Optitrack Export

This example shows how to clean the .csv files outputted by Optitrack.

import rofunc as rf
import os
import numpy as np

# If input_path points to a folder, each element of objs and meta is the data corresponding to one file.
# In a folder, only the file with the following name format are considered: 'Take*.csv'
# If input_file points to a file, objs and meta are lists with only one element.
input_path = '../data/RAW_DEMO/optitrack/Take 2023-03-29 06.23.40 PM.csv'
parent_dir = os.path.dirname(input_path)
objs_list, meta_list = rf.optitrack.get_objects(input_path)

# You can delete object from the dictionary, and they will not be considered in the data
del_objects = ['left']
for obj in del_objects:
    del objs_list[0][obj]

# data is a numpy array of shape (n_samples, n_features)
# labels is a list of strings corresponding to the name of the features
data, labels = rf.optitrack.data_clean(parent_dir, legacy=False, objs=objs_list[0])[0]

# Accessing the position and attitude of an object over all samples:
# Coordinates names and order: ['x', 'y', 'z', 'qx', 'qy', 'qz', 'qw']
data_ptr = labels.index('box.pose.x')
assert data_ptr + 6 == labels.index('box.pose.qw')
box_pos_x = data[:, data_ptr:data_ptr + 7]

# TODO: export data as the format of objects
root_dir = '/home/ubuntu/Data/optitrack_export/'
rf.oslab.create_dir(root_dir)
exp_name = 'rigid_body.npy'
np.save(os.path.join(root_dir, exp_name), data)