- Dettagli della rivista
- Formato
- Rivista
- eISSN
- 2083-8492
- Prima pubblicazione
- 05 Apr 2007
- Frequenza di pubblicazione
- 4 volte all'anno
- Lingue
- Inglese
The paper concerns the design of a framework for implementing fault-tolerant control of hybrid assembly systems that connect human operators and fully automated technical systems. The main difficulty in such systems is related to delays that result from objective factors influencing human operators’ work, e.g., fatigue, experience, etc. As the battery assembly system can be considered a firm real-time one, these delays are treated as faults. The presented approach guarantees real-time compensation of delays, and the fully automated part of the system is responsible for this compensation. The paper begins with a detailed description of a battery assembly system in which two cooperating parts can be distinguished: fully automatic and semi-automatic. The latter, nonderministic in nature, is the main focus of this paper. To describe and analyze the states of the battery assembly system, instead of the most commonly used simulation, the classic max-plus algebra with an extension allowing one to express non-deterministic human operators’ work is used. In order to synchronize tasks and schedule (according to the reference schedule) automated and human operators’ tasks, it is proposed to use a wireless IoT platform called KIS.ME. As a result, it allows a reference model of human performance to be defined using fuzzy logic. Having such a model, predictive delays tolerant planning is proposed. The final part of the paper presents the achieved results, which clearly indicate the potential benefits that can be obtained by combining the wireless KIS.ME architecture (allocated in the semi-automatic part of the system) with wired standard production networks.
Keywords
- parallel system
- synchronization
- scheduling
- discrete event system
- wireless equipment
Global Behavior of a Multi–Group Seir Epidemic Model with Spatial Diffusion in a Heterogeneous Environment Bootstrap Methods for Epistemic Fuzzy Data Revisiting Strategies for Fitting Logistic Regression for Positive and Unlabeled Data Joint Feature Selection and Classification for Positive Unlabelled Multi–Label Data Using Weighted Penalized Empirical Risk Minimization Hybrid Deep Learning Model–Based Prediction of Images Related to Cyberbullying Fault–Tolerant Tracking Control for a Non–Linear Twin–Rotor System Under Ellipsoidal Bounding A Multi–Model Based Adaptive Reconfiguration Control Scheme for an Electro–Hydraulic Position Servo System Reliability–Aware Zonotopic Tube–Based Model Predictive Control of a Drinking Water Network A Graph Theory–Based Approach to the Description of the Process and the Diagnostic System On Some Ways to Implement State–Multiplicative Fault Detection in Discrete–Time Linear Systems A Kalman Filter with Intermittent Observations and Reconstruction of Data Losses Parameter Identifiability for Nonlinear LPV Models