摘要
针对传统ICP算法受初始位置限制而精度不高的问题,提出一种基于改进灰狼优化算法的配准方法。将差分进化算法引入到灰狼优化算法中并加以改进形成新的混合算法(Differential Evolution-Grey Wolf Optimizer, DE-GWO)解码获得最优的旋转矩阵参数,将其作为初始值引入到点云细配准使用的迭代最新点算法(Iterative Closest Point, ICP)中进一步提高配准精度,以避免传统ICP算法因点云初始位置相差大而导致配准失败,提高配准精度和配准效率。通过在斯坦福点云模型上进行仿真,验证了算法的有效性。
Aiming at the problem that the traditional ICP algorithm is limited by the initial position and the accuracy is not high, a registration method based on the improved gray wolf optimization algorithm is proposed. Firstly, the differential evolution algorithm was introduced into the gray wolf optimization algorithm and improved to form a new hybrid algorithm(de-gwo) to decode and obtain the optimal rotation matrix parameters. Then, it was introduced into the iterative closest point algorithm(ICP) used for point cloud fine registration as the initial value to further improve the registration accuracy, so as to avoid the registration loss caused by the large difference of the initial positions of the point clouds in the traditional ICP algorithm To improve the registration accuracy and efficiency. Finally, the effectiveness of the algorithm was verified by simulation on Stanford point cloud model.
作者
杨沐杰
钟羽中
郭斌
佃松宜
YANG Mu-jie;ZHONG Yu-zhong;GUO Bin;DIAN Song-yi(College of Electrical Engineering,Sichuan University,Chengdu Sichuan 610065,China)
出处
《计算机仿真》
北大核心
2022年第12期513-518,共6页
Computer Simulation
基金
中央高校基本科研业务费资助项目(2018CDZG-17)。
关键词
点云配准
差分进化
灰狼优化
Point cloud registration
Differential evolution
Grey wolf optimization
