Тип публикации: доклад, тезисы доклада, статья из сборника материалов конференций
Конференция: 5th International Conference on Control Systems, Mathematical Modeling, Automation and Energy Efficiency (SUMMA); Lipetsk; Lipetsk
Год издания: 2023
Идентификатор DOI: 10.1109/SUMMA60232.2023.10349438
Ключевые слова: firefighting robot, spray, YOLOv5, camera studio, fire detection
Аннотация: In recent years, there has been a lot of interest in robotic firefighting. These robots are critical in fighting fires in hazardous and difficult regions where human involvement is either detrimental or limited. The accurate fire size calculation capability of these robots is a critical component that permits more concentrated and Показать полностьюefficient suppression. Estimating the size of the fire is critical for determining how much water or extinguishing agent will be required to extinguish it. This computation becomes even more critical when self-adjusting water spray nozzles are utilized by firefighting robots because the nozzles modify their flow rate and spray pattern in response to the size of the fire. This research provides a method for calculating fire size that can be used to adjust water spray nozzles for use with firefighting robots. Deep learning and a camera studio were used to extract the data necessary to estimate the magnitude of the fire. The method has been shown to be useful in determining the length and width of a fire. The robots can ensure efficient firefighting operations and optimize water usage and waste reduction by precisely monitoring the extent of the fire. According to the method's simulation studies, the height and width of the fire may be estimated with 96% accuracy.
Журнал: IEEE Proccedings of 5th International Conference on Control Systems, Mathematical Modeling, Automation and Energy Efficiency (SUMMA)
Номера страниц: 231-234
Издатель: Lipetsk State Technical University