Heuristics-based parallelization of a class of control systems

The paper proposes an effective approach for the parallelization of control system models governed by ordinary differential equations. The method takes a mathematical model expressed in block diagram language and automatically assigns the code to be executed on several parallel standard microprocessors. The integrity of the simulations is preserved, the computing resources available are efficiently used, and the simulations are compliant with real-time constraints. Numerical simulations demonstrate the effectiveness of the proposed approach.