Modelling the Risk of Collision Between UAM Aircraft and Obstacles Using a Multivariate Gaussian Model

It is anticipated that urban air mobility (UAM) will be introduced to low-altitude airspace; therefore, it is necessary to analyse the risk of collision between the UAM aircraft and urban obstacles such as buildings. In the conventional air traffic system, an Instrument Landing System Collision Risk Model (ILS CRM) is used to calculate the probability of collision between an aircraft flying an ILS approach and ground obstacles. In this paper, we present a new type of collision risk model for UAM based on the multivariate Gaussian mixture model (GMM) to compensate for the limitations of the conventional ILS CRM. We propose a method to consider the aircraft deviation at different locations along the route as a multivariate random variable, allowing their correlation to be effectively considered. To validate the model, we created a toy problem with a synthetic obstacle environment and estimated the risk using the proposed model and ILS CRM method. Lastly, the proposed model was applied to aircraft arriving on runway 32R at Gimpo International Airport, with obstacle data.