基于测距特征向量匹配和模糊控制的室内清洁机器人路径规划

摘要

With regard to the inefficiency of random path planning method used by many indoor cleaning robots, a low-cost and efficient improvement method was proposed in this paper. The main idea was that the distance between the robot and its surrounding obstacles varied with the location of working robot. First, ultrasound distance measurement was used for obtaining the distances to form the feature vector which could identify different locations in work place. Then feature matching problem was solved by Earth Move's Distance (EMD), and the set of collision points was established. At last, fuzzy control theory was used for controlling rotation angle after collision to finish path planning procedure. As shown on the experiments, the proposed method can not only reserve the property of enjoyment and environment flexibility, but also promote the efficiency greatly.%随机路径规划算法大量用于商品化的室内清洁机器人,但规划效率低下,为此提出一种低成本、高效的改进方法.由于机器人到其四周障碍物的距离随空间位置不同而变化,先使用超声测距模块获得障碍距离并组成特征向量用以区分不同的空间位置；然后,用推土机距离(EMD)进行特征向量匹配,建立并维护碰撞点集；最后,利用模糊控制理论处理碰撞后的旋转角度,完成路径规划过程.实验结果表明,所提方法一方面保留了随机规划的工作趣味性和环境适应性,另一方面大大提高了清洁效率.