Tatiana Ortegon-Sarmiento

Driving in snowy conditions challenges both human drivers and autonomous systems. Snowfall and ice accumulation impair vehicle control and affect driver perception and performance. Road markings are often obscured, forcing drivers to rely on intuition and memory to stay in their lane, which can lead to encroachment into adjacent lanes or sidewalks. Current lane detectors assist in lane keeping, but their performance is compromised by visual disturbances such as ice reflection, snowflake movement, fog, and snow cover. Furthermore, testing these systems with users on actual snowy roads involves risks to driver safety, equipment integrity, and ethical compliance. This study presents a low-cost virtual reality simulation for evaluating winter lane detection in controlled and safe conditions from a human-in-the-loop perspective. Participants drove in a simulated snowy scenario with and without the detector while quantitative and qualitative variables were monitored. Results showed a 49.9% reduction in unintentional lane departures with the detector and significantly improved user experience, as measured by the UEQ-S (p = 0.023, Cohen’s d = 0.72). Participants also reported higher perceived safety, situational awareness, and confidence. These findings highlight the potential of vision-based lane detection systems adapted to winter environments and demonstrate the value of immersive simulations for user-centered testing of ADASs.

Diferentes avances han aparecido con la nueva era tecnológica y la transformación digital, destacándose los vehículos autónomos (VAs). La creciente visibilidad de estos ha dado paso a diversas investigaciones sobre la aceptación y percepción de esta tecnología, sin embargo, en muchas de estas, los participantes se basan únicamente en una creencia preconcebida ya que la mayoría no ha conducido o tenido contacto con este tipo de vehículo, debido a que los estudios de prueba con VAs reales son caros, peligrosos y poco éticos. La realidad virtual surge como una posible solución a este problema, ya que proporciona un entorno realista, seguro y controlado, en el que se pueden implementar diferentes situaciones viales y meteorológicas. En este artículo se presenta un diseño de investigación preliminar, basado en el Modelo de Aceptación de la Tecnología (TAM), que busca evaluar e inferir el nivel de aceptación de los VAs, así como el comportamiento de los usuarios a la hora de utilizarlos e interactuar con ellos. El estudio se basa en una simulación virtual inmersiva de un VA, donde el usuario es expuesto a una situación de riesgo en la que el vehículo es incapaz de reconocer el carril debido a dificultades meteorológicas.

Lane detection is an essential module for the safe navigation of autonomous vehicles (AVs). Estimating the vehicle’s position and trajectory on the road is critical; however, several environmental variables can affect this task. State-of-the-art lane detection methods utilize convolutional neural networks (CNNs) as feature extractors to obtain relevant features through training using multiple kernel layers. It makes them vulnerable to any statistical change in the input data or noise affecting the spatial characteristics. In this paper, we compare six different CNN architectures to analyze the effect of various adverse conditions, including harsh weather, illumination variations, and shadows/occlusions, on lane detection. Among all the aforementioned adverse conditions, harsh weather in general and snowy night conditions particularly affect the performance by a large margin. The average detection accuracy of the networks decreased by 75.2%, and the root mean square error (RMSE) increased by 301.1%. Overall, the results show a noticeable drop in the networks’ accuracy for all adverse conditions because the features’ stochastic distributions change for each state.

