一种基于SVM-RANSAC的路径规划算法
首发时间:2017-05-12
摘要:针对在非结构化道路环境下地面自主车辆的路径规划问题,提出了一种基于距离传感器以及SVM-RANSAC的路径规划算法. 首先利用非线性SVM分类器在栅格地图上提取出安全的路径;然后在多帧投影数据上使用RANSAC提取出路径,并使用三次多项式进行描述,从而计算出道路曲率;结合UGV自身状态在RANSAC路径上选取控制点,并使用贝塞尔曲线拟合出最终路径。算法有效地解决了非结构化道路环境下难于依靠视觉传感器进行路径规划的问题. 最后,通过实车试验对所提出的方法的有效性和正确性进行了验证。
关键词: 非结构化道路; 路径规划; UGV; SVM; RANSAC
For information in English, please click here
A Novel Path Planning Algorithm based on SVM-RANSAC
Abstract: Aiming at the path planning problem of unmanned ground vehicle (UGV) under the unstructured road environment, a novel path planning algorithm based on the distance-sensing and SVM-RANSAC is proposed. Firstly, a safe path is extracted from the grid map by the non-linear SVM; a path described with the cubic ploynomial can be extracted by using RANSAC on the multi-frame projection data and the path can be used to calculate the curvature of the road; then the control points are chosen by considering the UGV's state and the final path is generated by using the Bezier curve fitting. This algorithm effectively solves the problem which is difficult to rely on visual sensor for path planning in unstructured road environment. Finally, the validity and the correctness of the proposed algorithm is verified on a real vehicle..
Keywords: unstructured road path planning UGV SVM RANSAC
基金:
论文图表:
引用
No.4732822119947714****
同行评议
勘误表
一种基于SVM-RANSAC的路径规划算法
评论
全部评论0/1000