结合人工信标和视觉SLAM的自主快速定位方法

Autonomous Fast Localization Method Combining Artificial Beacons and Visual SLAM

针对环境复杂的工业现场中自主定位精度低、适应性差的问题,提出一种结合人工信标和视觉的同时定位与地图构建(Simultaneous localization and mapping,SLAM)的双相机定位系统。人工信标无需覆盖全空间布置,无需具备ID信息,通过两台相机分别拍摄信标及运行视觉SLAM算法完成全空间定位,利用信标定位算法恢复单目SLAM的尺度不确定因子并且建立起两种定位算法坐标系之间的关系,因此当相机视场由信标组成的控制场覆盖范围外再次移回其覆盖范围内时,可通过SLAM算法得到的系统位姿初值实现信标和像点的快速匹配,从而再次精确定位。将双相机定位系统安装在可穿戴的头盔上,测量人员头戴头盔便可实现高精度定位。分析了双相机头盔定位系统的工作原理,并进行了试验验证。试验结果表明,该方法精度较高、适应性强,可满足在信标无法全空间覆盖的复杂工业环境下的自主定位需求。

Aiming at the problem of low accuracy and poor adaptability of autonomous localization in complex industrial field environment,a dual-camera localization system combining artificial beacons and visual SLAM(simultaneous localization and mapping) is proposed.Artificial beacons need not cover the whole space layout,don’t need to have ID information.The two cameras complete the full space positioning by photographing the beacons and running the visual SLAM algorithm respectively.The scale uncertainty of monocular SLAM is recovered by using beacon localization algorithm and the relationship between the coordinate systems of the two localization algorithms is established.Therefore,when the camera moves out of the control field consisting of beacons and back into its coverage again,system initial pose can be obtained by SLAM algorithm.Then the fast matching of beacons and image points can be realized so as to locate accurately again.By installing the dual camera positioning system on the wearable helmet,the surveyors can achieve high precision positioning when wearing the helmet.The working principle of dual-camera helmet positioning system is analyzed and verified by experiments.Experimental results show that this method has high accuracy and strong adaptability,which can meet the requirements of autonomous positioning in the complex industrial environment where the beacons cannot cover the whole space.