akihikoy

from html.parser import HTMLParser
import re

class MyHTMLParser(HTMLParser):
    def __init__(self):
        HTMLParser.__init__(self)
        self.mode= None
        self.serialized= []

    def handle_starttag(self, tag, attrs):

akihikoy

#!/usr/bin/python
#Printing current position without enabling torque.

from dxl_util import *
from _config import *
import time

#Setup the device
DXL_ID= [1,2]  #IDs of the Dynamixels.
BAUDRATE= 57600

akihikoy

#!/usr/bin/python
#Change ID.

from dxl_util import *

DXL_ID= 1  #Current ID
NEW_ID= 2  #New ID
BAUDRATE= 57600
#DXL_TYPE= 'XM430-W350'  #Finger robot
DXL_TYPE= 'XH430-V350'  #Dynamixel gripper

akihikoy

// Changing exposure worked with the ELP camera on Windows.
#include "stdafx.h"
#include <opencv2/opencv.hpp>

bool ParamChanged(false);
void OnTrack(int,void*)
{
  ParamChanged= true;
}

akihikoy

#!/usr/bin/python
from core_tool import *
import cv2
import numpy as np


def Run(ct,*arg):
  cap = cv2.VideoCapture('http://aypi11:8081/?action=stream&dummy=file.mjpg')
  #q0= np.array(ct.robot.Q())
  t0= rospy.Time.now()

akihikoy

#!/usr/bin/python
from core_tool import *
def Help():
  return '''First script.
  Usage: a.ros'''
def Run(ct,*args):
  print len(args)
  print args[0]

  q= list(ct.robot.Q())

akihikoy

'''
Following codes are an example of processing marker tracker topic from FingerVision.
I just copied them from my system, which might be incomplete, but you could find the idea of the calculation.

The main function to process the marker tracker topic is BlobMoves.
(see `def BlobMoves(t,l,msg,side)`)
t: Interface of core tools (ROS connection, robot controller, etc.); you can replace t by adequate functions.
l: A container object. The simplest implementation is like:
  class Container(): pass
  l= Container()

akihikoy

#!/usr/bin/env python
# -*- coding: utf-8 -*-

################################################################################
# Copyright (c) 2016, ROBOTIS CO., LTD.
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
#

akihikoy

#!/usr/bin/python
#NOTE: Remove and Upgrade scikit-learn (v.1.8 is necessary)
#  sudo apt-get remove python-sklearn python-sklearn-lib
#  sudo pip install -U scikit-learn

import random, math
import numpy as np
import matplotlib.pyplot as plot
from mpl_toolkits.mplot3d import Axes3D
from sklearn.gaussian_process import GaussianProcessRegressor

akihikoy

#!/usr/bin/python
#Download CMA-ES for Python from https://www.lri.fr/~hansen/cmaesintro.html
#and put cma.py in the same directory
import numpy as np
import matplotlib.pyplot as plot
from fk import FK
import cma
plot.ioff()  #cma set matplotlib.pyplot.ion()

#Error function of inverse kinematics