EviLOG (Evidential Lidar Occupancy Grid Mapping)

数据集包含来自lidar点云的占用网格映射的综合训练，验证和测试数据。此外，还提供了来自测试车辆的真实lidar点云，其lidar设置与模拟lidar传感器相同。点云存储为PCD文件，占用网格图存储为PNG图像，而一个图像通道描述了免费的证据，另一个描述了占用单元状态的证据。

This dataset comprises comprehensive training, validation, and test data for occupancy grid mapping generated from LiDAR point clouds. Additionally, real LiDAR point clouds from test vehicles are provided, with their LiDAR settings matching those of the simulated LiDAR sensors. Point clouds are stored in PCD files, while occupancy grid maps are saved as PNG images. One image channel represents the evidence for free space, and the other represents the evidence for occupied cell states.

搜集汇总

背景与挑战

背景概述

EviLOG数据集由亚琛工业大学于2021年发布，专注于基于激光雷达点云的占用网格映射。它提供综合的训练、验证和测试数据，包括真实激光雷达点云，数据以PCD文件存储点云，PNG图像存储占用网格图，其中图像通道分别表示空闲和占用单元的证据。