Open-source System Research Articles

A Survey of Open-Source Autonomous Driving Systems and Their Impact on Research

A Survey of Open-Source Autonomous Driving Systems and Their Impact on Research

Tradutor: Building a Variety Specific Translation Model

Language models have become foundational to many widely used systems. However, these seemingly advantageous models are double-edged swords. While they excel in tasks related to resource-rich languages like English, they often lose the fine nuances of language forms, dialects, and varieties that are inherent to languages spoken in multiple regions of the world. Languages like European Portuguese are neglected in favor of their more popular counterpart, Brazilian Portuguese, leading to suboptimal performance in various linguistic tasks. To address this gap, we introduce the first open-source translation model specifically tailored for European Portuguese, along with a novel dataset specifically designed for this task. Results from automatic evaluations on two benchmark datasets demonstrate that our best model surpasses existing open-source translation systems for Portuguese and approaches the performance of industry-leading closed-source systems for European Portuguese. By making our dataset, models, and code publicly available, we aim to support and encourage further research, fostering advancements in the representation of underrepresented language varieties.

Predicting forest above-ground biomass using SAR imagery and GEDI data through machine learning in GEE cloud

The estimation of Forest above-ground biomass (AGB) is critical for comprehending forest ecosystems and promoting biodiversity restoration. The study was conducted to develop an effective approach to predict Forest Above-Ground Biomass (AGB) using Machine Learning, Image Classification, and GEE open-source fast processing system in the Similipal Tiger Reserve (STR), India. The study utilized six machine learning models and integrated various data sources, including Sentinel-1 and Sentinel-2 imagery, forest canopy height data from NASA’s GEDI Global Ecosystems Dynamics Investigation-LiDAR, geo-environmental Shuttle Radar Topography Mission (SRTM) data, and Climate Hazards Group Infrared precipitation with Station (CHIRPS) data. Sentinel-based optical and Synthetic Aperture Radar (SAR) signatures were also extracted before and after the monsoon season to evaluate AGB in a subtropical region. The Random forest-based Boruta method was used to examine the importance of multiple factors contributing to the prediction’s accuracy. In addition, the assessment of multicollinearity, which is accomplished by measuring the variance inflation factor (VIF), was carried out to address the issue of interrelatedness among variables that could affect the accuracy of the AGB mapping. The Random Forest model exhibited superior accuracy compared to other models, achieving an R2 of 0.71, MAE of 16.12Mg/ha, RMSD of 22.27Mg/ha, NRMSD of 0.212, and AUROC of 75%. The scatterplot analysis revealed a positive correlation between forest biomass and factors, such as forest canopy height, elevation, normalized differential vegetation index, normalized differential moisture index, and the ratio of VV/VH after the monsoon season. The VIF values ranged between 1.12 and 8.73, with VH_Jan_Mar having the maximum VIF and forest canopy height having the minimum VIF. As per the Boruta algorithm, 16 attributes were deemed important, while 14 attributes had less influence on AGB. The study presented a novel approach for estimating biomass in subtropical regions using remote sensing data set and machine learning models in Google platform. These results can be successfully used by the planners of STR for monitoring variation in AGB ensuring better habitat for wild animals.

Easy scalable, low-cost open-source magnetic field detection system for evaluating low-field MRI magnets using a motion-tracked robot

Abstract Objective Low-field magnetic resonance imaging is currently developing into a valuable diagnostic tool due to its simplicity of magnet array designs. Particularly, this allows the development of scanners as part of educational workshops, thus ensuring knowledge transfer and empowering local scientists to design tailored solutions for specific local problems. To obtain the maximum performance, the magnet needs to be shimmed requiring an automated system measuring the spatial magnetic field distribution. Methods A self-designed measuring probe based on commercial integrated Hall sensor chips is used and optimized by calibrating it in an easy-to-build calibration system. For positioning of the sensor, a low-cost five-degree-of-freedom robot arm is used and improved by camera-based motion tracking for precise localization of the sensor. Results The system is able to map the field of a $$45\text{mT}$$ 45 mT -Halbach desktop MR magnet, as well as a self-designed x-gradient (used inside the magnet) with an efficiency of $$2\text{mT}/\text{m}/\text{A}$$ 2 mT / m / A . The built-up Hall sensor demonstrates a level of precision that is competitive with commercial sensors. The entire positioning system can be freely scaled to accommodate larger designs by adjusting the kinematics. Conclusion The presented system is demonstrated to be comparable to already established measurement systems, while the costs, setup times, and mapping duration are greatly reduced.

