Rising demands for bandwidth,speed,and energy efficiency are reshaping the landscape of computing beyond the limits of von Neumann electronics.Neuromorphic photonics—using light to emulate neural computation—offers ...Rising demands for bandwidth,speed,and energy efficiency are reshaping the landscape of computing beyond the limits of von Neumann electronics.Neuromorphic photonics—using light to emulate neural computation—offers ultrafast,massively parallel,and low-energy information processing,positioning integrated photonic neural networks(IPNNs)as promising hardware for next-generation artificial intelligence(AI).By combining the architectural efficiency of neuromorphic models with the physical advantages of integrated photonics,IPNNs enable high-speed and programmable linear operations during the in-plane optical transmission,while leaving room for compact and reconfigurable on-chip optical nonlinearities and memory functions.Firstly,we review the concepts and principles of key building blocks in IPNN,that are photonic synapses,neurons,and photonic memristors which offer optical memory and storage capabilities.And then,we summarize the representative IPNN architectures and their recent advances,including coherent,parallel,diffractive,and reservoir computing,for photonic neuromorphic computing with high throughput and high efficiency.Finally,we outline practical considerations—calibration and stability of large-scale networks,routes toward co-integration with electronics,diffractive–interferometric hybrid architectures,and programmable photonic architectures for general AI purposes.We highlight a forward outlook on enabling IPNN with low energy consumption,robust photonic operations,and efficient training strategies,aiming to guide the maturation of general-purpose,low-power photonic AI.展开更多
Unmanned aerial vehicles(UAVs)face challenges in real-time high-precision tasks such as data acquisition and environmental monitoring,which demand low-latency decision-making and high-speed data transmission.A photoni...Unmanned aerial vehicles(UAVs)face challenges in real-time high-precision tasks such as data acquisition and environmental monitoring,which demand low-latency decision-making and high-speed data transmission.A photonic reservoir computing(PRC)system based on silicon-integrated micro-ring resonators(MRRs)was demonstrated for UAV flight direction recognition,achieving an accuracy of 95.6%.The proposed PRC exhibits short-term memory effects,enabling the detection of subtle changes in flight direction.The PRC operates at the MHz-rate with milliwatt-level power consumption,enabling photonic edge computing with high speed and energy efficiency,providing UAVs the potential to make decisions autonomously and avoid external interference when operating in extreme environments.展开更多
基金supported by the Shanghai Municipal Science and Technology Major Project, the Science and Technology Commission of Shanghai Municipality (No. 21DZ1100500)the Shanghai Frontiers Science Center Program (2021-2025 No. 20)+2 种基金the National Key Research and Development Program of China (No. 2021YFB2802000)the National Natural Science Foundation of China (No. 61975123, No. 62305217),the National Natural Science Foundation of China (No. 52075504)the Shanghai Pujiang Programme
文摘Rising demands for bandwidth,speed,and energy efficiency are reshaping the landscape of computing beyond the limits of von Neumann electronics.Neuromorphic photonics—using light to emulate neural computation—offers ultrafast,massively parallel,and low-energy information processing,positioning integrated photonic neural networks(IPNNs)as promising hardware for next-generation artificial intelligence(AI).By combining the architectural efficiency of neuromorphic models with the physical advantages of integrated photonics,IPNNs enable high-speed and programmable linear operations during the in-plane optical transmission,while leaving room for compact and reconfigurable on-chip optical nonlinearities and memory functions.Firstly,we review the concepts and principles of key building blocks in IPNN,that are photonic synapses,neurons,and photonic memristors which offer optical memory and storage capabilities.And then,we summarize the representative IPNN architectures and their recent advances,including coherent,parallel,diffractive,and reservoir computing,for photonic neuromorphic computing with high throughput and high efficiency.Finally,we outline practical considerations—calibration and stability of large-scale networks,routes toward co-integration with electronics,diffractive–interferometric hybrid architectures,and programmable photonic architectures for general AI purposes.We highlight a forward outlook on enabling IPNN with low energy consumption,robust photonic operations,and efficient training strategies,aiming to guide the maturation of general-purpose,low-power photonic AI.
基金supported by the Shanghai Municipal Science and Technology Major ProjectNational Key R&D Program of China(Nos.2021YFB2802000 and 2022YFB2804300)+4 种基金the Shanghai Frontiers Science Center Program(2021-2025No.20)the Science and Technology Commission of Shanghai Municipality(No.21DZ1100500)the National Natural Science Foundation of China(Nos.61975123 and 62305217)the Shanghai Science and Technology Innovation Program(No.23JC1403100)the Shanghai Pujiang Programme。
文摘Unmanned aerial vehicles(UAVs)face challenges in real-time high-precision tasks such as data acquisition and environmental monitoring,which demand low-latency decision-making and high-speed data transmission.A photonic reservoir computing(PRC)system based on silicon-integrated micro-ring resonators(MRRs)was demonstrated for UAV flight direction recognition,achieving an accuracy of 95.6%.The proposed PRC exhibits short-term memory effects,enabling the detection of subtle changes in flight direction.The PRC operates at the MHz-rate with milliwatt-level power consumption,enabling photonic edge computing with high speed and energy efficiency,providing UAVs the potential to make decisions autonomously and avoid external interference when operating in extreme environments.
