Location intelligence for building empathetic driving behavior to enable L5 cars

System and methods enable vehicles to make ethical/empathetic driving decisions by using deep learning aided location intelligence. The systems and methods identify moral islands/complex driving scenarios where a complex ethical decision is required. A Generative Adversarial Network (GAN) is used to generate synthetic training data to capture varied ethically complex driving situations. Embodiments train a deep learning model (ETHNET) that is configured to output one or more driving decisions to be taken when a vehicle comes across an ethically complex driving situations in the real world.