Sampling Effects in Classification of Using Point Cloud Data with Machine Learning

LiDAR has become an essential sensor for realtime data collection in various applications, ranging from autonomous vehicles to environmental mapping. However, increasing the scan speed results in a trade-off between resolution and accuracy, with higher speeds leading to reduced data quality. This study aims to optimize scan speed and accuracy in a LiDAR system by adjusting the rotation angle of the step motor. Point cloud data was collected and analyzed at different angles to assess the impact on resolution and accuracy. As a result, the optimal rotation angle that balances scan speed and data quality was determined, providing valuable insights for applications that require real-time, high-precision spatial data