Mechatronic Design and Optimization Using Knowledge-Based Engineering Applied to an Inherently Unstable and Unmanned Aerial Vehicle

A novel design method for mechatronic systems, based on knowledge-based engineering techniques, is proposed in this research study. The method is particularly suited for mechatronic vehicles which are inherently unstable and require control systems for stabilization. The method is implemented in a dedicated software tool in which physical entities of the product are defined as classes with attributes. Nonphysical elements of the system and procedures for the design and analysis of the system are defined as functions with variables. The method has two key features. First, multiphysics simulation models and associated analysis functions are generated automatically within a multidisciplinary analysis and optimization framework. These models are not restricted to geometrical aspects for mechanical design, but also include the system architecture, dynamics, aerodynamics, electronic control systems and associated software codes. Second, for each representation of a design, a dedicated control system is developed completely automatically, based on the multiphysics simulation model, using model inversion control. These two features make it possible to analyze the dynamics and performance of inherently unstable mechatronic vehicles already in the early design phases when the vehicle is still subject to large configuration design changes. The method is demonstrated for the design of a multirotor unmanned aerial vehicle. Thirty thousand possible design solutions are evaluated by the system without manual interference. For each design, a dedicated control system is created and five flight test maneuvers are simulated in order to assess the aircraft performance and flying qualities. A global optimization process is applied for two conflicting requirements and the process is convergent at two optimum solutions.