arXiv subjects

Pedro Machado

Publications and source records attributed to Pedro Machado.

At least 19 recordsLinked to original sources

arXiv2026

Detection of hydrocarbons in Titan using high-resolution cross-correlation spectroscopy

High-resolution cross-correlation spectroscopy (HRCCS) is a powerful technique for detecting molecules whose individual spectral lines are too weak to be identified directly, but its sensitivity is limited by the availability and quality of high-resolution opacity data. Many molecules of atmospheric and astrobiological interest lack complete line lists, restricting traditional template-based searches. In this work, we use Titan as a controlled testbed to develop and validate a new cross-section-based methodology for HRCCS template construction. We analyse K-band CRIRES+ observations of Titan (1.99 - 2.48 {\mu}m), and compute cross-correlation functions using both line-by-line and cross-section-based templates. Our analysis recovers known hydrocarbons such as methane (CH4) and acetylene (C2H2), and yields the first HRCCS detection of ethane (C2H6) with a significance of SNRpeak = 5.17 {\pm} 0.07. The ethane detection was made possible exclusively through cross-section-based templates, as no high-resolution line list currently exists for this molecule. These results demonstrate that cross-section-based template construction is a practical and powerful strategy for extending HRCCS to molecules that currently lack reliable line lists, and establish Titan as a benchmark for calibrating molecular detection techniques that can be applied to both solar system and exoplanet atmospheres. Future applications of this approach to other ground-based high-resolution spectrographs, as well as to JWST's highest-resolution modes and next-generation facilities such as the ELT, could significantly expand the inventory of molecules detectable in planetary atmospheres.

astro-ph.EP↗

arXiv2026

TriSAR: Task Coordination and Collision Avoidance for Aerial Robot Teams in Disaster Response

Multi-Unmanned Aerial Vehicle (UAV) disaster-response systems require coordinated task assignment and local trajectory control, yet the individual and combined contributions of these coordination layers to mission efficiency and operational safety remain insufficiently characterised under controlled experimental conditions. TriSAR is evaluated as a five-UAV coordination system operating in a physics-based Gazebo simulation of an earthquake-damaged urban environment. A 2 x 2 factorial design compares two task-allocation strategies (Genetic Algorithm and greedy fitness-based allocation) with reactive collision avoidance enabled or disabled. Each of the four configurations was evaluated over 30 stochastic episodes in a common scenario of five UAVs and eight targets. Under greedy allocation, enabling repulsion eliminated recorded collision-threshold violations, confirmed by a Mann-Whitney test (U = 885, p = 4.03 x 10^-12, rank-biserial r = 0.97). Under GA allocation, the same protective effect was confirmed (U = 675, p = 1.26 x 10^-5, rank-biserial r = 0.50). For mission-efficiency metrics, GA-based allocation showed no statistically detectable advantage over greedy allocation when repulsion was enabled, but a significant advantage in steps, path length, and energy when repulsion was disabled (Welch's t-tests, |g| between 0.92 and 1.76). These results show that reactive repulsion provides a substantial, allocation-dependent safety benefit, while the additional computational complexity of GA-based task allocation yields a detectable mission-efficiency benefit only when repulsion is disabled.

cs.RO↗

arXiv2026

From Thermal Preference Prediction to Adaptive Thermal Intervention: A Reinforcement Learning Approach Using Physiological and Environmental Sensing

Personalised thermal comfort is essential for occupant wellbeing and for the development of more responsive building-control strategies, yet conventional Heating, Ventilation, and Air Conditioning (HVAC) systems rely on static setpoints and population-level comfort models that fail to capture individual physiological variability. This paper presents a two-stage personalised thermal comfort approach integrating multimodal physiological and environmental sensing with reinforcement learning-based decision-making.

cs.LG↗

arXiv2026

An AI-Based Adaptive Learning Platform for Multilingual and Low-Resource Educational Contexts: A Case Study on Nigeria

