Point cloud registration method for multi-task agent assisted optimization

The invention discloses a point cloud registration method for multi-task agent-assisted optimization, and the method comprises the steps: carrying out the sampling of a source point cloud model and a target point cloud model for K times according to the sequentially reduced sampling rates, and obtaining K first point cloud models and K second point cloud models; according to the ith first point cloud model and the ith second point cloud model, constructing a sub-population i corresponding to the ith proxy task to obtain K different sub-populations; the ith proxy task is a registration task of the ith first point cloud model and the ith second point cloud model, and a rotation matrix and a translation vector of the ith first point cloud model and a rotation matrix and a translation vector of the ith second point cloud model are two different individuals in the sub-population i; the value of i ranges from 1 to K; and iteratively optimizing the K different sub-populations by adopting a genetic algorithm, and reg