基于多尺度网格细胞的路径整合模型

doi:10.12305/j.issn.1001-506X.2021.10.31

系统工程与电子技术 ›› 2021, Vol. 43 ›› Issue (10): 2961-2967.doi: 10.12305/j.issn.1001-506X.2021.10.31

基于多尺度网格细胞的路径整合模型

赵辰豪*, 吴德伟, 韩昆, 朱浩男, 代传金

空军工程大学信息与导航学院, 陕西西安 710077

收稿日期:2020-08-22 出版日期:2021-10-01 发布日期:2021-11-04
通讯作者: 赵辰豪
作者简介:赵辰豪(1996—), 男, 硕士研究生, 主要研究方向为智能自主导航理论与技术|吴德伟(1963—), 男, 教授, 博士, 主要研究方向为智能导航与协同控制|韩昆(1990—), 男, 博士研究生, 主要研究方向为智能自主导航理论与技术|朱浩男(1994—), 男, 博士研究生, 主要研究方向为智能自主导航理论与技术|代传金(1980—), 男, 副教授, 博士, 主要研究方向为智能导航与协同控制
基金资助:
国家自然科学基金(61973314)

Path integration model based on multi-scale grid cell

Chenhao ZHAO*, Deiwei WU, Kun HAN, Haonan ZHU, Chuanjin DAI

Information and Navigation College, Air Force Engineering University, Xi'an 710077, China

Received:2020-08-22 Online:2021-10-01 Published:2021-11-04
Contact: Chenhao ZHAO

摘要/Abstract

摘要：

针对类脑机制下空间位置自主推算问题, 提出一种基于新型叠加算法的路径整合模型, 通过引入多尺度网格细胞提高路径整合结果的准确性。在路径整合模型中, 基于吸引子网络构建网格细胞与头朝向细胞模型, 通过头朝向变化量来判断运行体的运行状态, 完成不同运动的整合过程; 采用动态权重值分别计算直线运动与曲线运动过程中多尺度网格细胞的位移变化量, 提高路径整合的精度。仿真结果表明, 单一尺度与多尺度网格细胞路径整合模型都能够完成路径整合, 且在3 000 m的距离内, 多尺度网格细胞模型误差不超过15 m, 证明了多尺度网格细胞模型能够提高路径整合的精度, 验证了该路径整合模型的有效性。

关键词: 多尺度, 网格细胞, 头朝向细胞, 吸引子网络模型, 路径整合

Abstract:

To solve the problem of autonomic spatial location estimation using brain-like mechanism, we propose a path integration model based on new superposition algorithm. By introducing multi-scale grid cell, the accuracy of path integration is improved. In path integration model, both grid cell model and head-direction model are realized by attractor network. The path integration model judges the state through the head direction's change to complete different integration process, and calculates the displacement changes of multi-scale grid cell model during linear motion and curvilinear motion via a dynamic weight to improve the accuracy of path integration. The results show that both single-scale and multi-scale model can complete path integration, and in the distance of 3 000 m, the error of multi-scale grid cell model does not exceed 15 m. It is proved that the multi-scale model can improve the accuracy of path integration and verified that the path integration model is effective.

Key words: multi-scale, grid cell, head direction cell, attractor network model, path integration

中图分类号:

TP183

赵辰豪, 吴德伟, 韩昆, 朱浩男, 代传金. 基于多尺度网格细胞的路径整合模型[J]. 系统工程与电子技术, 2021, 43(10): 2961-2967.

Chenhao ZHAO, Deiwei WU, Kun HAN, Haonan ZHU, Chuanjin DAI. Path integration model based on multi-scale grid cell[J]. Systems Engineering and Electronics, 2021, 43(10): 2961-2967.

