-
交互式人工智能系统的构建依赖于高性能人工感知系统和处理系统的开发. 传统的感知处理系统传感器、存储器和处理器在空间上是分离的, 感知数据信息的频繁传输和数据格式转换造成了系统的长延时与高能耗. 受生物感知神经系统的启发, 耦合感知、存储、计算功能的感存算一体化技术为未来感知处理领域提供了可靠的技术方案. 具有感知光、压力、化学物质等能力的忆阻器是应用于感存算一体系统的理想器件. 本文从器件层面综述了应用于感存算一体化系统忆阻器的研究方向和研究进展, 包括视觉、触觉、嗅觉、听觉和多感官耦合类别, 并在器件、工艺与集成、电路系统架构和算法方面指出现阶段的挑战与展望, 为未来神经形态感存算一体化系统的发展提供可行的研究方向.To develop future interactive artificial intelligence system, the construction of high-performance human perception system and processing system is vital. In a traditional perceptual and processing system, sensors, memory and processing units are physically separated because of their different functions and manufacture conditions, which results in frequent shuttling and format transformation of data resulting in long time delay and high energy consumption. Inspired by biological sensory nervous system, one has proposed the concept of in-sensor computing system in which the basic unit integrates sensor, storage and computing functions in the same place. In-sensor computing technology can provide a reliable technical scheme for the area of sensory processing. Artificial memristive synapse capable of sensing light, pressure, chemical substances, etc. is one type of ideal device for the application of in-sensor computing system. In this paper, at the device level, recent progress of sensory memristive synapses applied to in-sensor computing systems are reviewed, including visual, olfactory, auditory, tactile and multimode sensation. This review points out the challenge and prospect from the aspects of device, fabrication, integrated circuit system architecture and algorithms, aiming to provide possible research direction for future development of in-sensor computing system.
-
Keywords:
- in-sensor computing/
- memristors/
- artificial synapses/
- sensors
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82] [83] [84] [85] [86] -
忆阻器结构 响应类型 阻变机理 开启/关闭
电压/V开关比 PSC STP LTP 具体实现功能 文献 视觉 Ag/CH3NH3PbI3(OHP)/ITO — 碘空位导电细丝 0.32/–0.52 1×104 √ √ √ 数字识别分类 [47] Ni/Al2O3/Au UV 金属导电细丝 1.7/–1.6 1×102 — — — 图像记忆 [38] Pd/MoOx/ITO UV 界面效应 –2.13 40 √ √ √ 图像预处理 [39] Ag nanowire/TiO2 visible light (vis) 界面效应 — — √ √ √ 广角感知、处理存储 [50] glass/ITO/ZnO/PbS/ZnO/Al UV/infrared ray (IR) 氧空位导电细丝 — — √ √ √ 数字识别分类 [45] ITO/Nb:SrTiO3 vis 界面效应 — — √ √ √ 自适应光电突触 [48] ITO/PEDOT:PSS/CuSCN/CsPbBr3PNs/Au UV 界面效应 — — √ √ √ 回溯记忆功能的图像记忆 [51] ITO/SnO2/CsPbCl3/TAPC/TAPC:MoO3/MoO3/Ag/MoO3 UV/red light 界面效应 — — √ √ √ 双模式图像检测记忆 [42] RGO/GO-NCQD/graphene UV 氧空位导电细丝 — — √ √ √ 图像识别 [53] ITO/CsPbBr2I/P3HT/Ag vis/NIR 卤素空位导电细丝 0.4/–0.4 > 10 √ √ √ 图像预处理 [46] ITO/PCBM/MAPbI3:Si NCs/Spiro-OMeTAD/Au UV/NIR/vis 界面效应 — — √ √ — 图像预处理 [54] Au/Ag-TiO2/FTO vis/UV 表面等离子体共振效应/金属导电细丝 3.4/–1.8 1×103 √ √ √ 图像预处理及识别 [56] Ag/Cu3P/ITO λ= 660 nm 金属导电细丝 — 1×104 √ √ √ 回溯记忆功能的图像记忆 [57] Ni/p-NiO/n-ZnO/Ni UV 界面效应 — — √ — — 图像记忆 [40] ITO/MXene-ZnO/Al UV 氧空位导电细丝 -0.5/1.2 1×104 √ — √ 图像预处理及数字识别分类 [41] ITO/ZnO/Ag 白光 金属导电细丝 2/–2 — √ √ √ 人脸识别 [44] NiO/TiO2/FTO UV 界面效应 — > 10 √ √ √ 识别分类图像 [59] 触觉 Au/Nafion/ITO 压力 质子迁移 — — √ √ — 手写字母识别 [61] NiO/ZnO/ITO/PET 应变 界面效应 — — √ √ √ 外部应变的时空信息处理 [62] Si/NbOx/TiN 压力 晶体NbO2通道 VTH= 2.05 V
VH= 1.53 V— — — — 将压力模拟信号转换为动态振荡频率 [63] ITO/ZnO/NSTO 压力 界面效应 — 1×104 √ √ — 识别和记忆手写字母和单词 [64] Al/TiO2/Al 压力 氧空位导电细丝 — 14.2 √ — √ 压力实时感知、学习/推理、反馈可视化图像 [65] Pt/HfO2/TiN 压力 氧空位导电细丝 0.9–1.1/–1 > 100 √ — √ 触觉记忆学习 [66] ZnO/PVA基忆阻器 压力 界面效应 VTH= 3.25 V 1 × 103 √ √ √ 识别压力分布, 触觉可视化 [68] 嗅觉 Pd/W/WO3/Pd 乙醇、甲烷、乙烯、一氧化碳 氧空位导电细丝 — — √ — √ 气体识别 [73] Ti/rGO-CS/Au H2S 界面效应 — — √ √ — 气体识别 [75] 
下载:
导出CSV
-
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82] [83] [84] [85] [86]
下载:
计量
- 文章访问数:10212
- PDF下载量:667
- 被引次数:0