Software Complex Testing Models for the Evaluation of Group Intelligence Robots

Abstract

Problem statement: Questions of group interaction of intelligent electromechanical systems (SEMS) in complex robotic systems (CRS) play an important role in the analysis of their capabilities in performing various tasks. The paper proposes a solution to the problem of group intelligence assessment based on the results of test computer simulation of CRS. This problem occurs if necessary matching the best candidates to the RTK group from existing set of modules SEMS, for example, when you want to perform merge/split parts of RTK, the introduction of the group new robot or removal of existing or producing other transformation groups SEMS associated with the implementation of technological tasks. Purpose of research: Development of software complex for testing group intelligence assessment models in CRS based on SEMS in various environments. Results: A generalized structure of the software and hardware complex for testing models of intelligent robot groups and expert subsystems providing an assessment of group intelligence based on the results of test tests have been developed. The principles of teaching robots and methods of model correction based on the results of test tests are proposed. Practical importance: The generalized structure of the software and hardware complex for the evaluation of group intelligence CRS provides high functionality of models of groups of robots, taking into account the ideology of SEMS, which allows to evaluate group intelligence by computer modeling, taking into account the characteristics of the group members and the environment of their functioning.