Optimization of multidimensional dynamic systems using the criterion of minimum acceleration energy (Q1903465)

In the nineteenth century, Saint-Germain introduced the concept of acceleration energy by analogy with the concept of kinetic energy. This generalization proved fruitful in analytical mechanics: in 1899, Appell, using the function-definition of the energy of acceleration, obtained equations of dynamics that were suitable both for holonomic and nonholonomic systems. In previous investigations, the author has achieved a new trend in the theory of synthesis of algorithms for controlling the motion of dynamic systems. This trend is based on the concepts of the inverse problem in dynamics in combination with optimization (with respect to the functionals characterizing the energy of motion) of the system being controlled. The basic features of this approach are as follows. The dynamic characteristics of the system being designed are set using standard models whose structure and parameters correspond to the requirements of the speed of response, the nature of the transient, and the accuracy of the control in the steady-state motion and in the transient regime. The standard models form phase trajectories that the trajectory of motion of the object being controlled must follow with the necessary degree of approximation. The controlling functions are synthesized from the condition that the neighborhood of the assigned phase trajectories, the instantaneous value of the functional of the energy of acceleration belong, with respect to each degree of freedom, to a small neighborhood of the minimum. In the presented publication, these ideas are developed as they apply to problems of stabilizing stationary states and program trajectories of motion of multidimensional systems. Control algorithms are synthesized by optimizing a system using the criterion of minimum acceleration energy, calculated in the neighborhood of the phase trajectories of motion of standard models with respect to each degree of freedom. The properties of systems are investigated, and the problems involved in realizing the control algorithms are examined.