Antecedentes: La simulación de procesos es una herramienta útil en el diseño, análisis y evaluación de estos, facilitando así su optimización y monitoreo, y permitiendo la predicción de su comportamiento ante ciertas situaciones. Objetivo: Por lo anterior, en este artículo se presenta la simulación de un proceso cervecero, haciendo especial énfasis en los subprocesos de llenado, tapado y pasteurizado. Método: Se identificaron cada uno de los pasos de un proceso cervecero, y se caracterizaron los subprocesos de interés, definiendo cada una de las variables de entrada y salida. A continuación, se simularon los subprocesos haciendo uso del controlador lógico programable PLC S7-1500, y se supervisó el comportamiento de cada uno de estos mediante SIMATIC WinCC, de manera local en una pantalla HMI (Interfaz Hombre Máquina), y de manera remota en un ordenador. La dinámica del proceso se presenta en el software Automation Studio 6, el cual se conecta al PLC mediante el servidor OPC de IBH. Para observar el comportamiento del sistema ante ciertas situaciones de emergencia se implementaron alarmas. Resultados: La simulación realizada permite seguir paso a paso los subprocesos de interés, tanto de manera local como remota, permitiendo observar el comportamiento del sistema en condiciones normales o ante situaciones de emergencia a las cuales el simulador responde de manera inmediata. Conclusiones: Con base en los resultados obtenidos, se concluye que el simulador es una alternativa adecuada y de bajo costo que permite comprender, supervisar y predecir el comportamiento que tendría la planta real, en cuanto a los subprocesos de interés.Palabras Clave: Automatización, Control Automático, Controlador Lógico Programable, Diseño de Sistemas, Programación, Simulación de Procesos. Abstract Background: Process simulation is an useful tool in the design, analysis and evaluation of processes, facilitating its optimization and monitoring, and allowing the prediction of its behavior in certain situations. Objective: Therefore, this paper shows a simulation of a brewing process with special emphasis on the subprocesses of filling, capping and pasteurized. Method: Each of the steps of a brewing process was identified, and the subprocesses of interest were characterized, defining each of the input and output variables. For the simulation of the subprocesses, a programmable logic controller PLC S7-1500 was used, controlling and monitoring the behavior of each of these by SIMATIC WinCC, locally on an HMI (Human Machine Interface) screen and remotely on a computer. The dynamics of the process are displayed in the Automation Studio 6 software, which is connected to the PLC via the IBH OPC server. To observe the behavior of the system to certain emergency situations alarms were implemented. Results: The performed simulation, allowed to follow the subprocesses of interest step by step, both locally and remotely, allowing to observe the system behavior under normal conditions or emergency situations, when the simulator responds immediately. Conclusions: Based on the results obtained, it is concluded that the simulator is an adequate and inexpensive alternative for understanding, monitoring and predicting the behavior that would have the real plant in terms of the subprocesses of interest.Keywords: Automatic Control, Automation, Process Simulation, Programmable Logic Controller, Programming, Systems Design. Resumo Antecedentes: A simulação de processos é uma ferramenta útil na concepção, análise a avaliação destes, facilitando assim sua optimização e monitoramento, permitindo a predição do seu comportamento em determinadas situações. Objetivo: Portanto, neste artigo se apresenta a simulação de um processo cervejeiro, fazendo especial ênfase nos processos de enchimento, tapado e pasteurizado. Metodologia: Identificaram-se cada um dos passos de um processo cervejeiro, e caracterizaram-se os processos de interesse, definindo cada uma das variáveis de entrada e saída. A continuação, se simularam os processos fazendo uso do controlador lógico programável PLC S7-1500, e se supervisou o comportamento de cada um destes através de SIMATIC WinCC, localmente numa tela HMI (Interface Homem-Máquina), e remotamente em um computador. A dinâmica do processo se apresenta no software Automation Studio 6, que está ligado ao PLC através do servidor OPC de IBH. Para observar o comportamento do sistema em certas situações de emergência, foram implementados alarmes. Resultados: A simulação realizada permitiu acompanhar passo a passo os processos de interesse, tanto local como remotamente, permitindo observar o comportamento do sistema em condições normais ou situações de emergência nas quais o simulador responde imediatamente. Conclusões: Com base nos resultados obtidos, conclui-se que o simulador é uma alternativa adequada e de baixo custo que permite compreender, acompanhar e prever o comportamento que poderia ter a planta real no que respeita aos processos de interesse.Palavras-chave: Automatização, Controle Automático, Controlador Lógico Programável, Design de Sistemas, Programação, Simulação de Processos.