(371aj) Simulation Software for the Assessment of Nonlinear and Adaptive Multivariable Control Algorithms

Various novel control algorithms are designed to address the closed-loop stability and performance of nonlinear systems. The efficacy of the nonlinear control algorithms is typically demonstrated through application to tractable simulation examples, which limits the performance assessment and comparative analysis of the algorithms. Although novel applications of control systems promote progressive contributions in new fields, the new problems are accompanied by unique sets of challenges and considerations for the theory and implementation of control systems. Simulation software has the potential to accelerate the development and assessment of novel nonlinear and/or adaptive control algorithms in complex and challenging scenarios [1].

An emerging application of control systems deals with the regulation of biological or physiological processes [2]–[4]. Despite the potential, the assessment of new nonlinear multivariable control algorithms for emerging applications such as automated drug delivery is limited by the conventional simulation platforms that are designed primarily for evaluating classical control algorithms. Biological systems are characterized by inherent nonlinearity, significant time-delays, and stochastic disturbances that are difficult to model and degrade the realizable control performance. Readily measurable supplemental physiological variables that are indicative of the presence of disturbances must also be generated to progress beyond the single-input, single-output control architecture to multivariable control schemes [5]. A new simulation software platform is therefore necessary to enable the assessment and expedite the development of novel biomedical applications for multivariable control systems.

A new simulation software is developed to enable the testing and evaluation of nonlinear and adaptive control algorithms proposed for automated insulin delivery. Automated insulin dosing algorithms are developed to compute the optimal amount of insulin infusions to people with type 1 diabetes (T1D), a chronic disease characterized by insulin deficiency as a result of the autoimmune-mediated destruction of insulin-producing beta cells in the pancreas. Since insulin is a vital hormone in the regulation of blood glucose levels, people with T1D must administer exogenous insulin to maintain their blood glucose concentrations in a safe target range. The lack of durable glycemic control increases the risk of acute and chronic long-term complications associated with the disease. Preventing T1D-related morbidity and mortality requires novel insulin dosing algorithms that improve glycemic control.

A number of advanced simulators have been developed for T1D in recent years. They have detailed models on the effects of meals on glucose levels and compute a single output variable, the glucose concentration. This limits their use to controllers with single inputs. The multivariable glucose-insulin-physiological variable simulator (mGIPsim) developed by our research group includes several other “measured” variables that can be used by the control algorithm to detect and forecast the impact of measured disturbances such as physical activities.

The development of insulin dosing algorithms can be accelerated by using the mGIPsim software for the simulation of virtual subjects with T1D to recreate the complex biological systems in an inexpensive virtual environment where the control algorithms can be readily evaluated before migrating to clinical settings. The simulation software can also be harnessed to reinforce the theory, design, and computational implementations of nonlinear and stochastic control algorithms. The presentation will illustrate capturing the characteristics of various types and intensities of physical activities in addition to meal effects, and the performance assessment of adaptive MPC systems that leverage these measurable disturbances,

The simulator will be made available as a testbed to the research and education communities for testing their control algorithms for nonlinear systems with time-varying parameters.

[1] J. J. Downs and E. F. Vogel, “A plant-wide industrial process control problem,” Comput. Chem. Eng., vol. 17, no. 3, pp. 245–255, 1993.

[2] B. W. Bequette, “Challenges and recent progress in the development of a closed-loop artificial pancreas,” Annu. Rev. Control, vol. 36, no. 2, pp. 255–266, 2012.

[3] G. Birol, C. Ündey, and A. Çinar, “A modular simulation package for fed-batch fermentation: penicillin production,” Comput. Chem. Eng., vol. 26, no. 11, pp. 1553–1565, 2002.

[4] R. Gondhalekar, E. Dassau, and F. J. Doyle, “Periodic zone-MPC with asymmetric costs for outpatient-ready safety of an artificial pancreas to treat type 1 diabetes,” Automatica, vol. 71, pp. 237–246, 2016.

[5] K. Turksoy, E. Littlejohn, and A. Cinar, “Multimodule, multivariable artificial pancreas for patients with type 1 diabetes: regulating glucose concentration under challenging conditions,” IEEE Control Syst. Mag., vol. 38, no. 1, pp. 105–124, Feb. 2018.