lern-tracking-system/raspberry-pi/sensors/magneticSensor.py

from distutils.command.config import config
import queue
import statistics
from struct import calcsize
import numpy as np
import time
from configparser import ConfigParser
from sensors.calibration import CalibrationStateMashine
from sensors.connection import globalArduinoSlave
import logHandler

conn = globalArduinoSlave()


class MagneticSensor:
  def __init__(self, conf):
    self.conf = conf
    self.queue = queue.Queue()
    self.calibration_state = CalibrationStateMashine()
    self.success = False
    self.offset_x = float(conf["mag_sensor"]["off_x"])
    self.offset_y = float(conf["mag_sensor"]["off_y"])
    self.scale_x = float(conf["mag_sensor"]["scale_x"])
    self.scale_y = float(conf["mag_sensor"]["scale_y"])
    self.max_x = float(conf["mag_sensor"]["max_x"])
    self.max_y = float(conf["mag_sensor"]["max_y"])
    self.log_handler = logHandler.get_log_handler()
  
  def start(self):
    if not conn.isConnected():
      conn.open()
    self.success = True
    conn.addRecvCallback(self._readCb)

  def _readCb(self, raw):
    value = conn.getMagneticField()
    if value[0] >= 0 and value[1] >= 0:
      self.calibrate(value)
      position = self.calculate_position(value)
      if position != None:
        self.pass_to_gui(position + value)
  
  def start_calibration(self): ###
    self.calibration_state.reset_state()
    self.time_vals = [[],[],[]]
    self.calibration_state.next_state()
  
  def calibrate(self, value): ### öffnet erstmal popup :) # how?
     if self.calibration_state.get_state() == self.calibration_state.ACCUMULATING_1:
      self.time_vals[0].append(value[0])
      self.time_vals[1].append(value[1])
      self.time_vals[2].append(value[2])
      self.calibration_state.progress = len(self.time_vals[0]) / 2
      if len(self.time_vals[0]) >= 100:
        self.cal_values["front"][0]  = statistics.mean(self.time_vals[0])
        self.cal_values["front"][1] = statistics.mean(self.time_vals[1])
        self.cal_values["front"][2] = statistics.mean(self.time_vals[2])
        self.time_vals = [[],[],[]]
        self.calibration_state.next_state() # signal gui to get next position

      elif self.calibration_state.get_state() == self.calibration_state.ACCUMULATING_2:
        self.time_vals[0].append(value[0])
        self.time_vals[1].append(value[1])
        self.time_vals[2].append(value[2])
        self.calibration_state.progress = 50 + len(self.time_vals[0]) / 2
        if len(self.time_vals[0]) >= 100:
          self.cal_values["back"][0] = statistics.mean(self.time_vals[0])
          self.cal_values["back"][1] = statistics.mean(self.time_vals[1])
          self.cal_values["back"][2] = statistics.mean(self.time_vals[2])
        
          # all values have been captured
          
          # Hard iron distortion & Soft iron distortion 
          self.offset_x = (self.cal_values["back"][0] - self.cal_values["front"][0]) / 2
          self.offset_y = (self.cal_values["back"][1] - self.cal_values["front"][1]) / 2

          avg_delta_x = (self.cal_values["back"][0] - self.cal_values["front"][0]) / 2
          avg_delta_y = (self.cal_values["back"][1] - self.cal_values["front"][1]) / 2
          avg_delta_z = (self.cal_values["back"][2] - self.cal_values["front"][2]) / 2

          avg_delta = (avg_delta_x + avg_delta_y + avg_delta_z) / 3

          self.scale_x = avg_delta / avg_delta_x
          self.scale_y = avg_delta / avg_delta_y
          self.scale_z = avg_delta / avg_delta_z

          # max values for placeholder algorithm
          self.max_x = (self.cal_values["back"][0] - self.offset_x) * self.scale_x
          self.max_y = (self.cal_values["back"][1] - self.offset_y) * self.scale_y
        
          self.calibration_state.next_state()

          return self.offset_x, self.offset_y, self.scale_x, self.scale_y, self.max_x, self.max_y

  def read(self):
    value = conn.getMagneticField()
    return value
  
  def calculate_position(self): ###
    corrected_x = (conn.getMagneticField[0] - self.offset_x) * self.scale_x
    corrected_y = (conn.getMagneticField[1] - self.offset_y) * self.scale_y
    # placeholder algorithm (to see if the sensor even works)
    x = (corrected_x * 400) / self.max_x
    y = (corrected_y * 400) / self.max_y
    return (x, y)

  def stop(self):
    self.log_handler.log_and_print("stop magnetic sensor")
    self.success = False
    conn.close()

  def pass_to_gui(self, data):
    self.queue.put(("data", data))