MODEL PROPOSAL FOR A PORTABLE ROOFTOP GREENHOUSE: IMPROVING MIDDLE SCHOOL STUDENTS’ ENVIRONMENTAL CONSCIOUSNESS AND ATTITUDES

Environmental education is crucial for building a sustainable society; however, students in metropolitan areas often lack spaces for active learning and experiencing the environment directly. To solve this problem, this study explores the use of portable rooftop greenhouses (RTGs) in environmental education, utilizing the open-source Wikihouse system. The hypothesis suggests that these RTGs will actively engage students and enhance their environmental awareness. According to findings, students in the experimental group showed significantly higher levels of environmental awareness, attitude and thinking compared to those in the control group. These findings indicate how portable greenhouses can be used as creative tools for delivering engaging and innovative types of environmental education. Future implementations of this project at various schools and across scales will help assess its long-term effects on students' ecological consciousness and attitudes.

Harnessing Open-Source Solutions: Insights From the First Open Systems Pharmacology (OSP) Community Conference.

In 2017, the free and open-source software Open Systems Pharmacology (OSP) was launched. Since then, OSP has evolved from a small community into a diverse network of stakeholders committed to advancing open-source solutions for model-informed drug development (MIDD). In this context, the first OSP Community Conference was hosted by Novartis in Basel, Switzerland, on October 7-8, 2024, which gathered over 100 attendees from more than 40 institutions. This perspective synthesizes key insights from the conference.

Simplifying software compliance: AI technologies in drafting technical documentation for the AI Act

The European AI Act has introduced specific technical documentation requirements for AI systems. Compliance with them is challenging due to the need for advanced knowledge of both legal and technical aspects, which is rare among software developers and legal professionals. Consequently, small and medium-sized enterprises may face high costs in meeting these requirements. In this study, we explore how contemporary AI technologies, including ChatGPT and an existing compliance tool (DoXpert), can aid software developers in creating technical documentation that complies with the AI Act. We specifically demonstrate how these AI tools can identify gaps in existing documentation according to the provisions of the AI Act. Using open-source high-risk AI systems as case studies, we collaborated with legal experts to evaluate how closely tool-generated assessments align with expert opinions. Findings show partial alignment, important issues with ChatGPT (3.5 and 4), and a moderate (and statistically significant) correlation between DoXpert and expert judgments, according to the Rank Biserial Correlation analysis. Nonetheless, these findings underscore the potential of AI to combine with human analysis and alleviate the compliance burden, supporting the broader goal of fostering responsible and transparent AI development under emerging regulatory frameworks.

PyPSA-Earth sector-coupled: A global open-source multi-energy system model showcased for hydrogen applications in countries of the Global South

PyPSA-Earth sector-coupled: A global open-source multi-energy system model showcased for hydrogen applications in countries of the Global South

Uncovering the applications, developments, and future research directions of the open-source energy modelling system (OSeMOSYS): A systematic literature review

Uncovering the applications, developments, and future research directions of the open-source energy modelling system (OSeMOSYS): A systematic literature review

A Decision Support System for Irrigation Scheduling Using a Reduced-Size Pan

An automatic, weight-based, small 20 cm diameter pan was used for real-time calculations of evaporation and precipitation in a semiarid environment. The water evaporated from the evaporimeter (EP) was found to be a significant predictor of evapotranspiration (ETO; r2 = 0.84), which was calculated with the Penman–Monteith (P-M) equation by retrieving climatic data from a weather station. The results revealed seasonal variations of the pan coefficient (KP; dimensionless), with a mean value estimated at 0.84 (±0.16). Validation of ETO measurements using a calibrated regression model (ETO = 0.831*EP + 0.025), against the P-M equation indicated a high correlation coefficient (r2 = 0.99, slope of the regression line of 0.9). The present paper evaluates and discusses the potential of using a reduced-size pan for real-time monitoring of water evaporation and precipitation, proposing an open-source irrigation decision support system.

