Systém detekce kolizí ramen průmyslových robotů

Abstract

This diploma thesis deals with the design and implementation of a collision detection system for industrial robots operating in a shared workspace. The main objective is to develop an external solution independent of the internal control systems of individual robots, capable of monitoring their state in real time, evaluating their spatial relationships, and detecting potential collision situations. The proposed system utilizes a simplified geometric model of robot arms based on capsule approximations of individual segments. This approach enables efficient computation of minimum distances between robot segments while maintaining low computational complexity, which is essential for real-time applications. The spatial positions of the robots are determined using forward kinematics based on joint coordinates obtained from the robot controllers via the OpenShowVar communication interface. The solution also includes the implementation of a digital twin of the robotic workstation, allowing real-time 3D visualization of the system state. In the event of a detected collision, the system activates an output signal via an industrial IoT device, Siemens SIMATIC IOT2050, which controls a relay integrated into the safety circuit of the robots and ensures their immediate stop. The results demonstrate that the proposed system provides reliable collision detection while maintaining sufficient processing speed for real-world deployment. Furthermore, the solution offers high flexibility and the potential for future extensions, such as more advanced geometric modeling or predictive collision detection methods.