The rapid development of data science and machine learning in many research as well as industrial fields has drawn much attention to the fuel of these techniques – the data. In the domain of autonomous driving, the data are mostly collected from different sources which aims to endow the data with more versatility and diversity, and also having a wider coverage in order to get a more complete and accurate perception of the environment. This project aims to improve the reliability and safety of the perception systems for autonomous driving by fusing and analysing the spatiotemporal data from different sensors and different road users that are in the same communication sensor network. In this scenario, the reconstruction of static objects can rely both on asynchronous data from a specific time span of the same sensor as well as the synchronised data from different sensors, the dynamic objects can be tracked based on the later one and auxiliated by the static information obtained. During the fusion process, the accuracies and uncertainties should also be considered and propagated to the final result and then be efficiently visualised in addition to the visualisation of the aggregated environment in order to give the human driver or passenger a correct and precise impression about the current outside-environment so that they can also intervene the driving to fulfil their need without making mistakes.
