Sharpening the Future of Occupancy Grid Map Prediction Methods: An Investigation into Loss Functions and Semantic Segmentation Multi-Task learning for More Accurate OGM Predictions

For an Autonomous Vehicle (AV) to traverse safely in traffic, It is vital it can anticipate the behavior of surrounding traffic participants using motion prediction. Current motion prediction approaches can be categorized into object-centered and object-agnostic methods and are primarily based on deep learning. The former relies on a human...