基于三维点云的阶梯目标检测与参数估计  

Stair detection and parameter estimation using 3D point cloud

作　　者：钟晨[1] 王伟[1] 庄严[1] 

机构地区：[1]大连理工大学信息与控制研究中心,辽宁大连116024

出　　处：《控制理论与应用》2013年第6期724-733,共10页Control Theory & Applications

基　　金：国家自然科学基金资助项目(61075094;61035005);机器人学国家重点实验室基金资助项目(RL0200911)

摘　　要：阶梯目标检测与阶梯三维模型构建对移动机器人自主导航和运动规划具有重要意义.针对实际应用中阶梯目标结构的多样性以及点云分布的不确定性等特点,提出一种基于阶梯拓扑模型和模糊集理论的自适应阶梯目标检测与参数估计方法.利用阶梯剖面模型的拓扑关系与直方图算法,可有效提高阶梯边缘位置估计的精度及鲁棒性.采用同级线段提取与跨级线段接合策略,可实现对候选阶梯边缘线集合的准确构建.在此基础上,通过模糊变换和自适应模糊推理估计各级候选阶梯边缘线之间的级联概率,并采用模拟退火算法搜索全局最优的候选阶梯边缘线组合,从而实现对阶梯三维模型参数的有效估计.实验结果及数据分析验证了所提方法的有效性和实用性.Stair detection and 3D modeling are of great significance for autonomous navigation and motion planning of mobile robots. Considering the structure diversity of stairs and the distribution uncertainty of point cloud in practical application, we proposed an adaptive stair detection and parameter estimation method based on stair topology model and fuzzy set theory. By employing the topotogy of stair profile and a histogram algorithm, we improve the accuracy and robustness of the stair edge location estimation. Adopting the strategy of in-level extraction and cross-level linking, we construct a set of candidate edge lines. On the basis of this result, the cascade probability of candidate edge lines is estimated by using fuzzy transformation and adaptive fuzzy inference. A simulated annealing algorithm is applied to the global optimum search for the candidate edge-line combination to validate the estimation of 3D stair modeling parameters. Experimental results and data analysis show the validity and feasibility of the proposed approaches.

关 键 词：三维点云 阶梯目标检测 模糊推理 模拟退火 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]