System Composer vs. Model Predictive Control Toolbox

System Composer™ enables users to specify and assess architectures in the field of model-based systems engineering and software architecture modeling with great detail. By employing System Composer, individuals can proficiently assign requirements and refine an architectural model, which can later be designed and validated using Simulink®. Users have the option to develop architecture models with components and interfaces starting from the ground up, importing them from other applications, or using existing architectural elements found in Simulink designs. Moreover, it supports the depiction of a system through multiple architectural models, allowing for direct interconnections through model-to-model allocations. Behavioral characteristics can be effectively captured using sequence diagrams, state charts, or Simulink models, offering an extensive insight into system dynamics. In addition, users can define and simulate the execution sequence of component functions and even produce code from their software architecture models through tools such as Simulink and Embedded Coder®. This cohesive integration provides a solid framework for the efficient development of intricate systems while facilitating collaboration among engineering teams.

The Model Predictive Control Toolbox™ provides an extensive array of functions, an easy-to-use app, Simulink® blocks, and useful reference examples to streamline the development of model predictive control (MPC) systems. It effectively addresses linear problems by allowing the development of implicit, explicit, adaptive, and gain-scheduled MPC approaches. For more intricate nonlinear situations, users can implement both single-stage and multi-stage nonlinear MPC. Moreover, this toolbox comes equipped with deployable optimization solvers and allows for the incorporation of custom solvers as needed. Users can evaluate the performance of their controllers through closed-loop simulations within MATLAB® and Simulink environments. In the context of automated driving, the toolbox offers blocks and examples that comply with MISRA C® and ISO 26262 standards, which facilitates the rapid start of projects related to lane keeping assistance, path planning, path following, and adaptive cruise control. It enables the design of implicit, gain-scheduled, and adaptive MPC controllers that can solve quadratic programming (QP) problems while also facilitating the generation of explicit MPC controllers based on implicit designs. Furthermore, the toolbox accommodates discrete control set MPC for addressing mixed-integer QP challenges, thus expanding its versatility for various control systems. With its rich set of features, the toolbox guarantees that both beginners and seasoned professionals can successfully apply advanced control strategies in their projects. This versatility ensures that users across multiple domains can find relevant applications for their specific needs.