为提高机器人的自主导航能力和环境适应性,研究一种基于声音识别技术的机器人动态导航系统.该系统通过传声器阵列采集语音指令,利用深度学习算法实现语音识别和声源定位,并结合同步定位与建图(Simultaneous Localization And Mapping,SLAM)算法构建环境地图,采用A*和动态窗口(Dynamic Window Approach,DWA)算法进行路径规划,通过模型预测控制(Model Predictive Control,MPC)策略实现运动控制.实验结果表明,该系统在复杂环境下具有良好的健壮性和适应性.
Design of Dynamic Navigation System for Robots Based on Sound Recognition Technology
In order to improve the autonomous navigation ability and environmental adaptability of robots,a dynamic navigation system for robots based on voice recognition technology is studied.The system collects voice commands through microphone array,uses deep learning algorithm to realize speech recognition and sound source location,and combines Simultaneous Localization And Mapping(SLAM)algorithm to build environment map.A*and Dynamic Window Approach(DWA)algorithm were used for path planning,and Model Predictive Control(MPC)strategy was used for motion control.The experimental results show that the system has good robustness and adaptability in complex environment.
NETL
NSTL
万方数据
