Öz Kalabalık şehirlerde kent içerisinde itfaiye istasyonlarının doğru yer seçimi, yangınlara hızlı müdahale etmek, can ve mal kaybını en aza indirmek açısından çok hayati bir konudur. İtfaiye istasyonu yer seçiminde; kent bütününü belirli bölgelere ayırarak belirlenen her bir bölge için itfaiye istasyonu ihtiyacının sorgulanması gerekmektedir. Bu çalışmada da mevcut itfaiye istasyonlarından yola çıkarak makine öğrenmesi algoritmaları kullanarak bölgelere göre itfaiye istasyonu ihtiyacının sınıflandırılması gerçekleştirilmiştir. Çalışma kapsamında her bir bölgeye ait, itfaiye araçlarının o bölgeye ulaşım süreleri, bölgenin nüfus yoğunluğu, bölgeye giden ortalama ana ve yardımcı araç sayısı verileri ile bölgedeki itfaiye istasyonu bulunma durumu verileri kullanılarak istasyon ihtiyacının tahmini için sınıflandırılma çalışması gerçekleştirilmiştir. Bu çalışmadaki amaç İzmir Büyükşehir Belediyesinin belirlediği 808 bölgeye dair itfaiye istasyonu ihtiyacı sınıflandırılmasında en başarılı sınıflandırma algoritmasının tespit edilmesidir. 2015-2017 tarihleri arasındaki yangın kayıtları analiz edilerek bölgelerin sınıflandırılmasında %93.84 ile en başarılı algoritmanın Random Forest algoritması olduğu tespit edilmiştir. En başarılı algoritma tespit edilirken doğruluk, ortalama mutlak hata (MAE), kök hata kareler ortalaması (RMSE) ve Kappa değerleri göz önüne alınmıştır.

No abstract

In this paper, the problem of simultaneous localization and mapping (SLAM) using a modified Rao Blackwellized Particle Filter (RBPF) (a modified FastSLAM) is developed for a quadcopter system. It is intended to overcome the problem of inaccurate localization and mapping caused by inertial sensory faulty measurements (due to biases, drifts and noises) injected in the kinematics (odometery based) which is commonly used as motion model in FastSLAM approaches. In this paper, the quadcopter's dynamics with augmented bias and drift models is employed to eliminate these faults from the localization and mapping process. A modified FastSLAM is then developed in which both Kalman Filter (KF) and Extended Kalman Filter (EKF) algorithms are embedded in a PF with modified particles' weights to estimate biases and drifts and landmark locations, respectively. In order to make the SLAM process robust to model mismatches due to parameter uncertainties in the dynamics, measurements are incorporated in the PF and in the particle generation process. This leads to a cascaded two-stage modified FastSLAM in which the extended FastSLAM 1.0 (to include dynamics and sensory faults) is employed in first stage and the results are used in second stage in which probabilistic inverse sensor models are incorporated in the particle generation process of PF. The efficiency of the proposed approach is demonstrated through a co-simulation between MATLAB-2019b and Gazebo in the robotic operating system (ROS) in which the quadcopter model is simulated in Gazebo in ROS using a modified version of the Hector quadcopter ROS package. The collected pointcloud data using LiDAR is then utilised for feature extraction in the Gazebo. The simulation environment we are using is validated based on experimental data.

No abstract

In crowded cities, selection of the suitable location for fire stations within the town is a vital issue in terms of rapid response to fires and minimizing loss of life and property. For the selection of the suitable fire station location, at first it is necessary to divide the whole city into certain zones and the need for a fire station service should be questioned for each zone. In this study, based on existing fire stations service area, classification of fire station requirement by zones was carried out using machine learning classification algorithms. In order to estimate fire station requirement according to the zones, a classification study was conducted by using some data such as the travel time of the fire engines to zone from closed fire stations, population density of the zone, the mean number of main and assistant vehicles travelling to the zone from closed fire stations, and the fire station existence data in the zone. The purpose of this study was to determine the most successful classification algorithm for the classification of the fire station requirement of 808 zones determined by Izmir Metropolitan Municipality. As a result of the analysis of fire records between 2015 and 2017, it was found that for the classification of the zones, the most successful algorithm was Random Forest algorithm with 93.84% accuracy rate. Experimental evaluation of the study; according to the 5-minute access distance of the existing fire stations, the fire station requirements of the regions and the fire station needs of the regions covered by the machine learning algorithm classification results were found to be 85.43% similar.