Autonomous driving rely on LiDAR (Light Detection and Ranging) and camera technologies to navigate and perceive their surroundings.  Our research interests include point-cloud segmentation , adverse weather noise-tolerant  , AI-based point-cloud processing for related Lidar system applications, e.g., object segmentation in image and point cloud. Currently, we focus on a reliable LiDAR system implemented in an embedded system.  However, point cloud data can be significantly affected by adverse weather conditions such as heavy rain, snow, fog, or dust storms. These weather conditions can lead to reduced visibility and accuracy in the LiDAR data, posing a significant challenge to the safe and efficient operation of autonomous driving. To address these issues, the project statement focuses on developing a system or algorithm to enhance autonomous driving. Noisy point-cloud data is filtered out to mitigate the adverse effects of weather conditions, thereby improving the reliability and safety of autonomous driving in challenging environmental conditions. These challenges and goals not only present opportunities for innovation but also align with our project's overarching mission to create a cutting-edge solution in the field of semantic segmentation for objects. First is to increase the efficiency and accuracy of the existing system. The second is to execute the algorithm using an embedded system. The third is to reduce computational complexity. Finally, our research work needs can be well demonstrated in a real environment (e.g. heavy raining conditions).