Multi-Sensor Edge Angle Detection for Performance Analysis in Ski Jumping

In ski jumping, performance during the gliding phase depends on achieving an aerodynamic posture that maximizes the lift-to-drag ratio. In the V-style technique, the ski edge angle is a key determinant. Reducing the edge angle flattens the skis, increases their effective surface area, and improves aerodynamic lift, ultimately contributing to longer flight distances. Ski edge angles are biomechanically constrained by the limited range of ankle inversion. Current sensing solutions widely quantify these angles using multi-system approaches that combine sensor signals through geometric relations. Such configurations require instrumentation on both the boot and the ski, altering mass distribution, affecting balance during flight, and increasing system complexity. To overcome these limitations, this work presents a wearable sensing system that measures both boot inclination and ski edge angle without modifying the ski surface. Two ultrasonic Time of Flight (ToF) sensors and an in-shoe Inertial Measurement Unit (IMU) are integrated into a single boot-mounted unit. Edge angles are estimated by combining ultrasonic distance measurements with IMU data through geometric reconstruction of the boot-ski configuration. Laboratory experiments demonstrate an angle resolution of 0.4500°, a Mean Absolute Error (MAE) of 0.2640°, and a coefficient of determination exceeding 99\% when compared with reference measurements, indicating strong linear agreement between the two modalities. The system achieves an end-to-end latency of 30.31 ms, enabling real-time feedback suitable for athlete training, while consuming 1.28 mW of power. With a total weight of only 18.6 g the proposed system enables unobtrusive measurement of ski edge angle and boot orientation.

S-CEReBrO: Breaking the Memory Barrier in Continuous EEG Monitoring

Foundation models offer a promising paradigm for Electroencephalography (EEG) analysis, leveraging generalizable representations from vast unlabeled datasets. Yet, Transformer-based architectures face a critical bottleneck: global attention mechanisms couple the attention memory state to the signal duration, causing memory overflow during continuous monitoring. To address this, we introduce S-CEReBrO (Streaming CEReBrO), an evolution of the CEReBrO architecture designed for continuous monitoring. Our novel Windowed Alternating Attention mechanism factorizes attention computation into fixed-size spatiotemporal windows, guaranteeing constant KV cache memory as only the active window requires resident attention maps. Empirical scaling analysis confirms that windowed alternating attention can process signals 100X longer than full self-attention and 3X longer than low-rank linear attention. Compared to low-rank linear attention on long contexts, windowed alternating attention requires 55% of the memory while increasing inference throughput by 2.1X. Pre-trained on >25,000 hours of recordings from >12,000 subjects, S-CEReBrO achieves state-of-the-art performance on 7 of 11 downstream tasks, with up to 60% fewer parameters. This work represents a significant step toward the realization of efficient, generalizable, and continuous EEG monitoring. An accompanying code repository is available.

At-the-Roofline Sparse Tensor Contractions on Vector Processors for Transformer Inference

Fine-grained weight pruning and activation sparsification have emerged as effective approaches for reducing the compute and memory cost of inference for Transformer models. In the moderate-sparsity regime, Gustavson's dataflow provides a natural execution model for exploiting both activation and weight sparsity on vector processors through metadata-driven indexed accumulation. However, existing RVV architectures lack native support for this pattern, forcing kernels to rely on software index decoding and L1-backed indexed memory operations that keep sparse tensor contractions far below their roofline performance bound. We present Ventaglio, a runtime-configurable sparse execution unit coupled with RVV ISA extensions that drives sparse tensor contractions toward their roofline through indexed gather-accumulate-scatter support. Integrated into an open-source vector processing cluster and implemented in 12nm FinFET, Ventaglio accelerates sparse tensor contraction kernels by $6.9\text{--}7.4\times$ over optimized RVV baselines, with only $3.1\%$ area overhead for a cluster of tightly-L1 coupled vector processing elements. We build a performance-accurate instruction-level model of the Ventaglio extension, calibrate it against RTL implementation, and leverage it for scale-out performance analysis on a large $4\times4$ multi-cluster system. Using a DuoGPT-pruned LLaMA-3-8B model with practical $40\text{--}60\%$ dual sparsity, Ventaglio achieves $2.40\text{--}5.25\times$ and $2.06\text{--}3.16\times$ speedup over dense baselines during prefill and autoregressive decoding, respectively.

