论文部分内容阅读
基于神经网络和神经网络集成理论提出了一种多传感器信息的数据融合结构,并将其用于机器人的障碍物的识别,提高了系统的识别效率,增强了系统的可靠性。通过分别搭建识别各种障碍物的子网络,以并行集成的方式把各个个体网络组合起来,可以获得一个一个高性能的识别系统。在HEBUT-Ⅰ型移动机器人上进行了验证,取得了很好的识别效果,为机器人的正确导航奠定了基础。
Based on neural network and neural network integration theory, a data fusion structure of multisensor information is proposed and used to identify obstacle of robot, which improves recognition efficiency and enhances system reliability. By separately constructing subnetworks for identifying various obstacles and combining individual networks in a parallel integrated manner, a high-performance identification system can be obtained. The HEBUT-Ⅰ mobile robot has been verified and achieved good recognition results, which laid the foundation for the correct navigation of the robot.