def init():
global time, agents, gates, envir, states, orders_list, data_stats, total_stats
global envir, gates_locs, wall_locs, office_locs, charging_locs, waiting_points
# Initilizing the environement
envir, gates_locs, wall_locs, office_locs, charging_locs = envir_configuration(width, height)
# Initilizing the waiting_points
waiting_points.append(Waiting_Point(0, (6, 33)))
waiting_points.append(Waiting_Point(1, (6, 42)))
# Initilizing the gate objects
gates.append(Gate(0, (0,31), (0,32), (0,33), (6,29)))
gates.append(Gate(1, (0,37), (0,38), (0,39), (6, 38)))
gates.append(Gate(2, (0,43), (0,44), (0,45), (6, 47)))
# Initializing the agents
if(behavior_type == 1):
for i in range(0, N_AGV):
agents.append(AGV((42, i + 1 + (i*2) ), "darkred", 100 + i, []))
elif(behavior_type == 2):
for i in range(0, N_AGV):
agents.append(AGV((42, i + 1 + (i*2) ),"red", 100 + i, ["client", "status"]))
columns = orders_list.columns[2:]
n_col = len(columns)
i = 0
cont = 0
if(n_col > N_AGV):
for c in columns:
if(i < N_AGV):
temp = agents[i].articles_priority
temp.append(c)
agents[i].articles_priority = temp
i = i + 1
else:
i = 0
temp = agents[i].articles_priority
temp.append(c)
agents[i].articles_priority = temp
i = i + 1
else:
print("Error - Too many AGVs ")
elif(behavior_type == 3):
for i in range(0, N_AGV):
agents.append(AGV((42, i + 1 + (i*2) ),"red", 100 + i, []))
else:
print("Error - Behavior_Type is not existing.")
# Initializing the dataframe that is going to be used in order to collect the data about the simulation
total_stats, data_stats = init_dataStats(data_stats, total_stats, agents)
def init():
global time, agents, gates, n_col_per_ag, envir, states, orders_list, data_stats, wall_x, wall_y, gate_x, gate_y, total_stats, office_x, office_y
# Initilizing the environement
envir, wall_x, wall_y, gate_x, gate_y, office_x, office_y = envir_configuration(width, height)
# Initilizing the gate objects
gates.append(Gate(0, (0,31), (0,32), (0,33), (3,29)))
gates.append(Gate(1, (0,37), (0,38), (0,39), (3, 38)))
gates.append(Gate(2, (0,43), (0,44), (0,45), (3, 47)))
# Initializing the agents
if(behavior_type == 1):
for i in range(0, N_AGV):
agents.append(AGV((42, i + 1 + (i*2) ),"red", 100 + i))
elif(behavior_type == 2):
for i in range(0, N_AGV):
agents.append(AGV((42, i + 1 + (i*2) ),"red", ["client", "status"]))
columns = orders_list.columns[2:]
n_col = len(columns)
i = 0
if(n_col > N_AGV):
for c in columns:
if(i < N_AGV):
temp = agents[i].id
temp.append(c)
agents[i].id = temp
i = i + 1
else:
i = 0
else:
print("Error - Too many AGVs ")
elif(behavior_type == 3):
for i in range(0, N_AGV):
agents.append(AGV((42, i + 1 + (i*2) ),"red", 100 + i))
else:
print("Error - Behavior_Type is not existing.")
# Initializing the dataframe that is going to be used in order to collect the data about the simulation
total_stats, data_stats = init_dataStats(data_stats, total_stats, agents)
def create( spec, scheme = 'none', count = 20, *args ):
"""
@spec - Specification (size, endpoints, barriers); either exactly
specified in a file, or with numeric values in a list
@option_scheme - none|manual|optimal|small-world|random|ozgur's betweenness|ozgur's randomness|end
@n_actions - Number of steps that need to taken
comment : optimal(shortest path to destination)??|random|ozgur's betweenness|ozgur's randomness
"""
env = AGV.create( spec )
# Percentage
if isinstance(count,str):
count = int(count[:-1])
count = count*env.S/100
# Add options for all the optimal states
O = []
if scheme == "none":
pass
elif scheme == "random-node":
O = OptionGenerator.optimal_options_from_random_nodes( env, count, *args )
elif scheme == "random-path":
O = OptionGenerator.optimal_options_from_random_paths( env, count, *args )
elif scheme == "betweenness":
O = OptionGenerator.optimal_options_from_betweenness( env, count, *args )
elif scheme == "small-world":
O = OptionGenerator.optimal_options_from_small_world( env, count, *args )
elif scheme == "betweenness+small-world":
O = OptionEnvironment.optimal_options_from_betweenness( env, count )
count_ = count - len( O )
O += OptionEnvironment.optimal_options_from_small_world( env, count_, *args )
elif scheme == "load":
O = OptionGenerator.options_from_file( count, *args )
else:
raise NotImplemented()
return OptionEnvironment( AGVOptions, env.S, env.A, env.P, env.R, env.R_bias, env.start_set, env.end_set, O )
def reset_rewards( env, spec, *args ):
O = env.O
env = AGV.reset_rewards( env, spec )
return OptionEnvironment( AGVOptions, env.S, env.A, env.P, env.R, env.R_bias, env.start_set, env.end_set, O )
# print(2*x)
if __name__ == "__main__":
rootDic = sys.argv[1]
totalAGVno = 4
regNo = 4
AGVObj=[]
q = Queue()
ontObjList =[]
mapNo = 1
mObj = createMaps(mapNo, rootDic)
c = 1
while ( c <= totalAGVno):
newOntCreate(c,rootDic)
AGVObj.append(AGV(rootDic,mapNo, c, regNo, mObj))
ontObjList.append(newTT(c,rootDic))
c = c+1
#end while
q.put(ontObjList)
p1 = Process(target = AGVObj[0].pathPlanning,args=(ontObjList,))
p2 = Process(target = AGVObj[1].pathPlanning,args=(ontObjList,))
p3 = Process(target = AGVObj[2].pathPlanning,args=(ontObjList,))
p4 = Process(target = AGVObj[3].pathPlanning,args=(ontObjList,))
p1.start()
p2.start()
p3.start()
p4.start()
#end main