Impact of AID on Glycemic Profile and Maternal/Neonatal Outcomes in Pregnancy: A Review of the Evidence From Observational Studies.

The mainstay of type 1 diabetes (T1D) management in pregnancy is optimization of glucose levels in a tight range. Achieving euglycemia has been revolutionized by advances in diabetes technology, including the development of automated insulin delivery (AID) systems. A small but growing population of gravidas with T1D elects to pursue off-label use of AID systems in pregnancy, and their outcomes have been described in numerous observational cohorts. This review aims to aggregate data from all available observational studies examining glycemic, maternal, and neonatal outcomes associated with antenatal AID use. A total of 243 pregnancies managed antenatally with AID were described in 24 publications, with largely reassuring outcomes data. Time in range (TIR) with commercial AID systems was generally acceptable, with many patients reaching pregnancy target TIR > 70% by the third trimester. Time in range with open-source AID systems appeared even higher, although with the potential tradeoff of worse time below range (TBR). Clinically, there do not appear to be major differences in pregnancy outcomes between AID systems and other methods of insulin delivery, although this assumption is based largely on indirect comparisons with other population-level reports rather than direct comparisons within analytic observational cohorts. Clinical outcomes appear superior with open-source AID compared with commercial AID, although this should be interpreted with caution based on the small sample size of this subpopulation (n = 16) and potential confounding. The real-world evidence generated by these observational studies provides invaluable information for patients and providers seeking to improve outcomes for gravidas with T1D.

Comparison of Margin Quality for Intersegmental Plan Identification in Pulmonary Segmentectomy.

Background and Objectives: Insufficient margin in lung cancer is associated with an increased locoregional recurrence rate. In pulmonary segmentectomy, two commonly used methods for identifying the intersegmental plane are inflation-deflation and indocyanine green dyeing. The aim of this study was to compare these two methods in terms of quality margins and to evaluate their superiority. Materials and Methods: A total of 63 patients who underwent segmentectomy via video-assisted thoracoscopic surgery (VATS) for pulmonary nodules and underwent preoperative planning with 3D modeling between October 2020 and February 2024 were included in this study. The location of the nodule and the distance to the intersegmental margins were virtually measured preoperatively using an open-source 3D modeling system. Patients were grouped according to the method of identifying the intersegmental margins. Group 1 included segmentectomies performed by the inflation-deflation method (n = 42), and Group 2 included segmentectomies performed by systemic indocyanine green (ICG) injection (n = 21). The area where the histopathological nodule was measured closest to the intersegmental margin was recorded. Values within (+/-10 mm) compared to the value measured in the three-dimensional model were considered successful. The obtained data were statistically compared between the groups. Results: There was no difference between the groups in terms of virtual and pathological margins. However, in terms of margin quality, the rate of deviation detected in the pathological margin compared to the measured virtual margin was significantly different between the groups (p = 0.04). Accordingly, the success rate was 64.3% in Group 1 and 90.5% in Group 2 (p = 0.05). In Group 1, the failure rate was highly against the adjacent parenchyma. There was no significant difference between the groups in the analysis of simple and complex segmentectomies. Conclusions: Intersegmental plane identification with indocyanine green increases the margin quality by defining resection margins closer to the virtual margins. In the inflation-deflation method, unnecessary parenchymal loss occurs due to disadvantages in identifying intersegmental margins.

