amcl 0.1.2

AMCL Python Bindings

AMCL Python Bindings

Python bindings for Adaptive Monte Carlo Localization (AMCL) adapted from the Nav2 AMCL.

Vibe-coded with love ❤️

📦 Installation

You can install the package using pip:

pip install amcl

Or install from source:

git clone https://github.com/damanikjosh/amcl_python.git
cd amcl_python
pip install -e .

🎯 Quick Start

Basic Usage

import amcl
import numpy as np

# Create motion and laser parameters
motion_params = amcl.MotionParameters(
    alpha1=0.1,
    alpha2=0.1,
    alpha3=0.1,
    alpha4=0.1,
    alpha5=0.1
)
laser_params = amcl.LaserParameters(
    z_hit=0.5,
    z_short=0.05,
    z_max=0.05,
    z_rand=0.5,
    sigma_hit=0.2,
    lambda_short=0.1,
    chi_outlier=0.05,
    max_beams=60
)

# Create AMCL localizer
localizer = amcl.AMCL(
    min_particles=100,
    max_particles=500,
    alpha_slow=0.001,
    alpha_fast=0.1,
    motion_params=motion_params,
    laser_likelihood_max_dist=2.0,
    laser_params=laser_params,
    robot_model_type="differential"
)

# Create occupancy grid map
grid_data = [[0 for _ in range(100)] for _ in range(100)]
grid_data[50][50] = 100  # Add obstacle
grid = amcl.create_occupancy_grid_from_list(grid_data, 0.1, -5.0, -5.0)

# Set map and initial pose
localizer.set_map(grid)
initial_pose = amcl.Vector3D(0.0, 0.0, 0.0)
initial_cov = amcl.Matrix3D()
localizer.set_initial_pose(initial_pose, initial_cov)

# Set laser pose relative to robot base
localizer.set_laser_pose(amcl.Vector3D(0.0, 0.0, 0.0))

# Update with laser scan
ranges = [[i * 0.1, i * 0.017] for i in range(181)]  # [range, bearing] pairs
scan = amcl.create_laser_scan_from_list(ranges, 10.0)
odom_pose = amcl.Vector3D(0.1, 0.0, 0.0)

localizer.update(scan, odom_pose)
pose = localizer.get_pose_mean()
print(f"Estimated pose: ({pose.x:.3f}, {pose.y:.3f}, {pose.theta:.3f})")

📚 API Reference

Core Classes

• AMCL: Main localization class implementing particle filter
• Vector3D: 3D vector for poses (x, y, theta)
• Matrix3D: 3x3 matrix for covariances
• LaserScan: Laser scanner data structure
• OccupancyGrid: 2D occupancy grid map
• MotionParameters: Robot motion model parameters
• LaserParameters: Laser sensor model parameters

Utility Functions

• create_laser_scan_from_list(): Create LaserScan from Python lists
• create_occupancy_grid_from_list(): Create OccupancyGrid from Python lists
• get_version(): Get library version

NumPy Integration (amcl_numpy module)

• create_occupancy_grid_from_numpy(): Create OccupancyGrid from NumPy array
• create_laser_scan_from_numpy(): Create LaserScan from NumPy array
• occupancy_grid_to_numpy(): Convert OccupancyGrid to NumPy array
• laser_scan_to_numpy(): Convert LaserScan to NumPy array

📄 License

This project is licensed under the GPLv3 License - see the LICENSE file for details.

🙏 Acknowledgments

• Nav2 Team: Original AMCL implementation
• ROS Community: Foundational robotics algorithms
• nanobind: Excellent Python binding framework
• Claude Sonnet 4: For helping with code generation