). A mobile robot testbed with manipulator for security guard application. Proceedings of the 1995 IEEE International Conference on Robotics and Automation, Nagoya, Japan.
3. Human–Robot Interaction in Rescue Robotics10.1109/TSMCC.2004.826267 ↗
4. Robots for Environmental Monitoring: Significant Advancements and Applications10.1109/MRA.2011.2181683 ↗
5. Bapna, D., Rollins, E., Murphy, J., Maimone, E., Whittaker, W., and Wettergreen, D. (2002, January 20). The Atacama Desert Trek: Outcomes. Proceedings of the 1998 IEEE International Conference on Robotics and Automation (Cat. No. 98CH36146), Leuven, Belgium.
6. Rollins, E., Luntz, J., Foessel, A., Shamah, B., and Whittaker, W. (1998, January 20). Nomad: A Demonstration of the Transforming Chassis. Proceedings of the IEEE International Conference on Robotics and Automation, Leuven, Belgium.
7. Kimon, P.V., and George, J.V. (2015). Classification of UAVs. Handbook of Unmanned Aerial Vehicles, Springer.
8. Mei, Y., Lu, Y.H., and Hu, Y.C. (2005, January 18–20). A case study of mobile robot’s energy consumption and conservation techniques. Proceedings of the International Conference on Advanced Robotics, Seattle, WA, USA.
9. Yu, C., Liu, Z., and Liu, X. (2018, January 1–5). Ds-slam: A semantic visual slam towards dynamic environments. Proceedings of the 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Madrid, Spain.10.1109/IROS.2018.8593691 ↗
10. SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation10.1109/TPAMI.2016.2644615 ↗
11. Orb-slam2: An open-source slam system for monocular, stereo and rgb-d cameras10.1109/TRO.2017.2705103 ↗
12. Zhang, C., Li, P., and Sun, G. (2015, January 22–24). Optimizing FPGA-based Accelerator Design for Deep Convolutional Neural Networks. Proceedings of the ACM/SIGDA International Symposium on Field-Programmable Gate Arrays, Monterey, CA, USA.10.1145/2684746.2689060 ↗
13. Qiu, J., Wang, J., and Yao, S. (2016, January 21–23). Going Deeper with Embedded FPGA Platform for Convolutional Neural Network. Proceedings of the ACM/SIGDA International Symposium on Field-Programmable Gate Arrays, Monterey, CA, USA.10.1145/2847263.2847265 ↗
14. Khronos Group (2020, December 10). OpenCL-The Open Standard for Parallel Programming of Heterogeneous Systems. Available online: http://www.khronos.org/opencl.
15. Shi, X., Cao, L., and Wang, D. (2018, January 1–5). HERO: Accelerating Autonomous Robotic Tasks with FPGA. Proceedings of the 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Madrid, Spain.10.1109/IROS.2018.8593522 ↗
16. FPGA design of EKF block accelerator for 3D visual SLAM10.1016/j.compeleceng.2016.05.003 ↗
17. Boikos, K., and Bouganis, C.-S. (September, January 29). Semi-dense SLAM on an FPGA SoC. Proceedings of the 2016 26th International Conference on Field Programmable Logic and Applications (FPL), Lausanne, Switzerland.10.1109/FPL.2016.7577365 ↗
18. Embedding SLAM algorithms: Has it come of age?10.1016/j.robot.2017.10.019 ↗
19. Liu, R., and Yang, J. (2019). Eslam: An energy-efficient accelerator for realtime orb-slam on fpga platform. DAC, ACM.10.1145/3316781.3317820 ↗
20. Han, S., Mao, H., and Dally, W.J. (2015). Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding. arXiv.
21. The pascal visual object classes challenge: A retrospective10.1007/s11263-014-0733-5 ↗
22. Intel Corporation (2020, December 10). Intel Arria 10 Device Overview. Available online: https://www.intel.com/content/www/us/en/programmable/documentation/sam1403480274650.html.
23. Yu, M., Huang, H., Liu, H., He, S., Qiao, F., Luo, L., Xie, F., Liu, X.-J., and Yang, H. (August, January 29). Optimizing FPGA-based Convolutional Encoder-Decoder Architecture for Semantic Segmentation. Proceedings of the 2019 IEEE 9th Annual International Conference on CYBER Technology in Automation, Control, and Intelligent Systems (CYBER), Guangzhou, China.10.1109/CYBER46603.2019.9066759 ↗
24. Huang, H. (2020). EDSSA: An Encoder-Decoder Semantic Segmentation Networks Accelerator on OpenCL-Based FPGA Platform. Sensors, 20.10.3390/s20143969 ↗
25. Sturm, J., Engelhard, N., and Endres, F. (2012, January 7–12). A benchmark for the evaluation of rgb-d slam systems. Proceedings of the 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, Vilamoura-Algarve, Portugal.10.1109/IROS.2012.6385773 ↗