Reinforcement learning and convolutional neural network system for firefighting rescue robot

摘要: In this paper, we combine the machine learning and neural network to build some modules for fire rescue robot application. our research, legs module with Q-learning. We also finish face detection color sensors infrared sensors. It is usual that image fusion done when want use two kinds of Kalman filter chosen meet requirement. After indispensable steps, sliding windows choose region interest make system’s calculation lower. The least step convolutional network. design a seven layers find feature distinguish it or not.