Abstract:
The theory of emotional robots proposes a mathematical model that describes a uniform multi-level process of information accumulation by a robot. The process is uniquely defined by the following parameters: a piece of information that the robot receives at each event, and a coefficient of the information short-term memory of the robot. If the parameters are not known, then there is the problem of forecasting the piece of information accumulated by the robot for a certain time tact. We are the first to solve the problem on identification of the parameters of a uniform multi-level process of information accumulation by a robot on the basis of the experimental pieces of the accumulated information, which are measured in the robot at certain moments of time. The article introduces necessary restrictions. The problem is divided into two subproblems. The first is to generate all possible pairs of sequences of information accumulation levels and time steps at the levels. The second is to solve the optimization problem by the Lagrange multiplier method for each pair of sequences, which are obtained as a result of solving the first subproblem.
Keywords:robot, theory of emotional robots, information accumulation, optimization problem, Lagrange multiplier method.