The mission success of civil UAVs depends significantly on the sensor data quality obtained during the mission. Therefore, the interaction between the sensor and the environment should be taken into account as early as possible in conceptual UAV design optimization. For this purpose, a design framework has been developed, which allows to tailor a design to dedicated design missions, and to carry out mission simulations in a virtual environment using high-resolution terrain data. In this manner, arbitrary combinations of available sensors, different UAV configurations, and mission performance settings can be assessed in early conceptual design. An optimization algorithm allows to identify interesting regions in a given design space. The design space may then be further explored by means of parameter variations around promising points in the design space. These parameter studies are very time-consuming, especially if computationally expensive analysis methods are used. One approach to reduce the computational effort may be to make use of knowledge obtained from previous parameter variations, which were carried out at different design points. However, the impact of design variable variations on mission performance may change significantly throughout the design space. That is, knowledge from previous parameter variations cannot generally be transferred to any other design point in the design space. The goal of this paper is to investigate whether knowledge regarding relationships between design variables and mission performance can be transferred throughout an exemplary design space. For this purpose, the paper investigates how the impact of design variable variations on mission performance changes throughout the design space. A statistical analysis was carried out to determine whether changes in the relationships correlate with changes in the design variables or other aircraft data. Since mission performance also depends considerably on the operating environment, all analyses were performed for two different terrains. The results indicate that half of the parameter variations may be substituted by transferring knowledge from previous investigations at other design points. For the remaining parameter variations, more advanced methods are required to remove the associated uncertainties.     «

The mission success of civil UAVs depends significantly on the sensor data quality obtained during the mission. Therefore, the interaction between the sensor and the environment should be taken into account as early as possible in conceptual UAV design optimization. For this purpose, a design framework has been developed, which allows to tailor a design to dedicated design missions, and to carry out mission simulations in a virtual environment using high-resolution terrain data. In this manner,...     »