IoT and AI for smart agriculture in resource-constrained environments: challenges, opportunities and solutions

Climate change is an undeniable reality with far-reaching and profound implications for agriculture and, subsequently, global food security. The highly vulnerable farming communities in underdeveloped and developing countries can overcome resource and capacity constraints with the help of technology, particularly the Internet of Things (IoT) and Artificial Intelligence (AI), making agriculture smarter and more resilient to climate change. This work aims to enrich existing research on smart agriculture by focusing on its applicability in resource-constrained environments. In this context, it presents a detailed overview of the technologies enabling smart agriculture, highlights the challenges and opportunities for farming communities in developing countries, and proposes a framework for climate change-resilient smart agriculture. The framework is loosely based on McKinsey’s 7S model for change management, consisting of hard and soft elements that are defined and adapted for the desired context. The hard elements include IoT sensors, network communications, and data management and analysis using AI, whereas the soft elements consist of policies and regulations, capacity building measures, and a supportive developmental ecosystem. This novel approach has not been employed before in this context. Furthermore, the framework’s efficacy for environmental and crop growth monitoring is demonstrated through its implementation in a low-cost, open-source IoT system within a greenhouse using Edge-Cloud architecture. Here raw, extracted, and derived features are monitored to estimate irrigation requirements and crop maturity date. We conclude with an analysis of the results, recommendations for implementing climate change-resilient smart agriculture in resource-constrained environments, and the identification of areas for future research.

Enabling Automatic Solar PV Array Identification using Big Satellite Imagery

Recently, there has been a growing interest in automatically collecting distributed solar photovoltaic (PV) installation information in smart grid systems, including the quantity and locations of solar PV deployments, as well as their profiling information across a given geospatial region. Most recent approaches are still suffering low detection accuracy due to insufficient sample and principal feature learning when building their models and also separation of rooftop object segmentation and identification during their detection processes. In addition, they cannot report accurate multi-deployment results. To address these problems, we design a new system-SolarDetector + , which can automatically and accurately detect, and profile distributed solar PV arrays without any extra cost. In essence, SolarDetector + first leverages multiple data augmentation techniques, including CycleGAN, Latent Diffusion Models, and Generative Adversarial networks, to build a large rooftop satellite imagery dataset (RSID). Then, SolarDetector + employs Mask R-CNN algorithm to accurately identify rooftop solar PV arrays and learn the detailed installation information for each solar PV array simultaneously. We find that pre-trained SolarDetector + yields an average Matthews correlation coefficient (MCC) of 0.862 to detect solar PV arrays over RSID, which is ∼ 50% better than the most recent open-source detection system—SolarFinder.

Low-Cost, Open-Source, High-Precision Pressure Controller for Multi-Channel Microfluidics.

Microfluidics is a technology that manipulates liquids on the scales ranging from microliters to femtoliters. Such low volumes require precise control over pressures that drive their flow into the microfluidic chips. This article describes a custom-built pressure controller for driving microfluidic chips. The pressure controller features piezoelectrically controlled pressure regulation valves. As an open-source system, it offers high customizability and allows users to modify almost every aspect. The cost is roughly a third of what similar, alternative, commercially available piezoelectrically controlled pressure regulators could be purchased for. The measured output pressure values of the device vary less than 0.7% from the device's reported pressure values when the requested pressure is between -380 and 380 mbar. Importantly, the output pressure the device creates fluctuates only ±0.2 mbar when the pressure is cycled between 10 and 500 mbar. The pressure reading accuracy and stability validation suggest that the device is highly feasible for many advanced (low-pressure) microfluidic applications. Here, we compare the main features of our device to commercially and non-commercially available alternatives and further demonstrate the device's performance and accessibility in successful microfluidic hydrodynamic focusing (MHF)-based synthesis of large unilamellar vesicles (LUVs).

OpenNavSense platform: A low-cost, open-source inertial navigation system for the evaluation of estimation algorithms.

OpenNavSense platform: A low-cost, open-source inertial navigation system for the evaluation of estimation algorithms.

