Research: Project 1

Rapid Design and Implementation of Embedded Real-Time Control Systems

Goal

To streamline the process of implementing real-time feedback control system designs on embedded hardware

Background

The design of feedback control systems usually includes the following steps:

  1. System identification

  2. Designing and simulating a controller on a PC

  3. Implementing a prototype controller on an PC using an A/D board

  4. Converting the prototype controller to an embedded solution

This last step often causes a significant bottle neck in the design of real-time control systems.

Approach

This project will automate the process of converting a computer simulation of a control system to an embedded solution so that the need to implement prototype controllers on A/D boards is eliminated. This project will basically walk through the control design process, from system identification to embedded implementation, for several systems, intelligently automating at each step.

  • Use Python on a PC for automated processing and analyzing of system ID data and fitting that data to an object-oriented model of the physical system

  • Develop a Python module for feedback control design and simulation building on the existing Scipy suite of tools

  • Use Python to auto-generate C code for implementing the controller on a PSoC (Programmable System-on-Chip)

  • Compare this process to implementing the controller using the PSoC Express software from Cypress

  • Investigate the possibility of Python auto-generating an input file for PSoC Express that could also be user edited

Project Outcome

A suite of software tools that intelligently automate the control design process and integrate well with embedded hardware