Educational platforms in under-resourced and multilingual contexts, such as Nigeria, often struggle with limited personalisation, inadequate language support, and weak curriculum internationalisation, leading to reduced learner engagement and inclusivity. This paper presents an AI-based adaptive learning platform designed for multilingual and low-resource educational contexts, with a case study on Nigerian Pidgin English. The system integrates fine-tuned large language models (LLMs) within a personalised and adaptive learning (PAL) framework, addressing linguistic inclusivity and computational constraints in resource-limited environments. To enhance linguistic alignment, a curated Nigerian Pidgin corpus was developed and used to fine-tune an instruction-tuned LLM. The study further investigates model optimisation through multi-level quantisation (4-bit, 5-bit, and 8-bit), enabling systematic analysis of trade-offs between semantic fidelity and computational efficiency. Experimental evaluation combines automatic semantic metrics (BLEU, ROUGE-L, BERTScore, perplexity, lexical diversity) with human-centred cultural assessment conducted by native speakers. Results demonstrate that higher-bit quantisation improves semantic preservation and structural coherence, while lower-bit models offer reduced inference latency with minimal degradation in instructional quality. The findings establish a deployable, resource-aware intelligent learning system that balances semantic robustness, cultural relevance, and computational efficiency. This work contributes an experimentally validated framework for adapting large language models to low-resource languages while maintaining practical feasibility for scalable educational deployment.

cs.CY↗

arXiv2026

High Resolution Optical Methane Linelist from observations of Titan for Cross-Correlation studies

Exoplanet atmosphere characterization heavily relies on molecular spectroscopic data. Despite efforts to obtain comprehensive spectral libraries for the chemical characterization of exoplanet atmospheres, large gaps remain, particularly for larger molecules and higher frequencies at high spectral resolution. One key example is the methane (CH4) optical spectrum. CH4, the simplest hydrocarbon, is a crucial species for exoplanet atmosphere characterization and a possible biosignature. However, until now, high-resolution linelists at optical wavelengths for CH4 have been very challenging to obtain either experimentally or computationally, leaving the high resolution spectrum of CH4 uncharacterised across most of the visible spectrum. This restricts exploration of CH4 absorption in the optical regime, as upcoming instruments such as ELT-ANDES and VLT-RISTRETTO will start probing the atmospheres of ever smaller exoplanets in optical wavelengths. To address this spectroscopic data limitation, we observed Titan's optical spectrum, dominated by CH4 absorption, at the highest spectral resolution to date with VLT-ESPRESSO. From it, we produced an empirical, low-temperature high-resolution (R ~ 190000) linelist of CH4 in optical wavelengths which we present here, with thousands of previously unidentified lines. We employ this CH4 linelist (RRS-2026) to build a template suitable for high resolution cross-correlation spectroscopy (HRCCS) studies, a first for CH4 in optical wavelengths. With this new linelist, we performed the first HRCCS detection of CH4 in the atmospheres of Titan and Jupiter using optical high resolution spectra. This work sets the stage for the search for CH4 in exoplanet atmospheres through HRCCS with current and future ground-based high-resolution optical spectrographs, showcasing how Solar System observations provide useful products for exoplanet research.

astro-ph.EP↗

arXiv2026

Assessing the Impact of Varying HSO Cross Sections on Photochemical Models: Implications for the Spectral Characterization of Terrestrial Exoplanets

