Simulation Model of Two-Robot Cooperation in Common Operating Environment

The article considers a simulation modelling problem related to the chess game process occurring between two three-tier manipulators. The objective of the game construction lies in developing the procedure of effective control of the autonomous manipulator robots located in a common operating environment. The simulation model is a preliminary stage of building a natural complex that would provide cooperation of several manipulator robots within a common operating environment. The article addresses issues of training and research.

💡 Research Summary

**
The paper presents a virtual environment for studying cooperation between two three‑tier manipulator robots using a chess game as the test scenario. Implemented in Unity with C# scripts, the system reads a standard chess notation file (Game.txt), decodes it into numeric move codes, and separates the moves into two streams: odd‑indexed moves for the white‑piece robot (MR1) and even‑indexed moves for the black‑piece robot (MR2). Each robot’s structure—base, arm, and gripper—is modeled as a hierarchy of GameObjects, and movement is orchestrated through a sequence of functions: PovorotFrom (rotate the arm to pick up a piece), PovorotTo (translate the gripper to the target square), and DeFolt (return to the home position).

The user interface provides sliders to independently adjust the speed of each robot, buttons to start the simulation, switch between step‑by‑step and continuous autoplay modes, and display fields that show the duration of the most recent move for each robot. During execution the program records the length of the gripper’s trajectory and the elapsed time for each move, allowing the calculation of simple statistics such as average move time (1.07 s for MR1, 1.76 s for MR2) and the correlation coefficient between the two time series (r = ‑0.33). These metrics are intended to support hardware‑level optimization studies (e.g., joint dimensions, servo selection, power consumption).

The authors acknowledge several limitations. The simulation does not employ a physics engine, so collision detection, inertia, and real‑world dynamics are omitted. No sensory input (vision, force/torque feedback) is modeled, preventing real‑time perception and adaptive grasp control. Moreover, the system only replays pre‑recorded games; it does not integrate a chess engine or allow the robots to generate moves autonomously, limiting the assessment of strategic decision‑making and opponent modeling.

Future work outlined in the paper includes integrating a free‑standing chess engine to enable on‑the‑fly move generation, adding virtual cameras and vision algorithms for piece detection, and expanding the manipulator library to include two‑finger, three‑finger, and anthropomorphic five‑finger grippers equipped with strain gauges. The authors also propose a human‑machine collaboration mode where a human operator can control one robot via somatosensory gloves, gyroscopes, or brain‑computer interfaces, using Arduino‑based hardware and the EZ‑Robot development environment. Such extensions would transform the current educational demonstrator into a comprehensive research platform for cooperative robotics, adaptive training algorithms, and human‑robot interaction studies.

In conclusion, the study delivers a functional, modular simulation that visualizes and measures cooperative robot actions in a well‑structured task (chess). While the current implementation is limited to replaying fixed games without physical realism, it provides a solid foundation for teaching robot coordination concepts and for conducting parametric optimization experiments. The proposed roadmap toward sensory integration, autonomous decision‑making, and human‑in‑the‑loop control promises to elevate the platform from a pedagogical tool to a versatile testbed for advanced collaborative robotics research.