45c9918 · opencog-rocca

Changed files (2) +81 -48

examples/flatworld.py +16 -17

…	…	@@ -11,7 +11,7 @@ from rocca.agents.utils import agent_log
11	11	from rocca.envs.malmo_demo.flatworld_env import *
12	12	from rocca.envs.wrappers import MalmoWrapper
13	13	from opencog.ure import ure_logger
14		from rocca.envs.wrappers.utils import mk_action
	14	from rocca.envs.wrappers.utils import mk_evaluation
15	15	from functools import wraps
16	16
17	17
…	…	@@ -24,28 +24,27 @@ class AbstractMalmoWrapper(MalmoWrapper):
24	24	@wraps(step)
25	25	def wrapper(ref, action, update_callback=None):
26	26	name = action.out[0]
27		ws = step(ref, name, update_callback)
28
29		return MalmoWrapper.parse_world_state(ws)
	27	return step(ref, name, update_callback)
30	28
31	29	return wrapper
32	30
33	31	@step_decorator.__func__
34		def step(self, action, update_callback=None):
	32	def step(self, abstract_action, update_callback=None):
35	33	try:
36		func = globals()[action.name]
37		func(self.agent_host)
	34	func = globals()[abstract_action.name]
	35	observations, reward, done = func(self.agent_host)
38	36	time.sleep(0.2)
39		self.world_state = self.agent_host.getWorldState()
40
41		for error in self.world_state.errors:
42		print("Error: ", error.text)
43
44		if update_callback is not None:
45		update_callback(action, self.world_state)
	37	obs_list = []
	38	if observations:
	39	for k in observations:
	40	if isinstance(observations[k], list):
	41	for i in observations[k]:
	42	obs_list.append(mk_evaluation(i, k))
	43	else:
	44	obs_list.append(mk_evaluation(k, observations[k]))
46	45	except RuntimeError as e:
47	46	print("Error sending command:", e)
48		return self.world_state
	47	return obs_list, mk_evaluation("Reward", reward), done
49	48
50	49	if __name__ == "__main__":
51	50	atomspace = AtomSpace()
…	…	@@ -56,7 +55,7 @@ if __name__ == "__main__":
56	55	# log.set_sync(True)
57	56	agent_log.set_level("debug")
58	57	# agent_log.set_sync(True)
59		ure_logger().set_level("debug")
	58	ure_logger().set_level("info")
60	59
61	60	#
62	61	wrapped_env = AbstractMalmoWrapper(missionXML=missionXML, validate=True)
…	…	@@ -78,7 +77,7 @@ if __name__ == "__main__":
78	77
79	78	# Training/learning loop
80	79	lt_iterations = 3 # Number of learning-training iterations
81		lt_period = 10 # Duration of a learning-training iteration
	80	lt_period = 50 # Duration of a learning-training iteration
82	81	for i in range(lt_iterations):
83	82	cog_agent.reset_action_counter()
84	83	par = cog_agent.accumulated_reward # Keep track of the reward before