Characterization of exoplanet atmospheres requires a close interplay between observations, modelling and experimental data. The accuracy of input data used in atmospheric models is essential, as it impacts our interpretation of planetary spectra with retrieval codes. Molecular absorption cross sections in the Ultraviolet-Visible range are fundamental input parameters, determining chemical kinetics, particularly on temperate terrestrial planets. However, several atmospheric species remain poorly, or even entirely uncharacterised. This is the case for HSO, a radical with unconstrained photolysis cross sections, often approximated by hydroperoxyl (HO2). Sulphur chemistry can strongly influence the composition of rocky exoplanets, particularly in anoxic environments where volcanic SO2 -- the main source of HSO -- is more efficiently photolysed, and sulphur aerosols like S8 can form. HSO photolysis contributes to the formation of such sulphur chains, which have been proposed as indirect signatures of volcanic outgassing. Assessing the sensitivity of photochemical models to different UV-Visible cross-section prescriptions for HSO is therefore important for guiding its prioritization among poorly characterised atmospheric species. Here, we derive an updated HSO cross-section prescription from simulated HSO2 data, providing a more reliable representation of HSO photolysis than HO2. We compare results from these new cross sections to the default prescription for temperate terrestrial planets with Archean-like atmospheres. We find that our updated HSO cross-section prescription enhances aerosol scattering and absorption signatures in transmission, emission and reflection spectra for planets orbiting G- and K-type stars.

astro-ph.EP↗

arXiv2026

EmotionAI: A Privacy-Preserving Computational Intelligence Pipeline for Speech-Emotion-Grounded Conversational Analysis

Reviewing recorded interviews for affective cues such as composure and agitation is slow and subjective, and cloud services that could automate the task require sensitive audio to leave the device. EmotionAI is a fully local Computational Intelligence (CI) pipeline that couples Speech Emotion Recognition (SER) with generative reasoning. Speaker diarisation, Whisper Automatic Speech Recognition (ASR) and a wav2vec2 emotion classifier produce per-segment affective evidence, and an adversarial three-model local Large Language Model (LLM) panel turns that evidence into timestamp-grounded, citation-constrained answers. Zero-shot evaluation on the RAVDESS four-class English subset (n = 672) measures the cost of cross-corpus transfer: the deployed classifier scores 48.8% accuracy, above random (24.9%) and majority (28.6%) baselines but below an in-domain MFCC + logistic-regression comparator (71.0%). The complete pipeline runs in a mean 157 s on CPU (real-time factor approximately 1.33) with zero external calls. The contribution is not state-of-the-art SER but an auditable, privacy-preserving integration of imperfect affective evidence into grounded conversational analysis.

cs.SD↗

arXiv2026

Detection of C3 in Titan with VLT-ESPRESSO

Titan is regarded as a natural laboratory in the Solar System for studying atmospheric photochemistry and the abiotic production of organic molecules on cold small exoplanets. Since the end of the Cassini-Huygens mission, telescope observations have enabled new detections of increasingly complex carbon-based molecules at infrared and sub-millimetre wavelengths, while the optical regime has been largely overlooked. Following a recent tentative detection of the 405 nm absorption band of C3 in Titan in archived optical VLT UVES spectra at resolving power R = 60000, this work reports an eight sigma detection of the C3 405 nm absorption band in Titan using dedicated ultra high resolution VLT ESPRESSO observations at R = 190000, the highest spectral resolution optical observations of Titan to date. The VLT ESPRESSO spectrum is compared to model spectra of Titan with varying C3 abundances. A chi squared analysis is used to assess the agreement between non solar spectral features and C3 absorption as the C3 abundance is varied, and a Bayesian Markov Chain Monte Carlo fit between model and observed spectra is performed. The chi squared analysis yields an eight sigma detection of C3, consistent with a C3 column density of approximately 1.5E13 cm-2, while the MCMC fit retrieves a C3 column density of 1.47E13 cm-2 at five sigma. These values are consistent with the order of magnitude predicted by photochemical models, which reach parts per million levels in the Titan mesosphere. This work demonstrates the usefulness of instruments and techniques originally developed for exoplanet research when applied to Solar System targets.