Scalable Attention for 5G NR Channel Estimation

Attention-based neural estimators achieve strong channel-estimation accuracy, but the computational cost of global attention over the time-frequency resource grid grows quadratically with the number of subcarriers, and these estimators are typically tied to a single resource allocation. This paper proposes Channel Estimation Attention (CHEA), a low-complexity channel estimator for 5G New Radio (5G NR) multi-user multiple-input multiple-output (MU-MIMO). CHEA replaces global attention with a multi-resolution windowed design: a high-resolution encoder preserves local pilot detail, a low-resolution encoder captures wider frequency-domain context, and a local cross-attention decoder transfers this coarse context back to the high-resolution pilot tokens. A per-Physical Resource Block (PRB) upsampling module then reconstructs the channel over the full slot. Because every attention operation is confined to a fixed-size window and reconstruction is performed per PRB, the cost of CHEA scales linearly with the number of subcarriers, and a single trained model supports different PRB allocations without retraining. On a standard-compliant Physical Uplink Shared Channel (PUSCH), CHEA achieves the lowest Mean Squared Error (MSE) among conventional and state-of-the-art neural estimators, while requiring 2.8\(\times\) to 22.0\(\times\) lower operations than existing attention-based estimators.

Improving Autonomous Nano-drones Performance via Automated End-to-End Optimization and Deployment of DNNs

The evolution of energy-efficient ultra-low-power (ULP) parallel processors and the diffusion of convolutional neural networks (CNNs) are fueling the advent of autonomous driving nano-sized unmanned aerial vehicles (UAVs). These sub-10 cm robotic platforms are envisioned as next-generation ubiquitous smart-sensors and unobtrusive robotic-helpers. However, the limited computational/memory resources available aboard nano-UAVs introduce the challenge of minimizing and optimizing vision-based CNNs -- which to date require error-prone, labor-intensive iterative development flows. This work explores methodologies and software tools to streamline and automate all the deployment of vision-based CNN navigation on a ULP multicore system-on-chip acting as a mission computer on a Crazyflie 2.1 nano-UAV. We focus on the deployment of PULP-Dronet, a state-of-the-art CNN for autonomous navigation of nano-UAVs, from the initial training to the final closed-loop evaluation. Compared to the original hand-crafted CNN, our results show a 2x reduction of memory footprint and a speedup of 1.6x in inference time while guaranteeing the same prediction accuracy and significantly improving the behavior in the field, achieving: i) obstacle avoidance with a peak braking-speed of 1.65 m/s and improving the speed/braking-space ratio of the baseline, ii) free flight in a familiar environment up to 1.96 m/s (0.5 m/s for the baseline), and iii) lane following on a path featuring a 90 deg turn -- all while using for computation less than 1.6% of the drone's power budget. To foster new applications and future research, we open-source all the software design in a ready-to-run project compatible with the Crazyflie 2.1

WULPUS PRO: Multi-mode Ultra-Low-Power Wearable Ultrasound and Array Imaging with CMUT Support

Wearable ultrasound enables continuous monitoring of physiological processes such as muscle dynamics, bladder volume, and cardiovascular activity. Existing fully wearable ultra-low-power platforms are limited to shallow, low-channel A-mode sensing, while larger multi-mode systems are too bulky and power-hungry for true wearability. We present WULPUS PRO, a runtime-programmable wearable ultrasound acquisition platform measuring $39\times21\times6 \mathrm{mm}$ and weighing $5 \mathrm{g}$. It integrates $30 \mathrm{V}$ excitation, 16 time-multiplexed channels, a low-noise receive front-end with up to $70 \mathrm{dB}$ gain, $9.9 \mathrm{MHz}$ bandwidth, time-gain compensation, and $32 \mathrm{dB}$ SNR. The platform supports deep-tissue echo acquisition up to $2.2 \mathrm{MHz}$ in RF-sampling mode and $8 \mathrm{MHz}$ in envelope-detection mode. We demonstrate B-mode imaging in a 16-channel ultra-low-power wearable with sub-millimeter axial and millimeter-scale lateral resolution in phantom experiments, while consuming $40 \mathrm{mW}$ at $50 \mathrm{Hz}$ PRF and under $60 \mathrm{mW}$ at $300 \mathrm{Hz}$ PRF. WULPUS PRO supports both piezoelectric and capacitive micromachined ultrasonic transducers, enabling integration with skin-conformal polymer-based CMUT arrays. As a host-agnostic acquisition front-end, it exposes standard data and power interfaces for BLE- and Wi-Fi-based wearable hosts. We demonstrate wireless transmission with external BLE and Wi-Fi modules and project 1-2 days of BLE operation at $50 \mathrm{Hz}$ PRF and over 3 h of Wi-Fi streaming at $300 \mathrm{Hz}$ PRF using a $300 \mathrm{mAh}$, $6.4 \mathrm{g}$ Li-Po cell. WULPUS PRO establishes a new class of fully programmable, B-mode-enabled, ultra-low-power wearable ultrasound platforms.

