Real-Time Control

Robert W. Brennan

The primary objective of this research is the development of a real-time distributed intelligent control system for the lower process/machine control level that is stable in the face of disturbances, adaptable and flexible in the face of change, and makes efficient use of available resources.

Although recent research has proven the concept of distributed intelligent control at the higher planning and scheduling level (e.g., (Brennan, 2000)), the lower process/machine control level has been left virtually untouched. The main barriers at this level result from the difficulty of implementing multi-agent systems concepts in a stochastic environment where hard real-time constraints must be met to achieve safe system operation.

Current industrial control is typically implemented using large, and often expensive, hardware platforms that support monolithic computer control applications. As a result, when the control system is installed, commissioning can take months to complete, and once the system is operational, changes are often complex and difficult. Both of these factors contribute significantly to the total cost of the industrial control project.

This research is intended to rectify these problems through the development of a distributed intelligent control solution that is inherently adaptable and dynamically reconfigurable. This system will take advantage of distributed artificial intelligence at the planning and control levels to achieve significantly shorter up-front commissioning times as well as significantly more responsiveness to change. These potential benefits, in combination with the current trend towards low-cost, distributed computing platforms, will result in a much more attractive, low-cost automation solution for future manufacturers than current centralised solutions.

Projects

• Distributed intelligent control modelling and design. This research focuses primarily on how distributed artificial intelligence concepts can be applied to hard real-time control problems. In particular, the emerging IEC 1499 standard for industrial control systems is being used as a general modelling and design approach and its relationship with other modelling approaches is being investigated (e.g., conventional programmable logic controller languages, objects, agents, Petri nets, etc.).
• Fault monitoring and recovery. This research involves the identification of the types and nature of faults that a real-time manufacturing system controller must be deal with, the classification of the ways in which control agents should handle fault recovery (e.g., fail-safe modes, homogeneous and diverse redundancy), as well as the development of preliminary models of control agents using the IEC 1499 function block model.

Selected Publications

R.W. Brennan, M. Fletcher, and D.H. Norrie, “An agent-based approach to reconfiguration of real-time distributed control systems,” IEEE Transactions on Robotics and Automation, 18(4), pp. 444-451, 2002.

R.W. Brennan, X. Zhang, Y. Xu, and D.H. Norrie, “A reconfigurable concurrent function block model and its implementation in real-time Java,” Integrated Computer-Aided Engineering, 9(3), pp. 263-279, 2002.

Contact Information

        Email:          brennan@enme.ucalgary.ca
        Telephone:      (403) 220-5798

Last updated: 9 June 2003