基于案例推理的轮式移动机器人仿变色龙视觉受污偶然性规划

为使移动机器人能够有效应对视觉受污这一突发事件，提出一种基于案例推理（CBR）的偶然性规划方法.首先，对搭载仿变色龙视觉系统的轮式移动机器人（WMR）进行描述，分析其双目负相关运动机制，并给出一种融合隔帧差分法与背景差分法的改进的污染物提取算法，实现动态场景中静止污染物的检测.然后，通过详细分析机器人视觉受污后的环境感知行为建立轮式移动机器人仿变色龙视觉受污环境感知模型，对基于CBR的视觉受污偶然性规划进行建模，并详细分析视觉受污后CBR的推理过程.最后，设计基于机器人目标跟踪常规规划的视觉受污偶然性规划实验，实验结果显示目标跟踪误差基本介于±15个像素之间，表明在视觉受污情况下跟踪效果良好.

The Contingency Planning of Chameleon-Inspired Visual Contamination forWheeled Mobile Robot Based on Case-Based-Reasoning

To effectively cope with the unexpected event of visual contamination of mobile robots, a contingency planning approach is proposed based on case-based reasoning (CBR). Firstly, a wheeled mobile robot (WMR) equipped with the chameleon-inspired visual system is described, and the negative-correlation mechanism of binocular movement is analyzed. In order to realize the detection of static contaminants in dynamic scene, an improved contaminant extraction algorithm is proposed, which combines the frame difference method and the background difference method. Then, an environment perception model with chameleon-inspired visual contamination for WMR is built through the analysis of environment perception when visual contamination occurs. A CBR-based contingency planning model of visual contamination is established, and the reasoning process of CBR for visual contamination is analyzed in detail. Finally, a contingency planning experiment for visual contamination is designed based on the robot general planning of target tracking, and the experimental results show that the tracking error is basically between ±15 pixels, which demonstrates that the tracking effect is better under the condition of visual contamination.