Industrial Automation & Logistics

Industrial manufacturing, good transport and logistics are traditional and long-standing applications of artificial intelligence and automation.

Applications

In the industrial field, a variety of application examples for the successful use of Artificial Intelligence respectively SoftComputing methods emerged:

  • Development and design of control algorithms for autonomous, mobile robots in various applications such as automated guided systems
  • Swarm intelligence for the coordination of various (mobile) robots equivalent to the approaches in the VERONET traffic control
  • Monitoring, analysis and ongoing optimization for the control of goods flow
  • Robot controls including adaptive path planning and collision avoidance (offline and/or online)
  • Automatic identification and classification of workpieces with pattern recognition methods
  • Various robotic applications, such as collision avoidance of robots for improved human/machine interaction
  • Scenario-based approaches to optimize production processes under various operational conditions
  • Usage of data mining methods for creating data-based forecasting models
  • Cooperative/interconnected control for the coordination of various production stations
  • Automatic anomaly and fault detection in production processes
  • Failure prediction of operation-dependent predictive maintenance
  • ...

References

Certain application projects of ANDATA were, for example:

  • Multi-criteria offline optimization of the robot paths for stationary industrial robots (Kuka KR500, Kuka KR16, Kuka LBR) including implementation of the control unit
  • Recognition and interpretation of arbitrary workpieces and adaptive planning and control of the robot paths for a specific machining procedures (e.g. for casting plastering)
  • Development and implementation of the complete control software with free navigation and any freely configurable missions of automated guided vehicles
  • Development and implementation of automated sensor data analysis for the failure prediction of certain production machines or for the detection of improper machine usage
  • Automated detection of faulty workpieces with Neural Networks (by Deep Learning) and alternative image recognition methods