rocca/envs/malmo_demo/flatworld_env.py +65 -31

…	…	@@ -76,14 +76,17 @@ key_x, key_y, key_z = 7, 228, -10
76	76	door_x, door_y, door_z = -4, 228, -3
77	77	diamond_x, diamond_y, diamond_z = -18, 228, -3
78	78	start_x, start_y, start_z, width, length, height = -4, 226, -1,6,15,5
	79	holds_key = False
79	80
80	81	# relocates the agent to the starting point
81	82	def relocate_agent(agent):
	83	print("Agent relocated to original place")
82	84	agent.sendCommand("tp {} {} {}".format(agent_x, agent_y, agent_z))
83	85	agent.sendCommand("setYaw {}".format(agent_yaw))
	86	time.sleep(0.2)
84	87
85	88	# stop the agent
86		def stop_condition(agent, blocktype, sec=5):
	89	def stop_condition(agent, blocktype, sec=10):
87	90	print("Moving to the {}".format(blocktype))
88	91	def stop(agent, blocktype):
89	92	done = False
…	…	@@ -96,27 +99,29 @@ def stop_condition(agent, blocktype, sec=5):
96	99	print("reached at {}".format(blocktype))
97	100	agent.sendCommand("move 0")
98	101	done = True
99		try:
100		with timeout(sec):
	102
	103	with timeout(sec):
	104	try:
101	105	stop(agent, blocktype)
102	106	return True
103		except TimeoutError:
104		print("Time out")
105		agent.sendCommand("move 0")
106		relocate_agent(agent)
107		return False
	107	except TimeoutError:
	108	print("********* Time out ************")
	109	agent.sendCommand("move 0")
	110	relocate_agent(agent)
	111	return False
108	112
109	113	# Get realtime location coordinates
110	114	def get_curr_loc(agent):
	115	loc = (agent_x, agent_y, agent_z)
111	116	world_state = agent.peekWorldState()
112		while world_state.is_mission_running:
113		world_state = agent.peekWorldState()
114		if world_state.number_of_observations_since_last_state > 0:
115		ob = json.loads(world_state.observations[-1].text)
116		return (int(ob[u'XPos']), int(ob[u'YPos']), int(ob[u'ZPos']))
117		else:
118		return (agent_x, agent_y, agent_z)
	117	# while world_state.is_mission_running:
	118	# world_state = agent.peekWorldState()
	119	if world_state.number_of_observations_since_last_state > 0:
	120	ob = json.loads(world_state.observations[-1].text)
	121	loc = (int(ob[u'XPos']), int(ob[u'YPos']), int(ob[u'ZPos']))
119	122
	123	return loc
	124
120	125	# Turns the agent towards the location given
121	126	def turn_to(agent, x1, y1, z1, x2, y2, z2):
122	127	def get_angle(x1, y1, z1, x2, y2, z2):
…	…	@@ -133,18 +138,27 @@ def turn_to(agent, x1, y1, z1, x2, y2, z2):
133	138	angle = yaw + angle if z1 < z2 else yaw - angle
134	139	agent.sendCommand("setYaw {}".format(angle))
135	140
	141	def is_mission_running(agent):
	142	world_state = agent.peekWorldState()
	143	return world_state.is_mission_running
	144
136	145	# Move the agent towards the key
137	146	# reward=0
138	147	def go_to_the_key(agent):
	148	global holds_key
139	149	curr_x, curr_y, curr_z = get_curr_loc(agent)
140	150	turn_to(agent,curr_x, curr_y, curr_z, key_x, key_y, key_z)
141	151	time.sleep(0.2)
142	152	agent.sendCommand("move 1")
143		if stop_condition(agent, 'tripwire_hook'):
144		# agent.sendCommand("tp {} {} {}".format(key_x, key_y, key_z))
	153	action_complete = stop_condition(agent, 'tripwire_hook')
	154	if action_complete:
145	155	agent.sendCommand("hotbar.2 1")
	156	observation = {"agent_status":["agent_holds_the_key"]}
	157	holds_key = True
146	158	else:
147		relocate_agent(agent)
	159	observation = []
	160	reward = 0
	161	return observation, reward, is_mission_running
148	162
149	163	# Move the agent towards the house and enter the house.
150	164	# reward = 0
…	…	@@ -153,10 +167,20 @@ def go_to_the_house(agent):
153	167	turn_to(agent, curr_x, curr_y, curr_z, door_x, door_y, door_z)
154	168	time.sleep(0.2)
155	169	agent.sendCommand("move 1")
156		if stop_condition(agent, 'dark_oak_door'):
157		agent.sendCommand("tp {} {} {}".format(door_x, door_y, door_z))
	170	action_complete = stop_condition(agent, 'dark_oak_door')
	171	if action_complete:
	172	observation = {"agent_status":[]}
	173	if holds_key:
	174	agent.sendCommand("tp {} {} {}".format(door_x, door_y, door_z))
	175	observation["agent_status"].append("agent_inside_the_house")
	176	observation["agent_status"].append("agent_holds_the_key")
	177	else:
	178	observation["agent_status"].append("agent_nextto_closed_door")
	179
158	180	else:
159		relocate_agent(agent)
	181	observation = []
	182	reward = 0
	183	return observation, reward, is_mission_running
160	184
161	185	# Move the agent towards the diamond and collect it.
162	186	# reward = 1
…	…	@@ -165,10 +189,12 @@ def go_to_the_diamonds(agent):
165	189	turn_to(agent,curr_x, curr_y, curr_z, diamond_x, diamond_y, diamond_z)
166	190	time.sleep(0.2)
167	191	agent.sendCommand("move 1")
168		if stop_condition(agent, 'diamond_block'):
169		agent.sendCommand("tp {} {} {}".format(diamond_x, diamond_y, diamond_z))
170		time.sleep(0.2)
171		relocate_agent(agent)
	192	action_complete = stop_condition(agent, 'diamond_block')
	193	if action_complete:
	194	reward = 1
	195	else:
	196	reward = 0
	197	return [], reward, is_mission_running
172	198
173	199	missionXML = '''
174	200	<?xml version="1.0" encoding="UTF-8" standalone="no" ?>
…	…	@@ -249,17 +275,25 @@ if __name__ == "__main__":
249	275	rw, ob, done = malmoWrapper.restart()
250	276	total_reward = 0
251	277	agent = malmoWrapper.agent_host
252		go_to_the_key(agent)
253
	278	_, rw, _= go_to_the_key(agent)
	279	print(ob)
	280	total_reward += rw
254	281	world_state = agent.getWorldState()
255	282	while world_state.is_mission_running:
256	283	world_state = agent.getWorldState()
257	284	if world_state.number_of_rewards_since_last_state > 0:
258	285	total_reward += world_state.rewards[-1].getValue()
259		print("total reward = {}".format(total_reward))
260		go_to_the_key(agent)
261		go_to_the_house(agent)
262		go_to_the_diamonds(agent)
	286	print("total reward minecraft = {}".format(total_reward))
	287
	288	ob, rw, _ = go_to_the_house(agent)
	289	print(ob)
	290	total_reward += rw
	291
	292	ob, rw, _ = go_to_the_diamonds(agent)
	293	print(ob)
	294	total_reward += rw
	295
263	296	time.sleep(2)
	297	print("total reward = {}".format(total_reward))
264	298	print()
265	299	print("Mission ended")

opencog-rocca commits / 45c9918

Changed files (2) +81 -48