基于光流法的移动机器人避障平衡策略的改进

Improvement of Obstacle Avoidance Strategy of Mobile Robot Based on Optical Flow

平衡策略是利用光流场实现避障的最为简单直接的方法。将摄像头采集到的图像序列中的每一帧均分成左右2个视场并分别计算光流,控制移动机器人向光流值较小的一侧移动以实现避障。然而,当移动机器人视场中央存在左右对称的障碍物时,计算出的左右视场光流值基本相等,导致传统的平衡策略失去效果。该项目针对这种情况提出一种改进的平衡策略,使改进后能够满足上述情况的需求,提高移动机器人避障成功率。

The balance strategy is the simplest and most straightforward way to achieve avoidance by using the optical flow field.Each frame in the image sequence acquired by the camera was divided into left and right field of view,and the optical flow was calculated respectively.The mobile robot was controlled to move toward the smaller side of the optical flow value to achieve obstacle avoidance.However,when there are left and right symmetrical obstructions in the center of the mobile robot,the calculated left and right field of view light is basically equal,resulting in the traditional balance strategy to lose the effect.In this paper,an improved balance strategy is proposed to improve the success rate of mobile robots'obstacle avoidance.