astro-ph.EP↗

arXiv2025

Sex and age determination in European lobsters using AI-Enhanced bioacoustics

Monitoring aquatic species, especially elusive ones like lobsters, presents challenges. This study focuses on Homarus gammarus (European lobster), a key species for fisheries and aquaculture, and leverages non-invasive Passive Acoustic Monitoring (PAM). Understanding lobster habitats, welfare, reproduction, sex, and age is crucial for management and conservation. While bioacoustic emissions have classified various aquatic species using Artificial Intelligence (AI) models, this research specifically uses H. gammarus bioacoustics (buzzing/carapace vibrations) to classify lobsters by age (juvenile/adult) and sex (male/female). The dataset was collected at Johnshaven, Scotland, using hydrophones in concrete tanks. We explored the efficacy of Deep Learning (DL) models (1D-CNN, 1D-DCNN) and six Machine Learning (ML) models (SVM, k-NN, Naive Bayes, Random Forest, XGBoost, MLP). Mel-frequency cepstral coefficients (MFCCs) were used as features. For age classification (adult vs. juvenile), most models achieved over 97% accuracy (Naive Bayes: 91.31%). For sex classification, all models except Naive Bayes surpassed 93.23%. These strong results demonstrate the potential of supervised ML and DL to extract age- and sex-related features from lobster sounds. This research offers a promising non-invasive PAM approach for lobster conservation, detection, and management in aquaculture and fisheries, enabling real-world edge computing applications for underwater species.

cs.LG↗

arXiv2025

SteganoSNN: SNN-Based Audio-in-Image Steganography with Encryption

Secure data hiding remains a fundamental challenge in digital communication, requiring a careful balance between computational efficiency and perceptual transparency. The balance between security and performance is increasingly fragile with the emergence of generative AI systems capable of autonomously generating and optimising sophisticated cryptanalysis and steganalysis algorithms, thereby accelerating the exposure of vulnerabilities in conventional data-hiding schemes. This work introduces SteganoSNN, a neuromorphic steganographic framework that exploits spiking neural networks (SNNs) to achieve secure, low-power, and high-capacity multimedia data hiding. Digitised audio samples are converted into spike trains using leaky integrate-and-fire (LIF) neurons, encrypted via a modulo-based mapping scheme, and embedded into the least significant bits of RGBA image channels using a dithering mechanism to minimise perceptual distortion. Implemented in Python using NEST and realised on a PYNQ-Z2 FPGA, SteganoSNN attains real-time operation with an embedding capacity of 8 bits per pixel. Experimental evaluations on the DIV2K 2017 dataset demonstrate image fidelity between 40.4 dB and 41.35 dB in PSNR and SSIM values consistently above 0.97, surpassing SteganoGAN in computational efficiency and robustness. SteganoSNN establishes a foundation for neuromorphic steganography, enabling secure, energy-efficient communication for Edge-AI, IoT, and biomedical applications.

cs.CR↗

arXiv2025

Privacy-Preserving Spiking Neural Networks: A Deep Dive into Encryption Parameter Optimisation

Deep learning is widely applied to modern problems through neural networks, but the growing computational and energy demands of these models have driven interest in more efficient approaches. Spiking Neural Networks (SNNs), the third generation of neural networks, mimic the brain's event-driven behaviour, offering improved performance and reduced power use. At the same time, concerns about data privacy during cloud-based model execution have led to the adoption of cryptographic methods. This article introduces BioEncryptSNN, a spiking neural network based encryption-decryption framework for secure and noise-resilient data protection. Unlike conventional algorithms, BioEncryptSNN converts ciphertext into spike trains and exploits temporal neural dynamics to model encryption and decryption, optimising parameters such as key length, spike timing, and synaptic connectivity. Benchmarked against AES-128, RSA-2048, and DES, BioEncryptSNN preserved data integrity while achieving up to 4.1x faster encryption and decryption than PyCryptodome's AES implementation. The framework demonstrates scalability and adaptability across symmetric and asymmetric ciphers, positioning SNNs as a promising direction for secure, energy-efficient computing.

