Interactive Notebooks

These Jupyter notebooks provide hands-on tutorials covering key pytcl functionality. Each notebook is self-contained with explanations, code examples, and visualizations.

Note

These notebooks are rendered statically in the documentation. To run them interactively, clone the repository and open them in Jupyter:

git clone https://github.com/nedonatelli/TCL.git
cd TCL
pip install -e ".[dev]"
jupyter notebook docs/notebooks/

Filtering & Estimation

• Kalman Filters: From Fundamentals to Advanced Filtering
• Particle Filters: Sequential Monte Carlo Methods

These notebooks cover state estimation fundamentals from linear Kalman filtering to particle filters for nonlinear/non-Gaussian systems.

Multi-Target Tracking

• Multi-Target Tracking: Data Association and Tracking

Learn data association algorithms including GNN, JPDA, and track management for tracking multiple targets simultaneously.

Coordinate Systems

• Coordinate Systems: Transformations and Projections

Comprehensive coverage of geodetic, ECEF, ENU/NED frames, rotations, quaternions, and map projections.

Navigation

• INS/GNSS Integration: Navigation System Fusion

INS/GNSS integration covering strapdown mechanization, loosely-coupled fusion, and GNSS outage handling.

Advanced Topics

• GPU Acceleration: CuPy Integration for Tracking
• Network Flow for Assignment Problems
• Performance Optimization: Profiling and Benchmarking

GPU acceleration with CuPy, network flow algorithms for assignment problems, and performance optimization techniques.

Notebook Summary

#	Notebook	Description
1	Kalman Filters: From Fundamentals to Advanced Filtering	Linear KF, Extended KF, Unscented KF, Cubature KF, IMM estimator
2	Particle Filters: Sequential Monte Carlo Methods	Bootstrap PF, resampling strategies, effective sample size, degeneracy
3	Multi-Target Tracking: Data Association and Tracking	Data association, GNN, JPDA, track management, OSPA metrics
4	Coordinate Systems: Transformations and Projections	Geodetic/ECEF/ENU/NED, rotations, quaternions, map projections
5	GPU Acceleration: CuPy Integration for Tracking	CuPy basics, batch processing, particle filter GPU acceleration
6	Network Flow for Assignment Problems	Min-cost flow for assignment, successive shortest paths, Hungarian comparison
7	INS/GNSS Integration: Navigation System Fusion	INS mechanization, GNSS DOP, loosely-coupled integration, outage handling
8	Performance Optimization: Profiling and Benchmarking	Profiling, Numba JIT, vectorization, caching, memory optimization

Prerequisites

Most notebooks require only the core pytcl package:

pip install nrl-tracker matplotlib

Some advanced notebooks have additional requirements:

• GPU Acceleration: pip install cupy-cuda12x (NVIDIA GPU with CUDA)

• Network Flow: No additional requirements

• Performance: pip install numba line_profiler memory_profiler

Getting Started

1. Start with Kalman Filters (01) if you’re new to state estimation

2. Move to Particle Filters (02) for nonlinear problems

3. Explore Multi-Target Tracking (03) for data association

4. Study Coordinate Systems (04) for navigation applications

5. Advanced users: GPU acceleration, network flow, and optimization