Reinforced Learning for Navigation Equipment Test Table CAD Systems Integration

This paper proposes multi criteria decision making computer-aided design system to achieve the optimal configuration of navigation inertial equipment dynamic test table. The aim of the study is to improve the efficiency of the test table and its elements by building a computer-aided design system based on the use of intelligent approaches. A multiagent approach is proposed for automation of test table design, which uses reinforced learning to solve the optimization problem, resulting in an improved search for design solutions in terms of the use of prior knowledge. The novel method is used for solving the multi criteria decision making problem, which uses reinforcement learning agent in integrated decision process environment that allows direct search of the solution without the need to determine additional parameters of the design process model. Research includes experimental validation of the developed reinforcement learning agents by comparing results with existing multi criteria decision making methods.