Ant Colony Optimization Algorithm for Industrial Robot Programming in a Digital Twin

Abstract

Advanced manufacturing that is adaptable to constantly changing product designs often requires dynamic changes on the factory floor to enable manufacture. The integration of robotic manufacture with machine learning approaches offers the possibility to enable such dynamic changes on the factory floor. While ensuring safety and the possibility of losses of components and waste of material are against their usage. Furthermore, developments in design of virtual environments makes it possible to perform simulations in a virtual environment, to enable human-inthe-loop production of parts correctly the first time like never before. Such powerful simulation and control software provides the means to design a digital twin of manufacturing environment in which trials are completed at almost at no cost. In this paper, ant colony optimization is used to program an industrial robot to avoid obstacles and find its way to pick and place objects during an assembly task in an environment containing obstacles that must be avoided. The optimization is completed in a digital twin environment first and movements transferred to the real robot after human inspection. It is shown that the proposed methodology can find the optimal solution, in addition to avoiding collisions, for an assembly task with minimum human intervention.

A Survey on AI-Driven Digital Twins in Industry 4.0: Smart Manufacturing and Advanced Robotics

“…While RL has become popular in recent years, it is not the only AI strategy that is integrated with DTs. For example, Bansal et al developed an ant colony optimization algorithm for industrial robot programming in a digital twin [ 269 ].…”

“…265], Matulis et al integrated digital twin and reinforcement learning for a 6DOF robotic manipulator to plan pick-and-place motions; in [266], Liu et al proposed a multitaskingoriented robot arm motion planning scheme based on deep reinforcement learning and twin synchro-control; in [267], Xia et al proposed a DT to train deep reinforcement learning agent for automating smart manufacturing systems; and in [268], Zhao et al demonstrated collision avoidance for a number of UAVs in a confined airspace, using LSTM-MACG.While RL has become popular in recent years, it is not the only AI strategy that is integrated with DTs. For example, Bansal et al developed an ant colony optimization algorithm for industrial robot programming in a digital twin[269].…”

XOR Binary Gravitational Search Algorithm with Repository: Industry 4.0 Applications

“…Additionally, current robot programming techniques require highly talented professionals to perform programming, which increases response times and increases costs [27]. Real-time decision making and robot programming is therefore highly desirable and is a stated requirement of Industry 4.0 [28]. However, autonomous programming of robots will impact safety and certification requirements as well.…”

“…To perform the automated optimization, it is required that the robot movements be encoded in terms of an evolvable array which may then be translated to motions to evaluate cost function and make changes to optimize the problem. A similar approach has been previously done using Ant colony optimization in another software namely Gazebo [28]. However, the benefit of using V-REP versus Gazebo is that the former benefits from built-in inverse kinematic which makes it possible to perform the simulations in a timely manner.…”

Information-Based Model Discrimination for Digital Twin Behavioral Matching

“…In this paper, a decimal numeration (r = 10) system is used for look-up table codification, treating each table element n as a high order integer, removing decimal period and adding the corresponding sign to n. For example, if n = 10.34, it is codified as " + 1034" returning a length of 5, or if n = −0.45 its codification is "−45" returning a length of 3 always removing the leading zeros. Thus, the look-up table length is given by (8).…”