20.12 Research Roadmap - Control Co-Design

Core idea¶

Research Roadmap must be treated as a system-level decision rather than an isolated technique. For a reproducible repository with versioned models, data, formulations, solvers, tests, results, and hardware evidence, state what is fixed, what is optimized, what information is available, and what equations define feasibility.

The relevant quantities are interfaces, design and trajectory data, derivatives, metadata, seeds, tolerances, and provenance. The chapter-level formulation is

\mathrm{reproducibility}=\mathrm{model}+\mathrm{data}+\mathrm{configuration}+\mathrm{environment}+\mathrm{evidence}.

For this section, trace how the choice changes model–solver interfaces, the active constraints, and the implementable engineering design. A method is useful only when its assumptions are explicit and its result answers the same system question as the baseline.

Engineering interpretation¶

Ask three questions:

Which physical, informational, computational, or economic resource changed?
Which objective component or active constraint made the change valuable?
Does the conclusion survive model, disturbance, initialization, uncertainty, and implementation checks?

A practical action is to publish interfaces. Record units and assumptions before optimization, report component objectives and margins afterward, and verify the result using an independent calculation or higher-fidelity model.

Activity 20.12: build a quantitative research roadmap¶

Activity 20.12: build a quantitative research roadmap

A research group has 30 person-months for the next program cycle. Candidate projects are:

Project	Effort	Impact	Success probability	Reproducibility gain	Dependency
A: benchmark suite	8	8	0.90	9	—
B: differentiable models	10	9	0.65	6	—
C: hardware dataset	12	10	0.70	8	—
D: learning-based CCD	9	8	0.55	5	B
E: interface standard	6	7	0.85	10	—
F: digital-twin demo	11	9	0.60	7	A and C

Define expected value $V=P_s(0.7I+0.3R)$ , where $I$ is impact and $R$ reproducibility gain. Dependencies must be funded in the same cycle.

Compute $V$ and value per person-month for every project.
Enumerate feasible portfolios within 30 person-months, enforce dependencies, and select the maximum-total-value portfolio.
Require at least one project with reproducibility gain $\geq9$ and one project involving hardware. Reoptimize the portfolio.
If Project C effort rises to 15 and success probability falls to 0.55, recompute the constrained roadmap.
Allocate a 10% management reserve, leaving 27 person-months. Quantify lost expected value and identify which dependency creates the largest scheduling risk.

Chapter summary¶

The chapter connected benchmarks and reports, model–solver interfaces, differentiable physics and learning, system synthesis, research roadmap through one system formulation. Engineering conclusions require aligned models, information, numerical accuracy, and validation.

Common mistakes¶

changing assumptions while comparing alternatives;
reporting objective improvement without verified feasibility;
hiding information, architecture, or uncertainty;
treating solver convergence as validation; and
reporting runtime without accuracy, derivatives, and tolerances.

Exercises¶

Recreate the workflow for a reproducible repository with versioned models, data, formulations, solvers, tests, results, and hardware evidence.
State every variable, unit, dependency, and constraint.
Construct a common sequential or nominal baseline.
Identify active constraints and the physical bottleneck.
Design a test that could falsify the claimed benefit.

Principal sources¶

The wind CCD roadmap, UCCD taxonomy, information-availability work, and numerical CCD studies.

Open research question¶

What minimum common standard makes CCD studies genuinely comparable?