An FPGA processor for modelling wildfire spreading

In this paper, a model based on Cellular Automata (CAs) for predicting wildfire spreading is presented. The proposed model is inspired by existing fire spread models, but also includes a number of changes and additions, compared to them. Primary goal of the paper is to design and implement a software-based model as well as a corresponding hardware-based one that will sufficiently describe real fires, but will also have less stringent requirements on computational resources and computational power for execution. Therefore, an effort has been made to minimize the complexity of the model and the resulting computational burden aiming at an implementation that will have a practical significance in predicting the evolution of a fire. The proposed model is implemented on an Altera (R) Stratix IV (R) FPGA (Field-Programmable Gate Array), designed to execute in a parallel way in order to produce valuable information in real time that can be used to optimize response to a fire crisis. The FPGA design results from the automatically produced synthesizable VHDL code of the CA model and is advantageous in terms of low-cost, high speed and portability. The resulting implementation sufficiently depicts the natural phenomenon of wildfire spreading and provides short calculation times. Finally, the presented FPGA implementation of the proposed CA model offers the possibility a portable system to be designed, connected with GPS as well as GIS and/or wind monitoring systems able to provide real-time information concerning the wildfire propagation on the under test area. (C) 2012 Elsevier Ltd. All rights reserved.