机器人技术应用搜救机器人.pptx

Search and Rescue Robotics;Index;Development;Research motivation;Main task;Types of Rescue Robots;Application;Application;Post-disaster experiment;Fire fighting robots;The role of fire fighting robots;The role of fire fighting robots;Control processing communication function;The role of fire fighting robots;Fire fighting robots;Fire fighting robots;Ruin search and rescue robots;Ruin search and rescue robots;Ruin search and rescue robots;Route plan & Signal transmission;Demolition bomb robot;Water surface rescue robot;Performance characteristics;Performance characteristics;Underwater search and rescue robots;Underwater search and rescue robots;Underwater communication and data transmission;Key technical issues;Technological development direction;Summer;Ellipse proposal and convolutional neural network discriminant for autonomous landing marker detection;Index;Domain involved;MBZIRC;Related research;Related research;The main contributions;Elliptic detection algorithm;Preprocessing;Removal of a straight line;Elliptic arc selection;Elliptic arc selection;Evaluate whether an ellipse can be formed;Parameter fitting;Parameter fitting;The elliptical validation;Landing marker validation;Training data;Extraction of the inner region of the ellipse;Network structure;Experiment;Experiment;Experiment;Experiment;Experiment;After obtaining the simulation results, the algorithm needs to be transplanted into the onboard computer, and the algorithm is run and analyzed with the same data set. The algorithm is mainly divided into five stages. It can be seen from the results that the algorithm can be run in real time on an onboard computer, and the average detection time is more stable.;Influence of different ellipse detection algorithms on F-measure and running time;Landing marker X detection;And choose a series of points. The constraint formula is as follows:;Actual flight test;Summer