Neuromorphic sensory computing

Wan, Tianqing; Ma, Sijie; Liao, Fuyou; Fan, Lingwei; Chai, Yang

doi:10.1007/s11432-021-3336-8

Cited by 44 publications

(29 citation statements)

References 92 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The challenges associated with materials selection, device physics, and integration of large-scale array can be addressed, with further advances in material science, device physics, and integration technology. [114][115][116] Memristive devices based on metal/insulator/metal heterostructures have been regarded one of fundamental building blocks in neuromorphic electronics. The crystallinity of switching material and its interface quality with metal are critical to switching performance of memristive devices, as suggested in prior work.…”

Section: Challenges and Outlookmentioning

confidence: 99%

Emerging Low‐Dimensional Heterostructure Devices for Neuromorphic Computing

Liang

Cheng

et al. 2022

Small Structures

View full text Add to dashboard Cite

Over the last few decades, advance in material science has made it possible to control the dimensions of materials structures with atomic scale precision and to realize various low‐dimensional heterostructures. These heterostructures have been widely used in the electronic and spintronic as well as optoelectronic devices, with great success. Engineering the dimensionality and physical properties of low‐dimensional materials enables to design a large variety of low‐dimensional artificial heterostructures. The inherent physical properties that these emerging low‐dimensional heterostructures exhibit open unprecedented opportunities for neuromorphic computing applications, e.g., neuromorphic electronics, neuromorphic spintronics, and neuromorphic optoelectronics, which lay the foundation for building up intelligent system in the future. Herein, the current main status of the emerging low‐dimensional heterostructures used in neuromorphic electronic, neuromorphic spintronic, and neuromorphic optoelectronic devices is reviewed, and the relevant challenges and outlook for future advances are discussed.

show abstract

Section: Challenges and Outlookmentioning

confidence: 99%

Emerging Low‐Dimensional Heterostructure Devices for Neuromorphic Computing

Liang

Cheng

et al. 2022

Small Structures

View full text Add to dashboard Cite

show abstract

“…With the upsurge of artificial intelligence applications and the rapid development of internet of things (IOT), the time-efficient sensing information acquisition, energy-efficient data memorizing and processing capabilities of terminal electronic systems are indispensable 1 . However, in conventional chips, the separated sensing, storage, and processing units always need to collect signals by external sensors, convert signals into digital format data subsequently, and then transfer the date to memory units and processors for subsequent processing tasks, which limits the data conversion and movement and results in low computing efficiency and huge power consumption 2 .…”

Section: Introductionmentioning

confidence: 99%

Self-powered high-sensitivity all-in-one vertical tribo-transistor device for multi-sensing-memory-computing

Liu

Gao

et al. 2022

Nat Commun

View full text Add to dashboard Cite

Devices with sensing-memory-computing capability for the detection, recognition and memorization of real time sensory information could simplify data conversion, transmission, storage, and operations between different blocks in conventional chips, which are invaluable and sought-after to offer critical benefits of accomplishing diverse functions, simple design, and efficient computing simultaneously in the internet of things (IOT) era. Here, we develop a self-powered vertical tribo-transistor (VTT) based on MXenes for multi-sensing-memory-computing function and multi-task emotion recognition, which integrates triboelectric nanogenerator (TENG) and transistor in a single device with the simple configuration of vertical organic field effect transistor (VOFET). The tribo-potential is found to be able to tune ionic migration in insulating layer and Schottky barrier height at the MXene/semiconductor interface, and thus modulate the conductive channel between MXene and drain electrode. Meanwhile, the sensing sensitivity can be significantly improved by 711 times over the single TENG device, and the VTT exhibits excellent multi-sensing-memory-computing function. Importantly, based on this function, the multi-sensing integration and multi-model emotion recognition are constructed, which improves the emotion recognition accuracy up to 94.05% with reliability. This simple structure and self-powered VTT device exhibits high sensitivity, high efficiency and high accuracy, which provides application prospects in future human-mechanical interaction, IOT and high-level intelligence.

show abstract

“…In contrast to conventional sensory computing methods including analogue-to-digital signal conversion and digital-logic computing tasks (i.e., von Neumann computing), neuromorphic vision computing can dramatically improve the energy efficiency and data processing speed by minimizing unnecessary raw data transmissions between front-end photosensors and back-end post-processors (Fig. 1 ) [ 1 – 4 ]. Neuromorphic vision sensors are generally appropriately designed for neuromorphic vision computing tasks, such as denoising, edge enhancement, spectral filtering, and the recognition of visual information.…”

Section: Introductionmentioning

confidence: 99%

Progress of Materials and Devices for Neuromorphic Vision Sensors

Cho

Jo²,

Kim³

et al. 2022

Nano-Micro Lett.

View full text Add to dashboard Cite

The latest developments in bio-inspired neuromorphic vision sensors can be summarized in 3 keywords: smaller, faster, and smarter. (1) Smaller: Devices are becoming more compact by integrating previously separated components such as sensors, memory, and processing units. As a prime example, the transition from traditional sensory vision computing to in-sensor vision computing has shown clear benefits, such as simpler circuitry, lower power consumption, and less data redundancy. (2) Swifter: Owing to the nature of physics, smaller and more integrated devices can detect, process, and react to input more quickly. In addition, the methods for sensing and processing optical information using various materials (such as oxide semiconductors) are evolving. (3) Smarter: Owing to these two main research directions, we can expect advanced applications such as adaptive vision sensors, collision sensors, and nociceptive sensors. This review mainly focuses on the recent progress, working mechanisms, image pre-processing techniques, and advanced features of two types of neuromorphic vision sensors based on near-sensor and in-sensor vision computing methodologies. "Image missing"

show abstract

Neuromorphic sensory computing

Cited by 44 publications

References 92 publications

Emerging Low‐Dimensional Heterostructure Devices for Neuromorphic Computing

Emerging Low‐Dimensional Heterostructure Devices for Neuromorphic Computing

Self-powered high-sensitivity all-in-one vertical tribo-transistor device for multi-sensing-memory-computing

Progress of Materials and Devices for Neuromorphic Vision Sensors

Contact Info

Product

Resources

About