Positioner & Camera interface

This library allows communication with the fiber positioner though the CAN bus and with the Basler camera via USB.

Setup

Python version: 3.7 or after
In order to use it the modules in requirement.txt must be available for the used python interpreter
The Cameras folder contains the calibration files for the cameras
If not existing, create a folder Cameras in the Config folder and paste the calibration files from Drive.

Usage

The usage is pretty simple:

• import tp_init() from tp_astro
• use cam & pos objects to acquire data and move around the robot :-)

Examples

Here are all the python functions you need to get through this practicals

Setup communication for camera & positioner

Create cam and pos objects to call the functions needed to perform this practical

import tp_astro as tp
cam, pos = tp.tp_init()

Perform a manual move

Make a move. Each axis can rotate from 0° to 360°. Once the pos object is created you can either:

• Move to an absolute angular position, say (alpha = 30°, beta = 90°)

"""
Input: None
Output: robot moves to absolute angular positons
"""
pos.goto_absolute(30,90)

• IMPORTANT: Wait for the move to finish before continuing

"""
Input: None
Output: code pauses executing until robot finishes its move
"""
pos.wait_move()

• Change the angular speed of each axis, in RPM on the motor side (speed limits: [1000, 3000] RPM)

"""
Input: alpha speed [scalar in RPM], beta speed [scalar in RPM]
Output: speed set for next moves until new change
"""
pos.set_speed(2000,2000)

• Move to a relative angular position from current position, say (alpha = 30°, beta = 90°)

"""
Input: alpha [scalar in °], beta [scalar in °]
Output: robot moves to position relative to current one
"""
pos.goto_relative(30,90)

Acquire data with camera

The camera allows you to acquire the centroid of the fiber on the image coordinate

"""
Input: None
Output: x,y = coordinates [mm, mm] of centroid of fiber on image
"""
x_centroid, y_centroid = cam.getCentroid()

Example of scripted data acquisition

This code snippet summarizes in one shot:

• Initialize commmunication objects
• Make a move
• Get position of the fiber centroid

import tp_astro as tp
cam, pos = tp.tp_init() # Initialize camera & robot communication
pos.goto_absolute(30,90) # Move robot to alpha = 30°, beta = 90°
pos.wait_move() # Wait for trajectory to finish before continuing
x_centroid, y_centroid = cam.getCentroid() # Acquire position of centroid on image
print(x_centroid, y_centroid)

Fit circle

Finally you will also by provided the fit_circle function, i.e. determine the closest circle fit from a set of data points
Note: the code is optimized to handle a large number of data points. So take at least 4 sample points for circle fitting, even though 3 are enough in theory.

from miscmath import fit_circle
"""
Input : - xData (np array), yData (np array) = x,y coordinate of sample points
Output: - center_x [mm], center_y [mm] = center of fitted circle
        - radius [mm] = radius of fitted circle
"""

center_x, center_y, radius = fit_circle(xData, yData)