cs.CR↗

arXiv2025

Single pion-production and pion propagation in Achilles

We extend the applicability of Achilles (A CHIcagoLand Lepton Event Simulator) by incorporating the single-pion production mechanism in a fully exclusive fashion. The electroweak interaction vertex is modeled by combining the state-of-the-art Dynamical Coupled-Channels approach with realistic hole spectral functions, which account for correlations in both the initial target state and the residual spectator system. Final-state interactions are treated using a semi-classical intranuclear cascade that leverages nuclear configurations to determine the correlated spatial distribution of protons and neutrons. The meson-baryon scattering amplitudes used in the cascade are computed within the Dynamical Coupled-Channels framework, consistent with the electroweak vertex. To model pion absorption, we employ the optical potential approach of Oset and Salcedo. As an alternative approach, we explicitly model the production and propagation of resonances which mediate pion-nucleon scattering and pion absorption. We validate out approach against pion-nucleon and pion-nucleus scattering data, and present comparisons with electron- and neutrino-nucleus measurements from e4$\nu$, T2K, MINER$\nu$A, and MicroBooNE.

hep-ph↗

arXiv2025

Searching for neutrino polarizability at DUNE

We perform an initial study of DUNE's sensitivity to enhanced neutrino polarizability within models of light scalar mediators. We identify two possible signatures of polarizability due to neutrino scattering on electrons and due to coherent scattering on argon nuclei. These result in either one or two separated electromagnetic showers, respectively, with no associated hadronic activity. For each signature we compute the signal rates and the relevant backgrounds, obtaining the projected reach of the DUNE near detector. We then compare this with the current astrophysical and terrestrial bounds on light scalar models coupling to neutrinos and/or photons.

hep-ph↗

arXiv2025

Bearded Dragon Activity Recognition Pipeline: An AI-Based Approach to Behavioural Monitoring

Traditional monitoring of bearded dragon (Pogona Viticeps) behaviour is time-consuming and prone to errors. This project introduces an automated system for real-time video analysis, using You Only Look Once (YOLO) object detection models to identify two key behaviours: basking and hunting. We trained five YOLO variants (v5, v7, v8, v11, v12) on a custom, publicly available dataset of 1200 images, encompassing bearded dragons (600), heating lamps (500), and crickets (100). YOLOv8s was selected as the optimal model due to its superior balance of accuracy (mAP@0.5:0.95 = 0.855) and speed. The system processes video footage by extracting per-frame object coordinates, applying temporal interpolation for continuity, and using rule-based logic to classify specific behaviours. Basking detection proved reliable. However, hunting detection was less accurate, primarily due to weak cricket detection (mAP@0.5 = 0.392). Future improvements will focus on enhancing cricket detection through expanded datasets or specialised small-object detectors. This automated system offers a scalable solution for monitoring reptile behaviour in controlled environments, significantly improving research efficiency and data quality.

cs.CV↗

arXiv2025

Dynamical origin of Theia, the last giant impactor on Earth

Cosmochemical studies have proposed that Earth accreted roughly 5-10% of its mass from carbonaceous (CC) material, with a large fraction delivered late via its final impactor, Theia (the Moon-forming impactor). Here, we evaluate this idea using dynamical simulations of terrestrial planet formation, starting from a standard setup with a population of planetary embryos and planetesimals laid out in a ring centered between Venus and Earth's orbits, and also including a population of CC planetesimals and planetary embryos scattered inward by Jupiter. We find that this scenario can match a large number of constraints, including i) the terrestrial planets' masses and orbits; ii) the CC mass fraction of Earth; iii) the much lower CC mass fraction of Mars, as long as Mars only accreted CC planetesimals (but no CC embryos); iv) the timing of the last giant (Moon-forming) impact; and v) a late accretion phase dominated by non-carbonaceous (NC) bodies. For this scenario to work, the total mass in scattered CC objects must have been ~ 0.2 - 0.3 M$_{\oplus}$ , with an embryo-to-planetesimal mass ratio of at least 8, and CC embryos in the ~ 0.01 - 0.05 M$_{\oplus}$ mass range. In that case, our simulations show there are roughly 50-50 odds of Earth's last giant impactor (Theia) having been a carbonaceous object - either a pure CC embryo or an NC embryo that previously accreted a CC embryo. Our simulations thus provide dynamical validation of cosmochemical studies.