Enhancing energy transition with open-source regional energy system optimization models: TEMOA-Piedmont

Enhancing energy transition with open-source regional energy system optimization models: TEMOA-Piedmont

Design and implementation of an internet of things automatic solar shading system

The implementation of energy-saving solutions becomes increasingly crucial as global energy reserves decline. The integration of automated solar shading systems not only improves living conditions, but also reduces energy costs. This work presents the design and implementation of an Internet of Things (IoT) automatic solar shading system. The developed solar shading includes custom-made louvers that are able to reflect the external light, while being transparent. Moreover, the system is equipped with a microcontroller and appropriate sensors that enable its automated operation based on measurements of the internal and external environmental conditions. A Raspberry Pi acts as a server, enabling the communication between the shading devices and the users, through an open-source home automation operating system (Home Assistant OS). The user-friendly interface, accessible via a web browser or a mobile application, provides essential data such as temperature, humidity, and device status. Alert notifications are sent when specific conditions are met. The overall system is enclosed in two 3D-printed units, ensuring its durability and easy integration into existing or new installations. In summary, this system combines the advantages of automatic solar shadings, including energy efficiency and improved occupant comfort, with smart features for remote control and monitoring through a user-friendly interface. The proposed system has been installed in a window of our laboratory, performing successfully in real-life conditions.

Distributed MobilityDB: A Scalable Moving Object Database Management System

As the volume and complexity of spatiotemporal data continue to expand rapidly across various domains such as urban planning, environmental monitoring, and logistics, the demand for comprehensive data management systems becomes increasingly urgent. Handling such data entails intricate topological and analytical operations, emphasizing the necessity for robust and adaptable solutions capable of addressing diverse user queries. This paper introduces Distributed MobilityDB 1 , an open-source system engineered to manage big spatiotemporal trajectory datasets within SQL environments. Distributed MobilityDB offers capabilities for scalable spatiotemporal data management, facilitating efficient distributed query processing while seamlessly integrating with existing MobilityDB SQL operations. Key contributions highlighted in the paper encompass an adaptive spatiotemporal SQL query engine. This engine channels user SQL queries through various planning strategies for optimizing the distributed query plan, then distributing the query execution across cluster nodes transparently to the user. Various spatiotemporal query types are supported for distribution, including range selections, and joins proximity. Distributed MobilityDB is implemented as an add-on extension to PostgreSQL, which facilitates installing it on a readily running server. The paper further presents extensive experiments conducted on both cloud and on-premise environments using both real and synthetic datasets, including AIS for ship trajectories and BerlinMOD for simulated person trips.

Interval Analysis-Based Optimization: A Robust Model for Intensity-Modulated Radiotherapy (IMRT).

Background: Cancer remains one of the leading causes of mortality worldwide, with radiotherapy playing a crucial role in its treatment. Intensity-modulated radiotherapy (IMRT) enables precise dose delivery to tumors while sparing healthy tissues. However, geometric uncertainties such as patient positioning errors and anatomical deformations can compromise treatment accuracy. Traditional methods use safety margins, which may lead to excessive irradiation of healthy organs or insufficient tumor coverage. Robust optimization techniques, such as minimax approaches, attempt to address these uncertainties but can result in overly conservative treatment plans. This study introduces an interval analysis-based optimization model for IMRT, offering a more flexible approach to uncertainty management. Methods: The proposed model represents geometric uncertainties using interval dose influence matrices and incorporates Bertoluzza's metric to balance tumor coverage and organ-at-risk (OAR) protection. The θ parameter allows controlled robustness modulation. The model was implemented in matRad, an open-source treatment planning system, and evaluated on five prostate cancer cases. Results were compared against traditional Planning Target Volume (PTV) and minimax robust optimization approaches. Results: The interval-based model improved tumor coverage by 5.8% while reducing bladder dose by 4.2% compared to PTV. In contrast, minimax robust optimization improved tumor coverage by 25.8% but increased bladder dose by 23.2%. The interval-based approach provided a better balance between tumor coverage and OAR protection, demonstrating its potential to enhance treatment effectiveness without excessive conservatism. Conclusions: This study presents a novel framework for IMRT planning that improves uncertainty management through interval analysis. By allowing adjustable robustness modulation, the proposed model enables more personalized and clinically adaptable treatment plans. These findings highlight the potential of interval analysis as a powerful tool for optimizing radiotherapy outcomes, balancing treatment efficacy and patient safety.