Publications
Abstract
Robust visual odometry (VO) is a fundamental prerequisite for autonomous navigation in agile robotics, particularly during aggressive flight maneuvers. Artificial Neural Network (ANN) approaches for event-based VO achieve strong results through recurrent memory and joint trajectory optimization, but they discard the high-frequency temporal resolution of event data by compressing asynchronous events into fixed-time voxel grids. To bridge this gap, we propose NOVA, a hybrid framework that combines a voxel-based ANN state estimator with a Spiking Neural Network (SNN): the ANN tracks historical states, while the SNN extracts high-frequency optical flow directly from asynchronous events. A parameter-free fusion module combines both, using SNN flow to correct spatial drift and the directional alignment between the two modalities to adjust bundle adjustment weights. Evaluations on the UZH-FPV drone racing dataset show that NOVA achieves an average Mean Position Error (MPE) of 0.45%/m, a 15% reduction over the DEVO baseline. Tests on the MVSEC benchmark further confirm that the method mitigates trajectory overshoot and maintains stable tracking without relying on supplementary inertial sensors.
Abstract
Artificial intelligence (AI)-based multimodal sensor fusion is a relevant topic gaining ever more traction across ultra-low-power (ULP) embedded and cyber-physical systems, as it improves reliability, accuracy, and robustness under real-world constraints. However, adding more and more sensors to ultra-constrained sub-100 mW platforms requires balancing energy consumption against prediction accuracy. To achieve this ambitious goal, we present a novel adaptive AI methodology that combines multimodal sensor fusion (camera, ultrasound, and Time-of-Flight sensors) with a lightweight recurrent Transformer-based architecture (688 k parameters). We address the depth map estimation task with a mechanism that combines token propagation across iterations with incremental sensor utilization. At each iteration, a confidence-based gating mechanism dynamically decides whether to continue the computation by adding progressively richer but more power-demanding sensors as input. Token propagation ensures temporal consistency by forwarding context features across time. To deploy our algorithm and test a first real-world prototype, we design a novel printed circuit board featuring all three sensors, coupled with an ULP GWT GAP9 multicore System-on-Chip. When comparing our adaptive system against the same pipeline using all sensors and iterations on the NYUv2 dataset, we lose only 4.8% of the δ1 accuracy in exchange for 90% energy saving (2.44 mJ/frame). Finally, our adaptive method marks only 5.6% lower δ1 accuracy than MobileDepth despite using 9× fewer parameters. Compared with a state-of-the-art model also running on GAP9, our method improves δ1 accuracy by 31.8% thanks to our adaptive sensor fusion while operating within the same average power budget (∼400 mW)
Abstract
Autonomous nano-drones, powered by vision-based tiny machine learning (TinyML) models, are a novel technology gaining momentum thanks to their broad applicability and pushing scientific advancement on resource-limited embedded systems. Their small form factor, i.e., a few tens of grams, severely limits their onboard computational resources to sub-100mW microcontroller units (MCUs). The Bitcraze Crazyflie nano-drone is the de facto standard, offering a rich set of programmable MCUs for low-level control, multi-core processing, and radio transmission. However, roboticists very often underutilize these onboard precious resources due to the absence of a simple yet efficient software layer capable of time-optimal pipelining of multi-buffer image acquisition, multi-core computation, intra-MCUs data exchange, and Wi-Fi streaming, leading to sub-optimal control performances. Our NanoCockpit framework aims to fill this gap, increasing the throughput and minimizing the system's latency, while simplifying the developer experience through coroutine-based multi-tasking. In-field experiments on three real-world TinyML nanorobotics applications show our framework achieves ideal end-to-end latency, i.e. zero overhead due to serialized tasks, delivering quantifiable improvements in closed-loop control performance (-30% mean position error, mission success rate increased from 40% to 100%).
