Skip to content

ROADMAP.md

Latest commit

 

History

History
302 lines (244 loc) · 8.41 KB

ROADMAP.md

File metadata and controls

302 lines (244 loc) · 8.41 KB

CppPlot Development Roadmap

🎯 Vision

CppPlot - Thư viện C++ plotting chuyên nghiệp, header-only, không phụ thuộc, tối ưu cho hệ thống embedded, real-time, và scientific computing.

📊 Architecture Overview

cppplot/
├── include/cppplot/
│   ├── cppplot.hpp           # Main header (includes all)
│   ├── pyplot.hpp            # Matplotlib-style API
│   ├── figure.hpp            # Figure management
│   ├── axes.hpp              # Axes and plotting
│   │
│   ├── core/                 # Core utilities
│   │   ├── types.hpp
│   │   ├── color.hpp
│   │   ├── style.hpp
│   │   ├── utils.hpp
│   │   └── latex.hpp
│   │
│   ├── backends/             # Output backends
│   │   ├── svg_backend.hpp   # ✅ Done
│   │   ├── pdf_backend.hpp   # 🔜 v2.5
│   │   └── png_backend.hpp   # 🔜 v3.0
│   │
│   ├── control/              # 🔜 v1.5 - Control Systems
│   │   ├── control.hpp       # Main control header
│   │   ├── transfer_function.hpp
│   │   ├── state_space.hpp
│   │   ├── bode.hpp
│   │   ├── nyquist.hpp
│   │   ├── root_locus.hpp
│   │   └── time_response.hpp
│   │
│   ├── signal/               # 🔜 v2.0 - Signal Processing
│   │   ├── signal.hpp
│   │   ├── fft.hpp
│   │   ├── spectrogram.hpp
│   │   ├── filter.hpp
│   │   └── window.hpp
│   │
│   ├── stats/                # 🔜 v2.0 - Statistics
│   │   ├── stats.hpp
│   │   ├── histogram.hpp
│   │   ├── boxplot.hpp
│   │   ├── violin.hpp
│   │   └── heatmap.hpp
│   │
│   └── plot3d/               # 🔜 v3.0 - 3D Plotting
│       ├── plot3d.hpp
│       ├── surface.hpp
│       ├── contour.hpp
│       └── wireframe.hpp
│
├── examples/
│   ├── basic_demo.cpp
│   ├── subplot_demo.cpp
│   ├── control_system_demo.cpp    # ✅ Done
│   ├── signal_processing_demo.cpp
│   └── statistics_demo.cpp
│
└── tests/

🗓️ Release Schedule

✅ v1.0 - Core (RELEASED)

  • Line plots, scatter, bar, histogram
  • SVG backend
  • Format strings (matplotlib-style)
  • Subplots and layouts
  • GridSpec
  • Error bars, fill_between
  • Log scale
  • Annotations (text, axhline, axvline)
  • Basic LaTeX rendering

✅ v1.5 - Control Systems (COMPLETE)

Status: All planned features implemented

Phase 1: Transfer Function Library ✅

  • TransferFunction class (polynomials num/den)
  • StateSpace class (A, B, C, D matrices)
  • Conversion: TF ↔ SS ↔ ZPK
  • Series, parallel, feedback connections
  • Stability analysis (poles, zeros)

Phase 2: Frequency Domain ✅

  • bode() - Automatic Bode plot function
  • nyquist() - Automatic Nyquist plot
  • nichols() - Nichols chart
  • Gain margin, phase margin calculation
  • Bandwidth calculation

Phase 3: Time Domain ✅

  • step() - Step response
  • impulse() - Impulse response
  • initial() - Initial condition response
  • lsim() - Arbitrary input simulation
  • Rise time, settling time, overshoot (stepinfo())

Phase 4: Root Locus & Design ✅

  • rlocus() - Automatic root locus
  • pzmap() - Pole-zero map
  • sisotool()-style interactive (basic)
  • Lead/Lag compensator design helpers

Phase 5: Advanced Control (BONUS - COMPLETE) ✅

  • Discrete-time systems (DiscreteTransferFunction, c2d_tustin, c2d_zoh)
  • LQR/DLQR controller design
  • Kalman filter and LQG
  • MPC (Model Predictive Control)
  • H∞ synthesis (robust control)
  • Sliding mode control (nonlinear)
  • μ-analysis (uncertainty modeling)

