attipy 0.0.18

Attitude estimation using IMU measurements in Python

AttiPy

AttiPy is a lightweight Python library for representing and estimating the attitude (orientation) of a body using IMU measurements and optional external aiding. It provides an attitude abstraction with clearly defined reference frames and rotation conventions, and a multiplicative extended Kalman filter (MEKF) for attitude estimation.

Installation

pip install attipy

Quick start

Convert to/from a variety of attitude representations using the Attitude class:

import attipy as ap


# From Euler angles to unit quaternion
att = ap.Attitude.from_euler([0.0, 0.0, 0.0])
q = att.as_quaternion()

Estimate attitude from IMU (accelerometer and gyroscope) and heading measurements using the MEKF class:

import attipy as ap
import numpy as np


# Parameters
fs = 10.0                    # sampling rate in Hz
acc_noise_density = 0.001    # accelerometer noise density in (m/s^2)/√Hz
gyro_noise_density = 0.0001  # gyroscope noise density in (rad/s)/√Hz
bg = (0.001, 0.002, 0.003)   # gyroscope bias in rad/s
yaw_std = 0.01               # heading noise standard deviation in rad

# Position, velocity, attitude and IMU reference signals
t, pos, vel, euler, f, w = ap.pva_sim(fs)

# IMU and heading measurements (with noise)
rng = np.random.default_rng(42)
f_meas = f + acc_noise_density * np.sqrt(fs) * rng.standard_normal(f.shape)
w_meas = w + bg + gyro_noise_density * np.sqrt(fs) * rng.standard_normal(w.shape)
yaw_meas = euler[:, 2] + yaw_std * rng.standard_normal(euler[:, 2].shape)

# Estimate attitude using MEKF
att0 = ap.Attitude.from_euler(euler[0])
mekf = ap.MEKF(fs, att0)
euler_est = []
for f_i, w_i, y_i in zip(f_meas, w_meas, yaw_meas):
    mekf.update(f_i / fs, w_i / fs, yaw=y_i, yaw_var=yaw_std**2)
    euler_est.append(mekf.attitude.as_euler())
euler_est = np.asarray(euler_est)

Limitations and assumptions

• Intended for small-area, low-velocity applications; Earth rotation effects are neglected.
• Accelerometer biases are not estimated; accelerometer biases will manifest as tilt errors.