sight = fc.get_car_sight(name)
wight = len(sight[0])
hight = len(sight)
r, ti = 0, fc.get_ti(name)
mid = ti['mid']
if sight[mid[0][1]][mid[0][0]] < 0.5:
r = -35
else:
r = -np.abs(mid[0][0] - wight / 2)
return r
FRAME_SPACING = 20
fc.game_init('FCMAP0.PNG')
fc.add_car('0', (90, 255))
fc.set_car('0', (90, 255), np.pi / 2, 0)
#fc.show_car_sight('0')
wight = len(fc.get_car_sight('0')[0])
fc.set_time_speed(0)
ddpg = MDDPG.DDPG(11, 2, 2, np.array([np.pi / 3, 256]))
var = np.array([np.pi / 3, 100])
times = 1
fc.start()
while times:
start_time = fc.get_time()
normal_working = 0
batch_size=64,
name='target_ctrl',
op_name='ctrl_model')
value_model = tfl.regression(value,
placeholder=y,
optimizer='adam',
loss=value_loss,
trainable_vars=value_vars,
batch_size=64,
name='target_value',
op_name='value_model')
model = tfl.DNN(tf.concat([ctrl, value], 1))
fc.game_init("FCMAP0.PNG")
fc.add_car("0", (430, 240))
fc.set_car("0", (430, 240), np.pi / 2, 0)
fc.show_car_sight("0")
wight = len(fc.get_car_sight("0")[0])
train_sight = []
train_degree = []
train_velocity = []
train_y = []
degree = 0
velocity = 100
times = 1
fc.set_time_speed(1)