图/表 7

图1

图2

表1

图3

图4

图5

图6

参考文献 22

1	HAFTING T , FYHN M , MOLDEN S , et al. Microsruct-ture of a spatial map in the entorhinal cortex[J]. Nature, 2005, 436 (11): 801- 806.
2	OKEEFE J , DOSTROVSKY J . The hippocampus as a spatial map. preliminary evidence from unit activity in the freely-moving rat[J]. Brain Research, 1971, 34 (1): 171- 175. doi: 10.1016/0006-8993(71)90358-1
3	TAUBE J , MULLER R , RANCK J . Head-direction cells recorded from the postsubiculum in freely moving rats. I. description and quantitative analysis[J]. The Journal of Neuroence, 1990, 10 (2): 420- 435.
4	EMILIO K , JAMES E . Speed cells in the medial entorhinal cortex[J]. Nature, 2015, 523 (7561): 419- 424. doi: 10.1038/nature14622
5	SOLSTAD T , BOCCARA C N . Representation of geometric borders in the entorhinal cortex[J]. Science, 2008, 322 (5909): 1865- 1868. doi: 10.1126/science.1166466
6	吴德伟, 何晶, 韩昆, 等. 无人作战平台认知导航及其类脑实现思想[J]. 空军工程大学学报(自然科学版), 2018, 19 (6): 37- 42.
	WU D W , HE J , HAN K , et al. Cognitive navigation and its thought of brain-inspired realization in unmanned combat platform[J]. Journal of Air Force Engineering University (Natural Science Edition), 2018, 19 (6): 37- 42.
7	FYHN M , MOLDEN S , WITTER M P , et al. Spatial representation in the entorhinal cortex[J]. Science, 2004, 305 (5688): 1258- 1264. doi: 10.1126/science.1099901
8	TANG H J , YAN R , TAN K C , et al. Cognitive navigation by neuro-inspired localization, mapping, and episodic memory[J]. IEEE Trans.on Cognitive & Developmental Systems, 2017, (2): 30- 37.
9	斯白露, 罗壹凡. 空间记忆与类脑导航研究进展[J]. 人工智能, 2020, (1): 16- 31.
	SI B L , LUO Y F . Research progress of spatial memory and brain-like navigation[J]. Artificial Intelligence, 2020, (1): 16- 31.
10	吴德伟. 导航原理[M]. 北京: 电子工业出版社, 2015.
	WU D W . The principle of navigation[M]. Beijing: Publishing House of Electronics Industry, 2015.
11	赵辰豪, 吴德伟, 何晶, 等. 基于改进Q学习算法的导航认知图构建[J]. 空军工程大学学报(自然科学版), 2020, 21 (2): 53- 60. doi: 10.3969/j.issn.1009-3516.2020.02.008
	ZHAO C H , WU D W , HE J , et al. Navigation cognitive map construction based on improved q-learning algorithm[J]. Journal of Air Force Engineering University (Natural Science Edition), 2020, 21 (2): 53- 60. doi: 10.3969/j.issn.1009-3516.2020.02.008
12	GIOCOMO , LISA M , ZILLI G , et al. Temporal frequency of subthreshold oscillations scales with entorhinal grid cell field spacing[J]. Science, 2007, 315 (5819): 1719- 1722. doi: 10.1126/science.1139207
13	杨闯, 刘建业, 熊智, 等. 由感知到动作决策一体化的类脑导航技术研究现状与未来发展[J]. 航空学报, 2020, 41 (1): 30- 44.
	YANG C , LIU J Y , XIONG Z , et al. Brain-inspired navigation technology integrating perception and action decision: a review and outlook[J]. Acta Aeronautica et Astronautica Sinica, 2020, 41 (1): 30- 44.
14	YUAN M, TIAN B, SHIM V A, et al. An entorhinal-hippo-campal model for simultaneous cognitive map building[C]//Proc. of the 29th AAAI Conference on Artificial Intelligence, 2015: 586-592.
15	EDVARDSEN V. Long-range navigation by path integration and decoding of grid cells in neural network[C]//Proc. of the International Joint Conference on Neural Network, 2017: 4348-4355.
16	LU H, XIAO J, ZHANG L, et al. Biologically inspired visual odometry based on the computional model of grid cells for mobile robots[C]//Proc. of the IEEE International Conference on Robotics & Biomimetics, 2017: 595-601.
17	ISLAM T, FUKUZAKI R. A model of path integration and navigation based on head direction cells in entorhinal cortex[C]//Proc. of the Berlin Heidelberg Springer, 2010: 82-90.
18	WALTERS D M , STRINGER S M . Path integration of head direction: updating a packet of neural activity at the correct speed using neuronal time constants[J]. Biological Cyberne-tics, 2010, 103 (1): 21- 41. doi: 10.1007/s00422-009-0355-0
19	KUBIE J L , FENTON A . Heading-vector navigation based on head-direction cells and path integration[J]. Hippocampus, 2009, 19 (5): 456- 479. doi: 10.1002/hipo.20532
20	ZHOU Y , DEWEI W U . Biologically inspired model of path integration based on head direction cells and grid cells[J]. Frontiers of Information Technology & Electronic Engineering, 2016, 17 (5): 435- 448.
21	PAGE H J I , WALTERS D , STRINGER S M . A speed-accurate self-sustaining head direction cell path integration model without recurrent excitation[J]. Network Computation in Neural Systems, 2018, 29 (4): 37- 69.
22	BURAK Y , FIETE I R . Accurate path integration in continuous attractor network models of grid cells[J]. PLoS Computer, 2009, 5, e1000291. doi: 10.1371/journal.pcbi.1000291

