This article presents a machine learning-based technique to build a predictive model and generate rules of action to allow autonomous vehicles to perform roundabout maneuvers. The approach consists of building a predictive model of vehicle speeds and steering angles based on collected data related to driver–vehicle interactions and other aggregated data intrinsic to the traffic environment, such as roundabout geometry and the number of lanes obtained from Open-Street-Maps and offline video processing. The study systematically generates rules of action regarding the vehicle speed and steering angle required for autonomous vehicles to achieve complete roundabout maneuvers. Supervised learning algorithms like the support vector machine, linear regression, and deep learning are used to form the predictive models.
Among the reasons for traffic accidents, distractions are the most common. Although there are many traffic signs on the road that contribute to safety, variable message signs (VMSs) require special attention, which is transformed into distraction. ADAS (advanced driver assistance system) devices are advanced systems that perceive the environment and provide assistance to the driver for his comfort or safety. This project aims to develop a prototype of a VMS (variable message sign) reading system using machine learning techniques, which are still not used, especially in this aspect. The assistant consists of two parts: a first one that recognizes the signal on the street and another one that extracts its text and transforms it into speech. For the first one, a set of images were labeled in PASCAL VOC format by manual annotations, scraping and data augmentation. With this dataset, the VMS recognition model was trained, a RetinaNet based off of ResNet50 pretrained on the dataset COCO. Firstly, in the reading process, the images were preprocessed and binarized to achieve the best possible quality. Finally, the extraction was done by the Tesseract OCR model in its 4.0 version, and the speech was done by the cloud service of IBM Watson Text to Speech.
Advancements in autonomous driving have seen unprecedented improvement in recent years. This work addresses the challenge of enhancing the navigation of autonomous vehicles in complex urban environments such as intersections and roundabouts through the integration of computer vision and unmanned aerial vehicles (UAVs). UAVs, owing to their aerial perspective, offer a more effective means of detecting vehicles involved in these maneuvers. The primary objective is to develop, evaluate, and compare different computer vision models and reduced-board (and small-power) hardware for optimizing traffic management in these scenarios. A dataset was constructed using two sources, several models (YOLO 5 and 8, DETR, and EfficientDetLite) were selected and trained, four reduced-board computers were chosen (Raspberry Pi 3B+ and 4, Jetson Nano, and Google Coral), and the models were tested on these boards for edge computing in UAVs. The experiments considered training times (with the dataset and its optimized version), model metrics were obtained, inference frames per second (FPS) were measured, and energy consumption was quantified. After the experiments, it was observed that the combination that best suits our use case is the YoloV8 model with the Jetson Nano. On the other hand, a combination with much higher inference speed but lower accuracy involves the EfficientDetLite models with the Google Coral board.
This publication presents a dataset of Spanish roundabouts aerial images taken from a UAV, along with annotations in PASCAL VOC XML files that indicate the position of vehicles within them. Additionally, a CSV file is attached containing information related to the location and characteristics of the captured roundabouts. This work details the process followed to obtain them: image capture, processing, and labeling. The dataset consists of 985,260 total instances: 947,400 cars, 19,596 cycles, 2262 trucks, 7008 buses, and 2208 empty roundabouts in 61,896 1920 × 1080 px JPG images. These are divided into 15,474 extracted images from 8 roundabouts with different traffic flows and 46,422 images created using data augmentation techniques. The purpose of this dataset is to help research into computer vision on the road, as such labeled images are not abundant. It can be used to train supervised learning models, such as convolutional neural networks, which are very popular in object detection.
This paper explores the application of large language models (LLMs) to automate the evaluation of programming assignments in an undergraduate “Introduction to Programming” course. This study addresses the challenges of manual grading, including time constraints and potential inconsistencies, by proposing a system that integrates several LLMs to streamline the assessment process. The system utilizes a graphic interface to process student submissions, allowing instructors to select an LLM and customize the grading rubric. A comparative analysis, using LLMs from OpenAI, Google, DeepSeek and ALIBABA to evaluate student code submissions, revealed a strong correlation between LLM-generated grades and those assigned by human instructors. Specifically, the reduced model using statistically significant variables demonstrates a high explanatory power, with an adjusted R2 of 0.9156 and a Mean Absolute Error of 0.4579, indicating that LLMs can effectively replicate human grading. The findings suggest that LLMs can automate grading when paired with human oversight, drastically reducing the instructor workload, transforming a task estimated to take more than 300 h of manual work into less than 15 min of automated processing and improving the efficiency and consistency of assessment in computer science education.
This article presents a remote management architecture of an unmanned aerial vehicles (UAVs) fleet to aid in the management of solar power plants and object tracking. The proposed system is a competitive advantage for sola r energy production plants, due to the reduction in costs for maintenance, surveillance, and security tasks, especially in large solar farms. This new approach consists of creating a hardware and software architecture that allows for performing different tasks automatically, as well as remotely using fleets of UAVs. The entire system, composed of the aircraft, the servers, communication networks, and the processing center, as well as the interfaces for accessing the services via the web, has been designed for this specific purpose. Image processing and automated remote control of the UAV allow generating autonomous missions for the inspection of defects in solar panels, saving costs compared to traditional manual inspection. Another application of this architecture related to security is the detection and tracking of pedestrians and vehicles, both for road safety and for surveillance and security issues of solar plants. The novelty of this system with respect to current systems is summarized in that all the software and hardware elements that allow the inspection of solar panels, surveillance, and people counting, as well as traffic management tasks, have been defined and detailed. The modular system presented allows the exchange of different specific vision modules for each task to be carried out. Finally, unlike other systems, calibrated fixed cameras are used in addition to the cameras embedded in the drones of the fleet, which complement the system with vision algorithms based on deep learning for identification, surveillance, and inspection.