Abstract
A key task in embedded vision is visual odometry (VO), which estimates camera motion from visual sensors, and it is a core component in many embedded power-constrained systems, from autonomous robots to augmented and virtual reality wearable devices. The newest class of VO systems combines deep learning models with bio-inspired event-based cameras, which are robust to motion blur and lighting conditions. However, state-of-the-art (SoA) event-based VO algorithms require significant memory and computation. For example, the leading approach DEVO requires 733 MB of memory and 155 billion multiply-accumulate (MAC) operations per frame. We present TinyDEVO, an event-based VO deep learning model designed for resource-constrained microcontroller units (MCUs). We deploy TinyDEVO on an ultra-low-power (ULP) 9-core RISC-V-based MCU, achieving a throughput of approximately 1.2 frames per second with an average power consumption of only 86 mW. Thanks to our neural network architectural optimizations and hyperparameter tuning, TinyDEVO reduces the memory footprint by 11.5x (to 63.8 MB) and the number of operations per frame by 29.7x (to 5.2 billion MACs per frame) compared to DEVO, while maintaining an average trajectory error of 27 cm, i.e., only 19 cm higher than DEVO, on three state-of-the-art datasets. Our work demonstrates, for the first time, the feasibility of an event-based VO pipeline on ultra-low-power devices.
Abstract
Sub-30g nano-sized aerial robots can leverage their agility and form factor to autonomously explore cluttered and narrow environments, like in industrial inspection and search and rescue missions. However, the price for their tiny size is a strong limit in their resources, i.e., sub-100 mW microcontroller units (MCUs) delivering ∼100 GOps/s at best, and memory budgets well below 100 MB. Despite these strict constraints, we aim to enable complex vision-based tasks aboard nano-drones, such as dense 3D scene reconstruction: a key robotic task underlying fundamental capabilities like spatial awareness and motion planning. Top-performing 3D reconstruction methods leverage neural radiance fields (NeRF) models, which require GBs of memory and massive computation, usually delivered by high-end GPUs consuming 100s of Watts. Our work introduces Tiny-DroNeRF, a lightweight NeRF model, based on Instant-NGP, and optimized for running on a GAP9 ultra-low-power (ULP) MCU aboard our nano-drones. Then, we further empower our Tiny-DroNeRF by leveraging a collaborative federated learning scheme, which distributes the model training among multiple nano-drones. Our experimental results show a 96% reduction in Tiny-DroNeRF's memory footprint compared to Instant-NGP, with only a 5.7 dB drop in reconstruction accuracy. Finally, our federated learning scheme allows Tiny-DroNeRF to train with an amount of data otherwise impossible to keep in a single drone's memory, increasing the overall reconstruction accuracy. Ultimately, our work combines, for the first time, NeRF training on an ULP MCU with federated learning on nano-drones.
Abstract
Monocular depth estimation (MDE) plays a crucial role in enabling spatially-aware applications in Ultra-low-power (ULP) Internet-of-Things (IoT) platforms. However, the limited number of parameters of Deep Neural Networks for the MDE task, designed for IoT nodes, results in severe accuracy drops when the sensor data observed in the field shifts significantly from the training dataset. To address this domain shift problem, we present a multi-modal On-Device Learning (ODL) technique, deployed on an IoT device integrating a Greenwaves GAP9 MicroController Unit (MCU), a 80 mW monocular camera and a 8 x 8 pixel depth sensor, consuming ≈300mW. In its normal operation, this setup feeds a tiny 107 k-parameter μPyD-Net model with monocular images for inference. The depth sensor, usually deactivated to minimize energy consumption, is only activated alongside the camera to collect pseudo-labels when the system is placed in a new environment. Then, the fine-tuning task is performed entirely on the MCU, using the new data. To optimize our backpropagation-based on-device training, we introduce a novel memory-driven sparse update scheme, which minimizes the fine-tuning memory to 1.2 MB, 2.2x less than a full update, while preserving accuracy (i.e., only 2% and 1.5% drops on the KITTI and NYUv2 datasets). Our in-field tests demonstrate, for the first time, that ODL for MDE can be performed in 17.8 minutes on the IoT node, reducing the root mean squared error from 4.9 to 0.6m with only 3 k self-labeled samples, collected in a real-life deployment scenario.