参数	数值
头朝向细胞个数	36
旋转细胞个数	72
α_HD	0
β_HD	0.2
网格细胞的个数	40×40
α_gc	1
λ₁	13
λ₂	15
λ₃	17
λ₄	19
λ₅	21
β_gc	0.017 8
γ	0.019 5
l	2
a	1
τ/ms	10
dt/ms	0.5

[1]	汤寓麟, 李厚朴, 张卫东, 边少锋, 翟国君, 刘敏, 张晓平. 侧扫声纳检测沉船目标的轻量化DETR-YOLO法[J]. 系统工程与电子技术, 2022, 44(8): 2427-2436.
[2]	范加利, 田少兵, 黄葵, 朱兴动. 基于Faster R-CNN的航母舰面多尺度目标检测算法[J]. 系统工程与电子技术, 2022, 44(1): 40-46.
[3]	金煌煌, 庄志洪, 付梦印, 郭健, 王宏波. 最简多磁偶极子等效建模方法[J]. 系统工程与电子技术, 2021, 43(8): 2066-2075.
[4]	赵辰豪, 吴德伟, 韩昆, 代传金. 无环境信息下多尺度网格细胞群空间表征模型[J]. 系统工程与电子技术, 2021, 43(3): 814-822.
[5]	花思齐, 赵伟, 刘建业. 基于改进滤波器和图像多尺度变换的背景抑制算法[J]. 系统工程与电子技术, 2020, 42(8): 1679-1684.
[6]	韩昆, 吴德伟, 来磊. 类脑导航中基于差分Hebbian学习的网格细胞构建模型[J]. 系统工程与电子技术, 2020, 42(3): 674-679.
[7]	蒋铁军, 张怀强, 周成杰. 事件影响下的时间序列多尺度集成预测[J]. 系统工程与电子技术, 2020, 42(2): 390-397.
[8]	刘明骞, 颜志文, 张俊林. 空中目标辐射源的个体识别方法[J]. 系统工程与电子技术, 2019, 41(11): 2408-2415.
[9]	蔡军, 黄袁园, 李鹏泽, 赵子硕, 邓撬. 基于视觉对比度机制的红外弱小目标检测算法[J]. 系统工程与电子技术, 2019, 41(11): 2416-2423.
[10]	常莉红, 冯象初, 张瑞. 四元数小波变换联合稀疏表示的图像融合[J]. 系统工程与电子技术, 2017, 39(7): 1633-1639.
[11]	刘斌, 刘维杰, 魏艳萍. 6通道MSVD构造及其在多聚焦图像融合中的应用[J]. 系统工程与电子技术, 2015, 37(9): 2191-2197.
[12]	曹向海, 邓湖明, 黄波. 背景感知的显著性检测算法[J]. 系统工程与电子技术, 2014, 36(8): 1668-1672.
[13]	王浩，周祚峰，曹剑中，闫肃. 基于边缘轮廓上多尺度自相关矩阵的角点检测算法[J]. 系统工程与电子技术, 2014, 36(6): 1220-1224.
[14]	徐海霞，温显斌，张建光. 基于多尺度贝叶斯网络的SAR图像分割[J]. 系统工程与电子技术, 2014, 36(6): 1075-1080.
[15]	黄波，赵晓晖，赵继印，时公涛，陈涛. 一种新的多尺度仿射几何不变量提取方法[J]. Journal of Systems Engineering and Electronics, 2012, 34(5): 1051-1057.

基于多尺度网格细胞的路径整合模型

Path integration model based on multi-scale grid cell

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 7

参考文献 22

相关文章 15

编辑推荐

Metrics

本文评价