Automated Steering with Real-Time Model-Based Control

2020 ◽  
Author(s):  
Nazli Demirer ◽  
Umut Zalluhoglu ◽  
Julien Marck ◽  
Robert Darbe
Keyword(s):  
Real Time ◽  
Time Model ◽  
Model Based Control ◽  
Model Based

Closed Loop Control of Resin Flow in VARTM Using a Multi-Segment Injection Line and Real-Time Adaptive, Model-Based Control

Materials ◽  
2005 ◽  
Author(s):  
Ajit R. Nalla ◽  
James L. Glancey
Keyword(s):  
Real Time ◽  
Closed Loop ◽  
Performance Metrics ◽  
Void Formation ◽  
Resin Flow ◽  
Time Model ◽  
Loop Control ◽  
Model Based Control ◽  
Model Based ◽  
Injection Line

To improve process controllability during VARTM, a new resin injection line was designed and tested. The injection line, which consists of multiple segments each independently operated, allows for the control of resin flow to different locations within the mold. Simulation of different injection line configurations for various mold geometries is studied. Performance of a prototype line is quantified with a laboratory size mold used to demonstrate the potential value and benefits of this approach. Specific performance metrics, including resin flow front controllability, total injection time and void formation are used to compare this new approach to conventional VARTM injection methods. Computer-based closed loop controller strategies are designed that use point sensor feedback of resin location. In addition, an adaptive control algorithm that uses a finite element model to provide real-time updates of the injection line configuration is presented. Experimental validation of two different control strategies is presented, and demonstrates that real-time, model-based control is possible in VARTM.

Distributed Network System for Real-Time Model Based Control of Industrial Gas Turbines

2011 ◽  
Author(s):  
V. Panov
Keyword(s):  
Control System ◽  
Real Time ◽  
Gas Turbine ◽  
Distributed Network ◽  
Hardware In The Loop ◽  
Time Model ◽  
Model Based Control ◽  
Model Based ◽  
On Line ◽  
Dedicated Hardware

This paper describes the development of a distributed network system for real-time model based control of industrial gas turbine engines. Distributed control systems contribute toward improvements in performance, testability, control system maintainability and overall life-cycle cost. The goal of this programme was to offer a modular platform for improved model based control system. Hence, another important aspect of this programme was real-time implementation of non-linear aero-thermal gas turbine models on a dedicated hardware platform. Two typical applications of real-time engine models, namely hardware-in-the-loop simulations and on-line co-simulations, have been considered in this programme. Hardware-in-the-loop platform has been proposed as a transitional architecture, which should lead towards a fully distributed on-line model based control system. Distributed control system architecture offers the possibility of integrating a real-time on-line engine model embedded within a dedicated hardware platform. Real-time executing models use engine operating conditions to generate expected values for measured and non-measured engine parameters. These virtual measurements can be used for the development of model based control methods, which can contribute towards improvements in engine stability, performance and life management. As an illustration of model based control concept, the example of gas turbine transient over-temperature protection is presented in this study.

scholarly journals Implicit Contact Dynamics Modeling With Explicit Inertia Matrix Representation for Real-Time, Model-Based Control in Physical Environment

2021 ◽  
pp. 1-18
Author(s):  
Takeshi D. Itoh ◽  
Koji Ishihara ◽  
Jun Morimoto
Keyword(s):  
Neural Network ◽  
Real Time ◽  
Network Model ◽  
Matrix Representation ◽  
Contact Dynamics ◽  
Time Model ◽  
Model Based Control ◽  
Model Based ◽  
Inertia Matrix ◽  
Manipulation Task

Model-based control has great potential for use in real robots due to its high sampling efficiency. Nevertheless, dealing with physical contacts and generating accurate motions are inevitable for practical robot control tasks, such as precise manipulation. For a real-time, model-based approach, the difficulty of contact-rich tasks that requires precise movement lies in the fact that a model needs to accurately predict forthcoming contact events within a limited length of time rather than detect them afterward with sensors. Therefore, in this study, we investigate whether and how neural network models can learn a task-related model useful enough for model-based control, that is, a model predicting future states, including contact events. To this end, we propose a structured neural network model predicting a control (SNN-MPC) method, whose neural network architecture is designed with explicit inertia matrix representation. To train the proposed network, we develop a two-stage modeling procedure for contact-rich dynamics from a limited number of samples. As a contact-rich task, we take up a trackball manipulation task using a physical 3-DoF finger robot. The results showed that the SNN-MPC outperformed MPC with a conventional fully connected network model on the manipulation task.

Sign in / Sign up

Export Citation Format

Share Document