HW-SW_CoDesign4CoC

Contribution to IEEE Transactions on Control Systems Technology Issue xx, Volume xx:

A Practical Multiobjective Learning Framework for Optimal Hardware-Software Co-Design of Control-on-a-Chip Systems

DOI - 10.1109/TCST.2024.3407582

Authors: Kimberly J. Chan, Joel A. Paulson, and Ali Mesbah

If you use our work, please cite:

@article{chan2024practical,
  title={A Practical Multiobjective Learning Framework for Optimal Hardware--Software Co-Design of Control-on-a-Chip Systems},
  author={Chan, Kimberly J and Paulson, Joel A and Mesbah, Ali},
  journal={IEEE Transactions on Control Systems Technology},
  year={2024},
  publisher={IEEE}
}

Implementation

Several components of this code was implemented using Python AND MATLAB.

To run the Python code on your own device, it is recommended to work within a virtual environment. You may create your own virtual environment and then install the required Python packages by using the command pip3 install -r requirements_wo_appj_testbed.txt (for simulations only) and pip3 install -r requirements_w_appj_testbed.txt (additional packages for experiments with the atmospheric pressure plasma jet (APPJ)).

This code requires MATLAB, specifically R2021a. Additionally, the following toolboxes were installed (not all may be necessary):

• Control System Toolbox
• Embedded Coder
• Deep Learning HDL Toolbox
• Deep Learning Toolbox
• Fixed-Point Designer
• Global Optimization Toolbox
• HDL Coder
• HDL Verifier
• MATLAB Coder
• Optimization Toolbox
• Statistics and Machine Learning Toolbox Note, if the code implementation requires additional toolboxes, then the user should install those additional toolboxes.

For multi-objective optimization, MATLAB function(s) are called by a Python script. This can be set up using the instructions at this link

Finally, this code requires the installation and use of Xilinx Vivado 2020.1 during the FPGA-in-the-loop simulations and experiments.

Illustrative Example

The illustrative example can be implemented completely in MATLAB.

The main file to run comparisons for the illustrative example is toy_problem/run_comparison.m.

Control-on-a-chip for Plasma Devices

The main file to run simulations with multi-objective Bayesian optimization for the artificial plasma control system is sim_src/run_mobo_hw_ax.py.

The main file to run experiments with the in-house Mesbah Lab APPJ testbed is exp_src/run_exp.py.

Additional details may be found within the README's within the individual source code folders as well as in commentary within the files.

(c) 2024 Mesbah Lab

in collaboration with Joel A. Paulson and Ali Mesbah.

Questions regarding this code may be directed to kchan45 (at) berkeley.edu