Rescue Robot Multi-sensor Information Fusion Technology Research

In recent years, earthquake, mine disaster, nuclear leakage, terrorism and local war of natural or man-made disasters such as a serious threat to human life safety, has received the widespread attention. The use of the rescue robot replace human into dangerous and complex rescue site for environmental exploration, searching for surviving the robot research personnel is an important research direction.This paper takes rescue robot as the research object, to the rescue robot multi-sensor information fusion technology of research. It mainly introduces the composition and rescue robot, and analyzes the architecture of the special rescue site after disasters in the environmental characteristic based superior, and design the environmental information collection robot, and design the three modules multi-sensor information fusion algorithm mainly include:1) Adaptive weighted fusion estimation algorithm of ultrasonic ranging array measured data fusion processing, to obtain more precise distance information.2) Based on BP neural network information fusion method of underground gas detection of data processing, it takes two level fusion ways, getting in-well environment characteristics, and it makes the comprehensive safety conditions of underground made global judgment.3) Based on adaptive weighted algorithm and T-S fuzzy neural network algorithm combining two level fusion algorithm, it realize the rescue robot real-time obstacle avoidance.This paper designed by multi-sensor information fusion algorithm for rescue robot in the post-quake environmental information detection, surviving the job such as personnel search provides great help; Meanwhile, this technology also can be used in similar function robot range finder, positioning, target recognition, autonomous navigation, obstacle-avoiding and path planning, it have higher on issues such as economic significance and social significance.