The project sets out to explore ML algorithms for embedded intelligent systems. In particular, the project is dedicated to exploring how to efficiently fuse various multi-view data(multiple modalities or uni-modalities captured from different viewpoints) at the edge. We use the HPC to simulate the results.