WANG Wenzhao1,2,3， TANG Jun1,2,3， SHEN Chong1,2,3， LIU Jun1,2,3
（1. School of Instrument and Electronics, North University of China, Taiyuan 030051, China； 2. National Key Laboratory of Dynamic Testing Technology, North University of China, Taiyuan 030051, China； 3. Shanxi Province Key Laboratory of Quantum Sensing and Precision Measurement, North University of China, Taiyuan 030051, China）
Abstract： Brain-inspired navigation algorithm controlled by simultaneous localization and mapping (SLAM) is prone to errors, primarily caused by complex environmental factors such as changes in light direction. To address this limitation, a brain-inspired SLAM approach is proposed where feature matching is supported by the speeded up robust features (SURF) algorithm. This model gathers environmental information via a set of mobile vision systems: the local view cell operators, designed to retrieve information about carrier direction/location via the SURF algorithm, and the head-direction cell and pose cell operators, which simultaneously represents current carrier position via continuous attractor neural networks. The position and time information retrieved by these cell operators are used to compute the position of the current carrier through path integration. In the final step, experience maps of the topology are constructed based on cognitive points. In addition, while the local view cell acquire environmental information, closed-loop detections are executed by the SURF algorithm to correct (if necessary) the current position. The optimized brain-inspired SLAM model successfully addresses the problem of scene matching errors faced by previous models in the presence of changing light direction. The effectiveness of the proposed method is validated through experimental verification.
Key words： brain-inspired navigation; simultaneous localization and mapping (SLAM); speeded up robust features (SURF) algorithm; experience map; closed-loop detection
References
［1］CADENA C, CARLONE L, CARRILLO H, et al. Past, present, and future of simultaneous localization and mapping: roward the robust-perception age. IEEE Transactions on Robotics, 2016, 32(6): 1309-1332.
［2］SHEN C, XIONG Y F, ZHAO D H, et al. Multi-rate strong tracking square-root cubature Kalman filter for MEMS-INS/GPS/polarization compass integrated navigation system. Mechanical Systems and Signal Processing, 2022, 163: 108146.
［3］MILFORD M, SCHULZ R. Principles of goal-directed spatial robot navigation in biomimetic models. Philosophical Transactions of the Royal Society B Biological Sciences, 2014, 369(1655): 20130484.
［4］FINKELSTEIN A, LAS L, ULANOVSKY N. 3D maps and compasses in the brain. Annual Review of Neuroscience, 2016, 39: 171-196.
［5］CAMPBELL M G, OCKO S A, MALLORY C S, et al. Principles governing the integration of landmark and self-motion cues in entorhinal cortical codes for navigation. Nature Neuroscience, 2018, 21(8): 1096-1106.
［6］JEFFERY K J, JOVALEKIC A, VERRIOTIS M, et al. Navigating in a three-dimensional world. Behavioral and Brain Sciences, 2013, 36(5): 523-543.
［7］JEFFERY K J, PAGE H J I, STRINGER S M. Optimal cue combination and landmark-stability learning in the head direction system. The Journal of Physiology, 2016, 594(22): 6527-6534.
［8］EVANS T, BICANSKI A, BUSH D, et al. How environment and self-motion combine in neural representations of space. The Journal of Physiology, 2016, 594(22): 6535-6546.
［9］COPE ALEX J, SABO C, VASILAKI E, et al. A computational model of the integration of landmarks and motion in the insect central complex. 2017, 12(2): e0172325.
［10］BJERKNES T L, DAGSLOTT N C, MOSER E I, et al. Path integration in place cells of developing rats. Proceedings of the National Academy of Sciences of the United States of America, 2018, 115(7): E1637-E1646.
［11］O’KEEFE J, DOSTROVSKY J. The hippocampus as a spatial map: Preliminary evidence from unit activity in the freely-moving rat. Brain Research, 1971, 34(1): 171-175.
［12］TAUBE J S, MULLER R U, RANCK J B. Head-direction cells recorded from the postsubiculum in freely moving rats. I. Description and quantitative analysis. Journal of Neuroscience, 1990, 10(2): 420-435.
［13］HAFTING T, FYHN M, MOLDEN S, et al. Microstructure of a spatial map in the entorhinal cortex. Nature, 2005, 436(7052): 801-806.
［14］KROPFF E, CARMICHAEL J E, MOSER M B, et al. Speed cells in the medial entorhinal cortex. Nature, 2015, 523: 419-424.
［15］LEVER C, BURTON S, JEEWAJEE A, et al. Boundary vector cells in the subiculum of the hippocampal formation. Journal of Neuroscience, 2009, 29(31): 9771-9777.
［16］MCNAUGHTON B L, BATTAGLIA F P, JENSEN O, et al. Path integration and the neural basis of the ‘cognitive map’. Nature Reviews Neuroscience, 2006, 7(8): 663-678.
［17］MOSER E I, KROPFF E, MOSER M B. Place cells, grid cells, and the brain’s spatial representation system. Annual Review of Neuroscience, 2008, 31: 69-89.
［18］MOSER M B, ROWLAND D C, MOSER E I. Place cells, grid cells, and memory. Cold Spring Harbor Perspectives in Biology, 2015, 7(2): a021808.
