Using SIGNAL for developing neural control systems

We present in this paper a new environment for developing neural networks integrated in process control loops. This environment allows the user to easily control the learning phase and assures the user that the obtained system is both logically and temporally correct. Those two aspects have to be proven in order to obtain a viable real-time system. The conventional languages can not provide the temporal aspect, so we have investigated the development of neural networks for control process with the synchronous languages and particularly the language SIGNAL and its developing environment SILDEX.<<ETX>>

[1]  Carlos Canudas de Wit,et al.  Performance oriented robust nonlinear control for subsea robots: experimental validation , 1994, Proceedings of the 1994 IEEE International Conference on Robotics and Automation.

[2]  Philip D. Wasserman,et al.  Advanced methods in neural computing , 1993, VNR computer library.

[3]  Thierry Gautier,et al.  Programming real-time applications with SIGNAL , 1991, Proc. IEEE.

[4]  Albert Benveniste,et al.  The synchronous approach to reactive and real-time systems , 1991 .