移动机器人声源定位与跟踪摄录系统开发  

Development of Sound Source Localization and Tracking Video Recording System for Mobile Robots

在线阅读下载全文

作  者:王一群 李兴旭 杨鸿波[1] 刘琼[1] 陈雯柏[1] WANG Yiqun;LI Xingxu;YANG Hongbo;LIU Qiong;CHEN Wenbai(School of Automation,Beijing Information Science and Technology University,Beijing 100096,China)

机构地区:[1]北京信息科技大学自动化学院,北京100096

出  处:《实验室研究与探索》2023年第5期97-101,共5页Research and Exploration In Laboratory

基  金:北京信息科技大学2021年教改立项重点项目(2021JGZD03);2021年度北京信息科技大学高教研究课题(重点)(2021GJZD04)。

摘  要:为满足短视频博主、网络主播等对低成本、简单可靠的跟随摄录设备的需求,提出了一种基于机器听觉的跟踪拍摄机器人设计方案,并进行实验验证。该设计使用GCC-PHAT算法估计目标声源到达各麦克风之间的时延,以时延大小作为声源定位系统的输入,根据空间几何关系解算出目标声源的方位信息,并使用PID控制算法控制底盘向目标靠近。测试结果表明,该设计方案具有一定抗干扰能力,能够在室内外环境下完成对在8 m范围内目标声源的定位,实际角度分辨率约为8.5°。基于机器听觉的跟踪拍摄机器人能够实现自动识别目标、运动跟随拍摄、室内外皆可稳定运行,且硬件成本相对较低,具有一定实用价值。A tracking and shooting mobile robot design based on sound source localization was proposed,and experimental verification was completed.The design uses generalized cross correlation with phase transform(GCC-PHAT) algorithm to estimate time delays for the target sound source to reach each unit of a microphone array.The delays are used as the input of the sound source localization,and the orientation information of the target sound source is solved according to the spatial geometric relationship.The PID control algorithm is used to control the robot to approach the target.Test results show that this design scheme has certain anti-interference capability,and can complete the localization of the target sound source within 8 meters in indoor and outdoor environments,and the actual angular resolution is about 8.5 degrees.The tracking and shooting robot based on sound source localization can realize automatic target recognition,motion following and shooting,indoor and outdoor stable operation,while the hardware cost is relatively low.

关 键 词:声源定位 跟踪摄录机器人 跟随控制 时延估计 

分 类 号:G642.0[文化科学—高等教育学]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象