Pendigard's picture
Upload folder using huggingface_hub (part 5)
4b876a7 verified
Raw History Blame Contribute Delete
9.28 kB
import pigpio
import threading
from time import sleep
import logging
from rotary_encoder.motorencoder import MotorEncoder
class WheelsAxel:
""" Class that handles both motor encoders, left and right
This class works like a wheels axle, coordinating left and right
wheels at the same time
It also tries to handle the inconsistent tension on wheels
that makes one wheel go slower than the other """
def __init__(self, pi, enable_pin,
left_forward_pin, left_backward_pin, left_encoder_feedback_pin_A, left_encoder_feedback_pin_B,
right_forward_pin, right_backward_pin, right_encoder_feedback_pin_A, right_encoder_feedback_pin_B):
# state variables
self._is_moving = False
# left motor
self._left_motor = MotorEncoder(pi,
enable_pin,
left_forward_pin,
left_backward_pin,
left_encoder_feedback_pin_A,
left_encoder_feedback_pin_B)
# right motor
self._right_motor = MotorEncoder(pi,
enable_pin,
right_backward_pin,
right_forward_pin,
right_encoder_feedback_pin_A,
right_encoder_feedback_pin_B)
# other
#self._wheelsAxle_lock = threading.RLock() # race condition lock
# STATE GETTERS
""" Distance and speed are calculated by a mean of the feedback
from the two motors """
def is_moving(self):
return self._left_motor.is_moving() or self._right_motor.is_moving()
# distance
def distance(self):
l_dist = self._left_motor.distance()
r_dist = self._right_motor.distance()
return (l_dist + r_dist) * 0.5
#speed
def speed(self):
l_speed = self._left_motor.speed()
r_speed = self._right_motor.speed()
return (l_speed + r_speed) * 0.5
#direction
def direction(self):
l_dir = self._left_motor.direction()
r_dir = self._right_motor.direction()
if(l_dir == r_dir):
return l_dir
else:
return 0
# MOVEMENT
""" Movement wrapper method
if time is specified and distance is not, control_time is called
if distance is specified and time is not, control_distance is called
if both distance and time are specified, control_velocity is called """
def control(self, power_left=100, power_right=100, time_elapse=0, target_distance=0):
if(time_elapse != 0 and target_distance == 0): # time
self.control_time(power_left, power_right, time_elapse)
elif(time_elapse == 0 and target_distance != 0): # distance
self.control_distance(power_left, power_right, target_distance)
else: # velocity
self.control_velocity(time_elapse, target_distance)
""" Motor time control allows the motors
to run for a certain amount of time """
def control_time(self, power_left=100, power_right=100, time_elapse=0):
#self._wheelsAxle_lock.acquire() # wheelsAxle lock acquire
# applying tension to motors
self._left_motor.control(power_left, -1)
self._right_motor.control(power_right, -1)
self._is_moving = True
# moving for desired time
# fixed for direct control that uses time_elapse -1 and stops manually
if(time_elapse > 0):
sleep(time_elapse)
self.stop()
""" Motor distance control allows the motors
to run for a certain amount of distance (mm) """
def control_distance(self, power_left=100, power_right=100, target_distance=0):
#self._wheelsAxle_lock.acquire() # wheelsAxle lock acquire
self._is_moving = True
# get desired direction from power, then normalize on power > 0
left_direction = power_left/abs(power_left)
right_direction = power_right/abs(power_right)
power_left = abs(power_left)
power_right = abs(power_right)
self._left_motor.reset_state()
self._right_motor.reset_state()
# applying tension to motors
self._left_motor.control(power_left * left_direction)
self._right_motor.control(power_right * right_direction)
#PID parameters
# assuming that power_right is equal to power_left and that coderbot
# moves at 11.5mm/s at full PWM duty cycle
MAX_SPEED = 180
target_speed_left = (MAX_SPEED / 100) * power_left #velocity [mm/s]
target_speed_right = (MAX_SPEED / 100) * power_right # velocity [mm/s]
# SOFT RESPONSE
#KP = 0.04 #proportional coefficient
#KD = 0.02 # derivative coefficient
#KI = 0.005 # integral coefficient
# MEDIUM RESPONSE
KP = 0.4 #proportional coefficient
KD = 0.1 # derivative coefficient
KI = 0.02 # integral coefficient
# STRONG RESPONSE
#KP = 0.9 # proportional coefficient
#KD = 0.05 # derivative coefficient
#KI = 0.03 # integral coefficient
SAMPLETIME = 0.01
left_derivative_error = 0
right_derivative_error = 0
left_integral_error = 0
right_integral_error = 0
# moving for certaing amount of distance
logging.info("moving? " + str(self._is_moving) + " distance: " + str(self.distance()) + " target: " + str(target_distance))
while(abs(self.distance()) < abs(target_distance) and self._is_moving == True):
# PI controller
logging.info("speed.left: " + str(self._left_motor.speed()) + " speed.right: " + str(self._right_motor.speed()))
if(abs(self._left_motor.speed()) > 10 and abs(self._right_motor.speed()) > 10):
# relative error
left_error = (target_speed_left - self._left_motor.speed()) / target_speed_left * 100.0
right_error = (target_speed_right - self._right_motor.speed()) / target_speed_right * 100.0
left_correction = (left_error * KP) + (left_derivative_error * KD) + (left_integral_error * KI)
right_correction = (right_error * KP) + (right_derivative_error * KD) + (right_integral_error * KI)
corrected_power_left = power_left + left_correction - right_correction
corrected_power_right = power_right + right_correction - left_correction
#print("LEFT correction: %f" % (left_error * KP + left_derivative_error * KD + left_integral_error * KI))
#print("RIGHT correction: %f" % (right_error * KP + right_derivative_error * KD + right_integral_error * KI))
# conrispondent new power
power_left_norm = max(min(corrected_power_left, 100), 0)
power_right_norm = max(min(corrected_power_right, 100), 0)
logging.info("ls:" + str(int(self._left_motor.speed())) + " rs: " + str(int(self._right_motor.speed())) +
" le:" + str(int(left_error)) + " re: " + str(int(right_error)) +
" lc: " + str(int(left_correction)) + " rc: " + str(int(right_correction)) +
" lp: " + str(int(power_left_norm)) + " rp: " + str(int(power_right_norm)))
# adjusting power on each motors
self._left_motor.adjust_power(power_left_norm * left_direction )
self._right_motor.adjust_power(power_right_norm * right_direction)
left_derivative_error = left_error
right_derivative_error = right_error
left_integral_error += left_error
right_integral_error += right_error
# checking each SAMPLETIME seconds
sleep(SAMPLETIME)
logging.info("control_distance.stop, target dist: " + str(target_distance) +
" actual distance: " + str(self.distance()) +
" l ticks: " + str(self._left_motor.ticks()) +
" r ticks: " + str(self._right_motor.ticks()))
# robot arrived
self.stop()
""" Motor speed control to travel given distance
in given time adjusting power on motors
NOT very intuitive, idea has been postponed"""
def control_velocity(self, time_elapse=0, target_distance=0):
pass
""" The stop function calls the two stop functions of the two
correspondent motors.
Locks are automatically obtained """
def stop(self):
# stopping left and right motors
self._left_motor.stop()
self._right_motor.stop()
# trying to fix distance different than zero after
# wheels has stopped by re-resetting state after 0.5s
#self._left_motor.reset_state()
#self._right_motor.reset_state()
# updating state
logging.info("stopping")
self._is_moving = False
# CALLBACK
def cancel_callback(self):
self._right_motor.cancel_callback()
self._left_motor.cancel_callback()