Intelligent Path and Task Planning in Robotic Manufacturing: Leveraging Artificial Intelligence for Industry 4.0

The smart manufacturing system enables a rapid evolution of Industry 4.0 in which the safety and efficiency are collaboratively enhanced in terms of intelligent controllers, sensors, and autonomous robots. Because of high-dimensional data and uncertainties in real-time decision making, the major challenges are to achieve an adaptive and robust path and task coordination. For autonomous robotic manufacturing, advanced Artificial Intelligence models of TemTransGIN-Planner and CQ-PathNet are leveraged to propose an intelligent task and path planning framework in this study. Initialize the proposed framework with data acquisition from the sources of EDAT24 and BridgeData V2 datasets. Ensures consistent data scaling and noise reduction in the data pre-processing module. The feature extraction module derives the features of Task, Dynamic, Kinematics, and Environmental features, which can build the robot environment interactions with its comprehensive representations. For task planning and scheduling, it proposes the Temporal Transformer Graph Isomorphism Network (TemTransGIN-Planner) for intelligent decision making. Next, incorporate Parameterized Graph Explainer (PGExplainer) to enhance interpretability and explainability, thereby elucidating the decision rationale of the graph-based planner. Under dynamic and complex manufacturing conditions, the precise and collision free trajectories are achieved by proposing CQ-PathNet. To enhance the performance over time, both TemTransGIN and CQ-PathNet, through feedback-driven learning, are continuously driven by the learning and adaptation module. An accurate and stable actuation of robotic systems is ensured via Proportional Integral Derivatives (PID). Over the traditional methods, the proposed model describes 95% resource utilization, 18.5s makespan, 0.5% collision rate, 95 m/s ² path smoothness and 8.2J energy consumption.