Neural network based feedback scheduler for networked control system with flexible workload
ICNC'05 Proceedings of the First international conference on Advances in Natural Computation - Volume Part II
Integrated computation, communication and control: towards next revolution in information technology
CIT'04 Proceedings of the 7th international conference on Intelligent Information Technology
Fuzzy logic based feedback scheduler for embedded control systems
ICIC'05 Proceedings of the 2005 international conference on Advances in Intelligent Computing - Volume Part II
IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews
Hi-index | 0.00 |
In this paper, a novel approach for networked control system (NCS) task scheduling is proposed. The proposed neural-genetic method utilizes the information about the quality of service (QoS) over the communication network and enables online reconfigurable scheduling on distributed environment. In this way the NCS's bandwidth can be shared properly among different parallel control tasks. For NCS, two significant factors of QoS that affect validity of scheduling results are the packet loss and delay, which occurred in the communication among tasks. By adopting a Elman neural network based prediction model, the one-step ahead packet loss and time delay are obtained. The knowledge about the predict QoS factors, combined with the task execution features and the resources available in the system, are used as an entry to improve the decisions of the proposed scheduling algorithm. Such algorithm uses genetic algorithm techniques to find out the appropriate task scheduling scheme to adapt changes of application and communication circumstance. The proposed neural-genetic approach is evaluated through simulation by using a model parameterized with the features obtained from a real scenario of Ethernet based control system. The simulation results clearly show the effectiveness of the proposed approach in solving the task scheduling problems in NCS.