The article presents an analysis and synthesis of existing research in the field of forming a generalized architecture of land-based autonomous unmanned vehicles (UV) intended for the transport of goods and passengers within a smart city. The resulting architecture of a land-based UV allows to identify its main subsystems and sensors as well as their characteristics. And based on this, the general description of the approach to identifying components and changes in time of the characteristics of the UV’s assets. Among the UV parameters and components critical ones are selected, i.e. those whose invalid values or state over a certain period of time will lead to the greatest negative impact, including the destruction, damage or malfunction of the vehicle itself, other UVs, and/or the relevant smart city infrastructure, as well as causing harm to the passengers, cargo transported or other traffic participants. Probabilistic and event approaches are proposed in determining the level of criticality of the UV’s characteristic and its elements respectively. This paper is the first stage of a project intended to develop methods and approaches for assessing the critical characteristics of the UVs.