Anonym-2045/passage

该数据集是一个用于路径规划研究的合成地形数据集，基于真实世界高程数据生成。它包含多种分辨率（从64x64到4096x4096）的地形网格，每个样本代表一个地理瓦片区域，并带有起点和终点标记。数据集通过A*算法计算了三种不同成本模型（高程、能量、坡度）下的最优路径，包括无障碍物和有障碍物两种场景。元数据详细记录了地理来源、裁剪参数、障碍物生成设置、求解器配置、路径长度、成本和计时信息。数据集划分为训练、校准、验证和测试集，适用于机器学习模型在路径规划、地形分析和导航任务中的训练和评估。This dataset is a synthetic terrain dataset for path planning research, generated from real-world elevation data. It includes terrain grids at multiple resolutions (from 64x64 to 4096x4096), where each sample represents a geographic tile area with start and end point markers. The dataset computes optimal paths using the A* algorithm under three different cost models (elevation, energy, slope), in both obstacle-free and obstacle-included scenarios. Metadata comprehensively records geographic sources, cropping parameters, obstacle generation settings, solver configurations, path lengths, costs, and timing information. The dataset is split into train, calibration, validation, and test sets, suitable for training and evaluating machine learning models in path planning, terrain analysis, and navigation tasks.