Periodic Reporting for period 1 - NEHIL (Neuromorphic-Enhanced Heterogeneously-Integrated FMCW LiDAR)
Periodo di rendicontazione: 2024-10-01 al 2026-01-31
The NEHIL project, an EU-Korea partnership, is set to transform the landscape of digital technologies through groundbreaking neuromorphic architectures and advanced heterogenous integration such as LiDAR systems. This collaborative initiative aims to develop two innovative neuromorphic computing architectures that are crucial for tackling the complex demands of modern data-intensive applications. The first system utilizes FeFET-based Compute-in-Memory (CIM) accelerators, which are designed to support hybrid models of SNN and ANN. These accelerators enhance processing speeds and reduce power consumption, making them ideal for real-time, high-resolution data processing challenges like those found in autonomous vehicle navigation. The second system employs photonic integrated circuits based on reservoir computing (RC) principles, significantly easing manufacturing while enhancing the processing of dynamic data streams. The work involves integrating these neuromorphic systems with state-of-the-art FMCW LiDAR technologies. This integration aims to overcome traditional challenges such as high energy consumption and environmental sensitivity, setting new standards for resolution, accuracy, and cost-efficiency. Specific targets for the NEHIL project include reducing power consumption in object recognition tasks by 50% and achieving a proof-of-concept for ultra-low latency LiDAR signal processing using the FeFET-based CIM and RC architectures. With this approach we exploit the LiDAR’s high-resolution capabilities in adverse weather conditions; reducing power consumption, packaging size and manufacturing cost compared to the state of the art. This collaboration extends its benefits beyond the automotive industry, enhancing capabilities in diverse sectors such as telecom, healthcare, smart cities, security, predictive maintenance, infrastructure management, industrial automation.
The NEHIL project achieved significant technical progress across neuromorphic computing, photonic reservoir systems, and integrated FMCW LiDAR. In WP2, the consortium developed the Conceptor Control Loop (CCL), demonstrating robustness to 13.3% neuron failures and strong adaptability to input distortions, as well as advancing theory on emergent collective learning validated through published work. A feasibility pipeline for reservoir‑computing‑based LiDAR signal correction was also established.
WP3 delivered the first photonic reservoir computing chip with passive linear nodes and MZI‑based readout, showing effective compensation of nonlinear optical distortion and confirming functional resonator‑based RC behaviour.
In WP4, the consortium defined the chip‑scale FMCW LiDAR architecture, validated system variants through simulation, and selected the monostatic FPA approach due to superior range performance. Work progressed on pixel arrays, an FMCW‑grade on‑chip laser, and custom TX/RX ASIC designs.
WP5 demonstrated low‑temperature FeFETs, compute‑in‑memory circuit
WP3 delivered the first photonic reservoir computing chip with passive linear nodes and MZI‑based readout, showing effective compensation of nonlinear optical distortion and confirming functional resonator‑based RC behaviour.
In WP4, the consortium defined the chip‑scale FMCW LiDAR architecture, validated system variants through simulation, and selected the monostatic FPA approach due to superior range performance. Work progressed on pixel arrays, an FMCW‑grade on‑chip laser, and custom TX/RX ASIC designs.
WP5 demonstrated low‑temperature FeFETs, compute‑in‑memory circuit
The project has already delivered convincing intermediate results that demonstrate the potential of neuromorphic‑enhanced FMCW LiDAR and integrated photonic–neuromorphic processing to enable more energy‑efficient, lower‑latency and higher‑performance sensing systems. Early outcomes show that local, neuromorphic, event‑driven processing can markedly reduce data‑transfer needs and computational load, while the photonic platform supports compactness and high‑resolution signal generation. These results position the technology as a promising foundation for next‑generation perception solutions in domains such as autonomous systems, robotics, smart infrastructure and industrial monitoring.
Beyond technical achievements, the project strengthens collaboration between European and Korean partners, establishing a transcontinental research ecosystem that accelerates knowledge exchange and contributes to long‑term impact. Stakeholder interactions and internal demonstrations indicate growing interest from potential adopters who value the combination of low power consumption, system integration potential and high sensing fidelity.
To ensure that these results translate into real‑world uptake, several needs have emerged. Continued research and system‑level optimisation are required to mature the technology toward higher TRLs and to validate performance in more demanding environments. Larger‑scale demonstrators and pilot deployments will be essential to build user confidence and reduce perceived adoption risks. For successful market uptake, clear exploitation strategies, stronger IP protection and tailored commercialisation support will be needed, along with access to markets, partnerships and financing mechanisms suited to deep‑tech hardware. Maintaining international cooperation and engaging with emerging regulatory and standardisation efforts will further support future deployment, particularly in application areas that require certification or compliance with safety‑critical standards.
Beyond technical achievements, the project strengthens collaboration between European and Korean partners, establishing a transcontinental research ecosystem that accelerates knowledge exchange and contributes to long‑term impact. Stakeholder interactions and internal demonstrations indicate growing interest from potential adopters who value the combination of low power consumption, system integration potential and high sensing fidelity.
To ensure that these results translate into real‑world uptake, several needs have emerged. Continued research and system‑level optimisation are required to mature the technology toward higher TRLs and to validate performance in more demanding environments. Larger‑scale demonstrators and pilot deployments will be essential to build user confidence and reduce perceived adoption risks. For successful market uptake, clear exploitation strategies, stronger IP protection and tailored commercialisation support will be needed, along with access to markets, partnerships and financing mechanisms suited to deep‑tech hardware. Maintaining international cooperation and engaging with emerging regulatory and standardisation efforts will further support future deployment, particularly in application areas that require certification or compliance with safety‑critical standards.