A vision measurement algorithm based on four-lines structured light for robots environment perception

Environment perception system is the base of autonomous navigation and control task of a mobile robot, therefore, the strength of the environment perception will impact the intelligence of a robot directly. A vision measurement algorithm based on four-lines structured light for robots environment perception is proposed in this paper. The algorithm is implemented on an embedded image processing board which contains an FPGA and a DSP. Four line lasers project four laser lines on the object to be measured, and a camera captures the image that contains light stripes which have been modulated by the surface of the object. Then the 3D coordinates of points on the laser lines can be obtained through the distortion of light stripes in the image. Each light stripe is limited in a certain region of the image so as to not overlap with each other. Therefore, the installation angle of every laser has been limited in a certain range. A sectional line fitting algorithm is proposed to improve the accuracy of centers of light stripes, thus improving the accuracy of the whole measurement system. A number of measurement results validate the effectiveness of the proposed method. Furthermore, we combine the 3D information obtained by the proposed method with the 2D edge feature of the image to construct a rough depth map of the environment.