论文部分内容阅读
机器人自定位是实现机器人自动导航及其他智能行为的前提, 一种基于光束平差法的移动机器人双目视觉里程计可以有效地实现机器人自定位. 为此, 首先采用点模式匹配方法建立相邻图像之间的特征匹配关系, 根据立体视觉算法得到匹配点对的三维对应关系; 然后, 计算摄像机的相对运动参数, 并采用光束平差分段优化算法对其进行优化. 所提出的双目视觉里程计能够避免车轮半径变化、空转、打滑等对里程计测量精度的影响, 相对定位精度较高.
Robot self-positioning is the precondition of realizing robot’s automatic navigation and other intelligent behaviors.A kind of binocular vision odometer based on the beam adjustment method can realize the robot self-localization effectively.To this end, Then, the relative motion parameters of the camera are calculated and optimized by using the beam-balancing segmentation optimization algorithm.The proposed binocular vision Odometer to avoid changes in the radius of the wheel, idling, slippage and other effects on the odometer measurement accuracy, relative positioning accuracy.