astro-ph.EP↗

arXiv2025

Neutrino Physics and Astrophysics at Colliders

Nonzero neutrino masses guarantee new physics and neutrinos are excellent probes of extreme environments in the Universe. The recent collider neutrino experimental program, including FASER$\nu$ and SND@LHC, along with the planned Forward Physics Facility at the High-Luminosity Large Hadron Collider, is opening a new window into neutrino physics and astrophysics. In this article, we review recent achievements and prospects of collider neutrino experiments, including key achievements such as the first measurements of collider neutrino interactions at unprecedented energies and the exploration of new physics scenarios, like dark matter candidates, sterile neutrinos, and non-standard neutrino interactions. For concreteness, we will focus on the significant scientific opportunities presented by the Forward Physics Facility, which will enable precision measurements of neutrino cross sections and proton structure at low parton momentum fraction. Furthermore, collider neutrino studies will substantially reduce systematic uncertainties in calculating atmospheric neutrino fluxes, thereby improving astrophysical neutrino observations as well as advancing our understanding of cosmic-ray interactions.

hep-ph↗

arXiv2025

Clash of the Titans: ultra-high energy KM3NeT event versus IceCube data

KM3NeT has reported the detection of a remarkably high-energy through-going muon. Lighting up about a third of the detector, this muon likely originated from a neutrino exceeding 10 PeV in energy. The crucial question we need to answer is where this event comes from and what its source is. Intriguingly, IceCube has been operating with a much larger effective area for a considerably longer time, yet it has not reported neutrinos above 10~PeV. We quantify the tension between the KM3NeT event and the absence of similar high-energy events in IceCube. Through a detailed analysis, we determine the most likely neutrino energy to be in the range of 23 - 2400 PeV. We find a $3.5\sigma$ tension between the two experiments, assuming the neutrino is from the diffuse isotropic neutrino flux. Alternatively, assuming the event is of cosmogenic origin and considering three representative models, this tension still falls within 3.1 - 3.6$\sigma$. The least disfavored scenario is a steady or transient point source, though still leading to $2.9\sigma$ and $2.0\sigma$ tensions, respectively. The lack of observation of high-energy events in IceCube seriously challenges the explanation of this event coming from any known diffuse fluxes. Our results indicate the KM3NeT event is likely the first observation of a new astrophysical source.

astro-ph.HE↗

arXiv2024

Does the Sun have a Dark Disk?

The Sun is not quite a perfect sphere, and its oblateness, thought to be induced through its rotation, has been measured using optical observations of its radius. Its gravitational quadrupole moment can then be deduced using solar models, or through helioseismology, and it can also be determined from measurements of its gravitational effects on Mercury's orbit. The various assessments do not appear to agree, with the most complete and precise orbital assessments being in slight excess of other determinations. This may speak to the existence of a non-luminous disk or ring, where we also note evidence for a circumsolar dust ring within Mercury's orbit from the Solar TErrestrial RElations Observatory (STEREO) mission. Historically, too, a protoplanetary disk may have been key to reconciling the Sun's metallicity with its neutrino yield. The distribution of the non-luminous mass within Mercury's orbit can modify the relative size of the optical and orbital quadrupole moments in different ways. We develop how we can use these findings to limit the mass of a dark disk, ring, or halo in the immediate vicinity of the Sun, and we note how future observational studies of the inner solar system can not only refine these constraints but also help to identify and to assess the mass of its dark-matter component.

hep-ph↗