Vision-Cloud Data Fusion for ADAS: A Lane Change Prediction Case Study

With the rapid development of intelligent vehicles and Advanced Driver-Assistance Systems (ADAS), a new trend is that mixed levels human driver engagements will be involved in transportation system. Therefore, necessary visual guidance for drivers vitally important under this situation to prevent potential risks. To advance systems, we introduce novel vision-cloud data fusion methodology, integrating camera image Digital Twin information from cloud help make better decisions. Target vehicle bounding box drawn matched with object detector (running on ego-vehicle) position (received cloud). The best matching result, 79.2% accuracy 0.7 intersection over union threshold, obtained depth images served as an additional feature source. A case study lane change prediction conducted show effectiveness proposed methodology. In study, multi-layer perceptron algorithm modified approaches. Human-in-the-loop simulation results Unity game engine reveal model can improve highway driving performance significantly terms safety, comfort, environmental sustainability.