Phase 6: Matrix Library (COMPLETE) ✅

  • Standalone Matrix class (no Eigen dependency)
  • LU, QR, Hessenberg, Schur decompositions
  • Eigenvalue computation (any size)
  • Matrix exponential (Padé with scaling)
  • Condition number, spectral radius
  • Cholesky decomposition, positive definiteness check
  • Lyapunov/Sylvester equation solvers

📅 v2.0 - Signal & Statistics (Month 2-3)

  • FFT (using kiss_fft or custom)
  • Spectrogram, PSD
  • Filter design (Butterworth, Chebyshev)
  • Window functions
  • Advanced histograms
  • Boxplot, violin plot
  • Heatmap, correlation matrix
  • Scatter matrix

📅 v2.5 - Enhanced Output (Month 4-5)

  • PDF backend (using libharu or custom)
  • Better font handling
  • Full LaTeX support (equations)
  • Color maps (viridis, plasma, etc.)
  • Custom themes

📅 v3.0 - 3D & Advanced (Month 6+)

  • 3D surface plots
  • Contour plots
  • Wireframe
  • 3D scatter
  • PNG backend (stb_image_write)

📋 v1.5 Detailed Checklist

Transfer Function Module

// Target API
#include <cppplot/control/control.hpp>
using namespace cppplot::control;

// Create transfer function: G(s) = (s+1) / (s^2 + 2s + 1)
TransferFunction G({1, 1}, {1, 2, 1});

// Or from zeros, poles, gain
auto G2 = zpk({-1}, {-1, -1}, 1.0);

// State-space
StateSpace sys(A, B, C, D);

// Conversions
auto ss = tf2ss(G);
auto tf = ss2tf(sys);

// Connections
auto G_series = G1 * G2;           // Series
auto G_parallel = G1 + G2;         // Parallel
auto G_feedback = feedback(G, H);  // Closed-loop

Plotting API

// One-liner plots (Python Control style)
bode(G);                    // Auto Bode plot
bode({G1, G2, G3});         // Multiple systems
bode(G, omega);             // Custom frequency range

nyquist(G);                 // Auto Nyquist
nyquist(G, opts({{"arrows", "true"}}));

step(G);                    // Step response
step(G, t);                 // Custom time vector
step({G1, G2});             // Compare systems

rlocus(G);                  // Root locus
rlocus(G, kvect);           // Custom K range

pzmap(G);                   // Pole-zero map
pzmap({G1, G2});            // Compare

Analysis Functions

// Margins
auto [gm, pm, wgc, wpc] = margin(G);

// Time-domain specs  
auto info = stepinfo(G);
// info.rise_time, info.settling_time, info.overshoot, info.peak

// Stability
bool stable = isstable(G);
auto p = poles(G);
auto z = zeros(G);

// Bandwidth
double wb = bandwidth(G);

🔧 Implementation Priority

HIGH Priority (Core functionality) ✅ ALL COMPLETE

  1. ✅ TransferFunction class
  2. ✅ bode() function
  3. ✅ nyquist() function
  4. ✅ step() function
  5. ✅ pzmap() function
  6. ✅ rlocus() function
  7. ✅ margin() function
  8. ✅ stepinfo() function

MEDIUM Priority (Convenience) ✅ ALL COMPLETE

  • ✅ StateSpace class
  • ✅ impulse() function
  • ✅ nichols() chart
  • ✅ bandwidth() function

LOW Priority (Advanced) ✅ ALL COMPLETE

  • ✅ lsim() arbitrary input
  • ✅ Discrete-time systems
  • sisotool() interactive (not planned)

📁 File Structure for v1.5

include/cppplot/control/
├── control.hpp              # Main header - includes all
├── transfer_function.hpp    # TF class
├── state_space.hpp          # SS class (optional for v1.5)
├── analysis.hpp             # margin, stepinfo, poles, zeros
├── plots/
│   ├── bode.hpp            # bode() function
│   ├── nyquist.hpp         # nyquist() function
│   ├── root_locus.hpp      # rlocus() function
│   ├── pzmap.hpp           # pzmap() function
│   └── time_response.hpp   # step(), impulse(), lsim()
└── utils/
    ├── polynomial.hpp       # Polynomial arithmetic
    └── frequency.hpp        # logspace, unwrap, etc.

🎯 Success Criteria for v1.5

  1. API Simplicity: One-liner for common plots

    bode(TransferFunction({1}, {1, 2, 1}));

  2. Compatibility: Follow Python Control naming conventions

  3. Performance: Fast enough for real-time visualization

  4. Documentation: Every function documented with examples

  5. Testing: Unit tests for all transfer function operations

📚 References

Last updated: January 2026