Researches On Multi-Sensor Detection System For Forest Fire Embers Detection And Cleaning Robot

Forest fires occur frequently,which cause serious damages to forest resources.After the fire,fire embers are not easy to be found.The detection and timely treatment of fire embers is very important to forest fire prevention.The complexity of forest environment and risk of artificial detection make forestry development and application become key problems to be solved.In view of the present forest fire and fire embers detection technology and lack of method,researches on multi-sensor detection system for forest fire embers detection and cleaning robot were conducted in the hope of the future application in the detection of forest fires and embers and cleaning work.Forest fire embers detection and cleaning robot multi-sensor detection system research including multi-sensor system's hardware and software design,the multi-sensor information fusion and MATLAB simulation.(1)According to the uniqueness of the forest environment,selected the hexapod bionic robot as the carrier of multi-sensor detection system and built the overall architecture of the bionic robot multi-sensor detection system.(2)According to the characteristics of fire embers and post-disaster environment requirements,multi-sensor detection system hardware was designed,including the components of the selection,the design of the hardware circuit diagram and reasonable layout of sensors.Besides,according to the designed hardware system,made plans for software processes.(3)Discussed the advantages of multi-sensor information fusion.Based on the method of combination of artificial neural network of BP neural network and fuzzy control,built the model of multi-sensor detection system.Selected fuzzy control method to design the robot's obstacle avoidance system.(4)Made experimental verification and optimization for forest fire embers detection and cleaning robot multi-sensor detection system.With the help of MATLAB software,embers detection system of the fuzzy neural network simulation and system simulation of obstacle avoidance were verified,which proved to be good.