［19］MILFORD M J, WYETH G, PRASSER D. RatSLAM: a hippocampal model for simultaneous localization and mapping//IEEE International Conference on Robotics and Automation, Apr. 26-May 1, New Orleans, LA, USA. New York: IEEE, 2004: 403-408.
［20］MILFORD M J, WYETH G F. Mapping a suburb with a single camera using a biologically inspired SLAM system. IEEE Transactions on Robotics, 2008, 24(5): 1038-1053.
［21］MILFORD M J, WYETH G F. Persistent navigation and mapping using a biologically inspired SLAM system. The International Journal of Robotics Research, 2010, 29(9): 1131-1153.
［22］STECKEL J, PEREMANS H. BatSLAM: simultaneous localization and mapping using biomimetic sonar. PLOS ONE, 2013, 8(1): e54076.
［23］SILVEIRA L, GUTH F, DREWS P, et al. 3D robotic mapping: a biologic approach//16th International Conference on Advanced Robotics, Nov. 25-29, 2013, Montevideo, Uruguay. New York: IEEE, 2013: 1-6.
［24］SILVEIRA L, GUTH F, DREWS-JR P, et al. An open-source bio-inspired solution to underwater SLAM. IFAC-PapersOnLine, 2015, 48(2): 212-217.
［25］YU F, SHANG J, HU Y, et al. NeuroSLAM: a brain-inspired SLAM system for 3D environments. Biological Cybernetics, 2019, 113(5-6): 515-545.
［26］BANINO A, BARRY C, URIA B, et al. Vector-based navigation using grid-like representations in artificial agents. Nature, 2018, 557: 429-433.
［27］TANG H, YAN R, TAN K C. Cognitive navigation by neuro-inspired localization, mapping, and episodic memory. IEEE Transactions on Cognitive and Developmental Systems, 2018, 10(3): 751-761.
［28］SHEN C, ZHANG Y, TANG J, et al. Dual-optimization for a MEMS-INS/GPS system during GPS outages based on the cubature Kalman filter and neural networks. Mechanical Systems and Signal Processing, 2019, 133: 106222.
［29］YAO Y, XU X, ZHANG T, et al. An improved initial alignment method for SINS/GPS integration with vectors subtraction. IEEE Sensors Journal, 2021, 21(16): 18256-18262.
［30］BERKVENS R, WEYN M, PEREMANS H. Asynchronous, electromagnetic sensor fusion in RatSLAM//IEEE Sensors, Nov. 1-4, 2015, Busan, Korea (South). New York: IEEE, 2015: 1-4.
［31］ZHAO D H, LIU X C, ZHAO H J, et al. Seamless integration of polarization compass and inertial navigation data with a self-learning multi-rate residual correction algorithm. Measurement, 2021, 170: 108694.
［32］ZHAO D H, LIU Y Z, WU X D, et al. Attitude-induced error modeling and compensation with GRU networks for the polarization compass during UAV orientation. Measurement, 2022, 190: 110734.
［33］MUR-ARTAL R, MONTIEL J M M, TARDOS J D. ORB-SLAM: a versatile and accurate monocular SLAM system. IEEE Transactions on Robotics, 2015, 31(5): 1147-1163.
［34］CIESLEWSKI T, CHOUDHARY S, SCARAMUZZA D. Data-efficient decentralized visual SLAM//IEEE International Conference on Robotics and Automation (ICRA), May 21-25, 2018, Brisbane, Australia. New York: IEEE, 2018: 2466-2473.
一种基于快速鲁棒特征匹配算法的类脑SLAM导航
王文照1,2,3， 唐军1,2,3， 申冲1,2,3， 刘俊1,2,3
（1. 中北大学 仪器与电子学院， 山西 太原 030051；2. 中北大学 动态测试技术国家重点实验室， 山西 太原 030051； 3. 中北大学 山西省量子传感与精密测量重点实验室， 山西 太原 030051）
摘要：类脑导航是模拟鼠类感知环境机制提出的一种同步定位与构图（Simultaneous localization and mapping, SLAM）的导航算法。 针对复杂环境如室内光线变化导致类脑SLAM导航产生误差的问题， 本文提出了基于特征匹配（Speeded up robust features, SURF）算法的优化类脑SLAM导航模型。 该模型通过一套移动视觉系统采集环境信息， 构建的局部场景细胞通过SURF特征匹配算法获取到载体在环境中的方向与位置信息； 头朝向细胞与位置细胞通过连续吸引子神经网络共同表示载体当前的位姿。 利用所获取的位姿与时间信息， 通过路径积分计算当前载体在坐标系中所处的位置； 最后， 构建基于认知点的拓扑经验地图。 此外， 在局部场景细胞获取环境信息的同时， 通过SURF特征匹配算法来进行闭环检测， 判断是否需要对当前位置进行修正。 本文提出的优化类脑SLAM模型很大程度改进了原有模型在有光线变化的室内情况下易产生场景误匹配的问题， 并通过实验验证了本文提出方法的有效性。
关键词：类脑导航； 同步定位和构图； 特征匹配算法； 经验地图； 闭环检测
引用格式：WANG Wenzhao， TANG Jun， SHEN Chong, et al． A brain-inspired SLAM navigation based on speeded up robust features matching algorithm． Journal of Measurement Science and Instrumentation， 2023， 14（2）： 137-147. DOI： 10．3969／j．issn．1674-8042．2023．02．002
[full text view]
