Multi-Sensor Information Fusion And Its Application In Robot

Multi-sensor information fusion is a rapidly evolving research area and requires interdisciplinary knowledge in control theory, signal processing, artificial intelligence, probability and statistics, etc. It provides a method technology for robot working in intricate, dynamic, uncertain or unknown environment. Multi-sensor information fusion refers to the synergistic combination of sensory data from multiple sensors to provide more reliable and accurate information. The potential advantages of multi-sensor fusion are redundancy, complementation, timeliness, and cost of the information.In the back of Heilongjiang Province's Natural Science Foundation "the Study of Shared Control Used in Remote Welding Robot", the thesis discusses sensor fusion methods based on robot and the main achievements are presented as follows:1. The key techniques and trend of robot and information fusion are introduced. Then, the application of multi-sensor information fusion in robot is analyzed. Finally, sensor fusion method is proved to be effective in order to improve the robot intelligence.2. The main research topics of multi-sensor information fusion such as information representation, sensor modeling and fusion level are studied. The classification of multi-sensor fusion algorithms is also presented.3. The principle of fuzzy neural network is analyzed and introduced. Then, we present fuzzy neural network with changeable structure based on T-S mode and it is used to navigation of robots. In order to avoid the obstacles successfully, detection results from CCD and ultrasonic sensors are fused by a fuzzy neural network, which acts as an avoidance controller. Simulation results show the validity of the proposed methods.4. B-Spline function based fuzzy neural network, which combine the concept of fuzzy variable, rule base and the learning function of neural network is proposed. This fuzzy neural network takes the role as decision layer in the system, and is the important chain of multi-sensor information fusion system.5. Robot navigation requires the global recognition of the environment. To fusion the multi-sensor information, a mobile robot can seek to accomplish tasks in an unstructured workspace. We design a robot model, which is equipped with multiple ultrasonic sensors. By fusion the sensor information, the mobile robot can move to the target position autonomously while avoiding obstacle. The safety and robustness of the fusion system are tested.