ACCUDRIVE

ACCUDRIVE is a Python recreation of LemLib: PID, odometry math, tank-drive motion helpers, driver control curves, pure pursuit path utilities, and a path.jerryio LemLib-format converter.

This repository is intentionally hardware-agnostic. It exposes the math and control surfaces in Python and lets robot projects provide motor/sensor adapters. That makes it useful for simulation, teaching, testing autonomous routines, and porting LemLib-style snippets before wiring them to a Python robotics runtime.

Included functions

• pid, controller_settings, and exit-condition helpers
• pose, vector_2d, angle math, slew limiting, desaturation, and tangent-arc curvature
• drivetrain, wheel, tracking_wheel, and tracking_wheel_odometry
• chassis driver helpers: tank, arcade, curvature
• chassis motion helpers: turn_to_heading, turn_to_point, move_to_point, move_to_pose, and follow
• LemLib path parsing plus ACCUDRIVE conversion for path.jerryio exports
• A docs-style website with all eight original tutorial topics adapted to Python

Local Installation

python -m pip install -e .

Example Usage

from accudrive import chassis, controller_settings, drivetrain, odom_sensors, omniwheel, pose

drive = drivetrain(
    track_width=10,
    wheel_diameter=omniwheel.new_325,
    wheels_per_side=3,
    wheel_sequence=["omni", "traction", "omni"],
    rpm=360,
    horizontal_drift=2,
)
lateral = controller_settings(kp=10, ki=0, kd=3, windup_range=3)
angular = controller_settings(kp=2, ki=0, kd=10, windup_range=3)

robot = chassis(drive, lateral, angular, odom_sensors())
robot.set_pose(pose(0, 0, 0))

signal = robot.move_to_point(0, 48)
print(signal.left, signal.right)

Conversion of path.jerryio LemLib format to ACCUDRIVE path

from accudrive.paths import convert_lemlib_to_accudrive

path_text = """0, 0, 80
12, 18, 90
24, 24, 0
endData
"""

print(convert_lemlib_to_accudrive(path_text, variable_name="auton_path"))

Website

The web app lives in website/.

cd website
npm install
npm run dev

It has an API reference with an intuitive documentation structure, all eight LemLib tutorial topics adapted to Python, and a browser-based LemLib path converter.

Attribution

This project is a Python port inspired by LemLib but completely independent of it. Many LemLib concepts have ACCUDRIVE Python counterparts, but they were developed separately. See NOTICE.md and LICENSE.