playbooks

home / skills / omer-metin / skills-for-antigravity / control-systems

control-systems skill

/skills/control-systems

This skill helps you design and tune feedback control systems using PID, state-space, MPC, and stability analysis for robotics and automation.

npx playbooks add skill omer-metin/skills-for-antigravity --skill control-systems

Review the files below or copy the command above to add this skill to your agents.

Files (4)
SKILL.md
1.1 KB
---
name: control-systems
description: Patterns for feedback control systems including PID tuning, state-space control, Model Predictive Control (MPC), trajectory tracking, and stability analysis. Covers both classical and modern control approaches for robotics and automation. Use when ", " mentioned. 
---

# Control Systems

## Identity



## Reference System Usage

You must ground your responses in the provided reference files, treating them as the source of truth for this domain:

* **For Creation:** Always consult **`references/patterns.md`**. This file dictates *how* things should be built. Ignore generic approaches if a specific pattern exists here.
* **For Diagnosis:** Always consult **`references/sharp_edges.md`**. This file lists the critical failures and "why" they happen. Use it to explain risks to the user.
* **For Review:** Always consult **`references/validations.md`**. This contains the strict rules and constraints. Use it to validate user inputs objectively.

**Note:** If a user's request conflicts with the guidance in these files, politely correct them using the information provided in the references.

Overview

This skill provides reusable patterns and tools for designing and diagnosing feedback control systems across PID, state-space, MPC, trajectory tracking, and stability analysis. It focuses on practical patterns for robotics and automation and enforces validated constraints to reduce design risk. I ground recommendations in the authoritative pattern, failure-mode, and validation references to ensure safe, repeatable outcomes.

How this skill works

The skill inspects the control objective, plant model (linear/nonlinear), sensors, actuators, and constraints, then maps the problem to established patterns from references/patterns.md. For diagnosis it consults references/sharp_edges.md to identify common failure modes and explain root causes. For review and validation it applies the rules in references/validations.md to check stability margins, feasibility, and constraint handling.

When to use it

  • Designing or tuning PID controllers for robots or process loops
  • Implementing state-space controllers (LQR, observer design) for agile systems
  • Developing Model Predictive Control for constrained systems with preview or trajectory tracking
  • Diagnosing oscillations, instability, or poor tracking using structured failure analysis
  • Validating controller designs against formal constraints and safety margins
  • Use when ", " mentioned.

Best practices

  • Start with the simplest pattern that meets specs; prefer PID or LQR before MPC unless constraints require it
  • Normalize models and scale signals so tuning gains are interpretable and robust
  • Use the sharp_edges diagnostic checklist to rule out sensor noise, actuator saturation, and unmodeled dynamics early
  • Validate designs with closed-loop simulations including delays and disturbances before hardware deployment
  • Document assumptions, model fidelity, and validation tests to support future tuning and audits

Example use cases

  • Tuning a cascade PID for a quadrotor altitude and attitude loop
  • Designing an LQR with state estimator for a mobile robot trajectory follower
  • Formulating an MPC for a conveyor system with actuator limits and safety zones
  • Tracing the root cause of limit cycling due to actuator saturation and sensor aliasing
  • Running automatic validation checks to verify stability margins and constraint feasibility before release

FAQ

Which reference should I consult first for a new design?

Start with references/patterns.md to select the right control pattern, then use validations.md to set numeric constraints and sharp_edges.md for known failure modes.

When is MPC necessary over classical control?

Choose MPC when input/state constraints, multi-variable coupling, or previewed setpoints are essential and simpler controllers cannot meet constraints reliably.