STEEL: Sparsity-Aware Fused Attention for Energy-Efficient Long-Sequence Inference on AMD's XDNA NPU

The growing adoption of large language model-based agents within operating system workflows has increased the importance of energy-efficient inference on laptop-class systems-on-chip (SoCs). While cloud offloading remains common, it introduces reliability and privacy concerns that are particularly problematic for agentic workloads. Recent laptop SoCs, therefore, incorporate neural processing engines (NPUs) optimized for energy efficiency; however, effectively mapping attention mechanisms onto NPUs remains challenging due to architectural diversity and explicit data-movement programming models. In this work, we present STEEL, the first open-source implementation of FlashAttention targeting XDNA-like NPUs. STEEL introduces a dataflow formulation of prefill attention, enabling efficient exploitation of spatial parallelism and on-chip memory. Furthermore, STEEL addresses the load imbalance induced by the causal mask by leveraging a sparsity-aware pipeline placement onto the NPU array, reducing synchronization overhead and improving utilization. We evaluate STEEL on the AMD Ryzen AI 9 HX 370 SoC and compare its performance against optimized CPU and GPU implementations. Experimental results show that STEEL reduces energy consumption by an average of 9.17x and 1.75x relative to CPU and GPU baselines, respectively. On XDNA 1, STEEL achieves an average 9.6x latency reduction over the prior state of the art, and delivers a 22.8x speedup on average compared to a layer-by-layer attention implementation on XDNA 2.

Low-Power License Plate Detection and Recognition on a RISC-V Multi-Core MCU-Based Vision System

In this paper, we present the first (to the best of our knowledge) demonstration of a low-power MCU-based edge device for Automatic License Plate Recognition (ALPR). The design leverages on a 9-core RISC-V processor, GAP8, coupled with a QVGA ultra-low-power greyscale imager. The proposed visual processing pipeline uses a multi-model inference approach based on SSDlite-MobilenetV2 for license plate detection and LPRNet for optical character recognition, reaching a 38.9% mAP score for the first task and a recognition rate of >99.13% for the latter on public datasets. On real-world data, the pipeline recognizes registration numbers when the size of LP crops is as small as 30x5 pixels. Thanks to the applied compression and optimization strategies, the multi-model inference (687 MMAC) achieves a throughput of 1.09 FPS at a power cost of 117 mW when running on GAP8. Our solution is the first MCU-class device embedding such a level of network complexity, resulting to be 73x more energy-efficient w.r.t. precedent mobile-class ALPR system featuring a Raspberry Pi3. The proposed design does not resort to any hardwired acceleration engines, thus retaining full flexibility for future algorithmic improvements.

LocalNav: Distilling Frontier VLMs and Embodied RL for On-Device Object Goal Navigation

Vision Language Models (VLMs) have emerged in the robotic domain as a powerful tool that enables environmental perception with language context, serving as a catalyst for open-vocabulary tasks like ObjectNav. Yet, their computational footprint typically confines them to cloud execution, hindering low-latency inference with local deployment on resource-constrained robots. To address this challenge, we present a distillation strategy that transfers complex spatial-semantic reasoning from large frontier models into a lightweight, 4B-parameter local VLM for edge execution on embedded GPU devices (e.g., Jetson Orin). We first establish a State of the Art (SotA), Scene Graph (SG)-based pipeline using Claude Sonnet 4.6, achieving a 39.7% Success Rate (SR) on the HM3D OVON benchmark. We then demonstrate that fine-tuning Qwen3.5-4B on just 500 frontier reasoning traces effectively enables navigation capabilities, yielding a SR of 34.5%, narrowing the gap to the performance of large cloud models. Finally, we introduce E-RLVR with Token Generation (TG) regularization to compress output sequence lengths for physical deployment while grounding the agent in its task. This downstream optimization reduces TG overhead by 72.1% and latency by 71.8%. Combined with quantization, this joint strategy yields a cumulative 82.8% reduction in overall inference latency without significantly sacrificing performance, presenting a viable paradigm for local, low-latency VLM execution on mobile robots.

Towards Whole Hand and Wrist Kinematic Tracking with a Wearable A-Mode Ultrasound Probe

A-mode ultrasound (US) has emerged as a promising modality for hand and wrist motion tracking. Prior works have mainly addressed static gesture classification or regression of a few degrees of freedom (DoFs), typically relying on non-wearable systems and external computing devices, and highlight the need for strategies to ensure robustness to sensor repositioning. In this work, we propose a framework for robust whole-hand and wrist kinematic tracking via wearable A-mode US using the WULPUS platform, tackling the regression of 23 DoFs directly on the probe. First, we introduce a compact (11285 parameters) multi-output convolutional neural network combined with an incremental training strategy, which improves inter-session generalization and reduces mean absolute error by more than 17% compared to a non-incremental approach. Second, we demonstrate, for the first time, the feasibility of end-to-end hand and wrist kinematic tracking entirely on-device. We deploy the model on the WULPUS nRF52832 microcontroller, achieving 0.73 mJ per inference, 29.1 ms latency, and showing the feasibility of full operation (data acquisition, online inference, and BLE streaming of results) within 33 mW, enabling up to 36 hours of continuous use and an 88% reduction in wireless bandwidth compared to raw data transmission.