Model-Based Development and Evaluation of Control for Complex Multi-Domain Systems: Attitude Control for a Quadrotor UAV


  • Ivan Grujic Department of Engineering, Aarhus University
  • René Nilsson Department of Engineering, Aarhus University


cyber-physicals-systems, modeling, co-model, cosimulation, drone, quadrotor, quadcopter, UAV, model predictive control, MPC, PID, attitude control, model fidelity


A Cyber-Physical System (CPS) incorporates sensing, actuating, computing and communicative capabilities, which are often combined to control the system. The development of CPSs poses a challenge, since the complexity of the physical system dynamics must be taken into account when designing the control application. The physical system dynamics are often defined within mechanical and electrical engineering domains, with the control application residing in software and control engineering domains. Therefore, such a system can be considered multi-domain.

With the constant increase in the complexity of such systems, caused by technological advances in all domains, new ways of approaching multi-
domain system development are needed. One methodology, which excels in complexity management, is model-based development. Multidomain systems require collaborative modeling, where the physical system dynamics are captured in the Continuous Time (CT) modeling domain and the digital control is captured in the Discrete Event (DE) modeling domain.

This thesis demonstrates how an extended CT-first model-based development approach can be applied to a complex multi-domain system. A collaborative model of a quadrotor Unmanned Aerial Vehicle (UAV) has been constructed and used to develop an attitude controller based on Model Predictive Control (MPC). The MPC controller has been compared to an existing open source Proportional Integral Derivative (PID) attitude controller.

This thesis contributes to the discipline of model-based development with a methodological extension to the CT-first approach, which extends the conventional approach by expanding the physical modeling process into three consecutive steps. An evaluation of the extension is presented, describing how and when the extended methodology provides increased value.




How to Cite

Grujic, I., & Nilsson, R. (2016). Model-Based Development and Evaluation of Control for Complex Multi-Domain Systems: Attitude Control for a Quadrotor UAV. Technical Report Electronics and Computer Engineering, 4(23). Retrieved from