Abstract: A point cloud data processing method and device, a computer device and a storage medium are provided. The method includes: acquiring point cloud data, and constructing a corresponding neighboring point set for each of data points in the point cloud data; calculating Hausdorff distances between the neighboring point set and a pre-constructed kernel point cloud to obtain a distance matrix; calculating a convolution of the neighboring point set with the distance matrix and a network weight matrix in a Hausdorff convolution layer in an encoder, to obtain high-dimensional point cloud features, the encoder and a decoder being two parts in a deep learning network; and reducing feature dimension of the high-dimensional point cloud features through the decoder, so that a classifier performs semantic classification on the point cloud data according to object point cloud features obtained by the dimension reduction.

Abstract: A method for generating a robot exploration path includes: establishing a point cloud scene according to single-frame point cloud data collected by a robot, the point cloud scene includes a plurality of data points; classifying and marking each of the data points, the classification mark includes a ground point and an obstacle point, and taking the data points marked as ground points as a ground point cloud; generating a confidence map according to the ground point cloud, the confidence map is configured to display a location and a confidence of each of the ground points in the ground point cloud; generating a plurality of access nodes according to the confidence map; and generating an exploration path according to the access nodes.

Abstract: A point cloud data processing method and device, a computer device and a storage medium are provided. The method includes: acquiring point cloud data, and constructing a corresponding neighboring point set for each of data points in the point cloud data; calculating Hausdorff distances between the neighboring point set and a pre-constructed kernel point cloud to obtain a distance matrix; calculating a convolution of the neighboring point set with the distance matrix and a network weight matrix in a Hausdorff convolution layer in an encoder, to obtain high-dimensional point cloud features, the encoder and a decoder being two parts in a deep learning network; and reducing feature dimension of the high-dimensional point cloud features through the decoder, so that a classifier performs semantic classification on the point cloud data according to object point cloud features obtained by the dimension reduction.