def test_scan():
earley = Earley('a')
earley.rules_list = [Rule('U', 'S'), Rule('S', 'a')]
earley.situations_dict[0].add(Situation('S', 'a', 0, 0))
earley.scan(0, 'a')
is_added = False
to_add = Situation('S', 'a', 0, 1)
for sit in earley.situations_dict[1]:
if sit == to_add:
is_added = True
break
assert is_added
def generate_trace(db_path):
db = DataBase(db_path)
t = Trace()
s = Situation()
# Robot/Operator moving status
moving_topics = {
"/safety/robot/moving": s.robot.is_moving,
"/safety/operator/moving": s.operator.is_moving,
}
for c in moving_topics.values():
t[c] = (0.0, False)
for m in from_table(db, "movablestatus"):
if m.topic in moving_topics:
t[moving_topics[m.topic]] = (m.timestamp, m.is_moving == 1)
# Collision occurrences
t[s.collision.occurs] = (0.0, False)
t[s.collision.force] = (0.0, 0.0)
for m in from_table(db, "collisionevent"):
if m.entity_id == "Operator-Tim":
t[s.collision.occurs] = (m.timestamp, True)
t[s.collision.force] = (m.timestamp, m.collision_force)
t[s.collision.occurs] = (m.timestamp + 0.01, False)
t[s.collision.force] = (m.timestamp + 0.01, 0.0)
# LIDAR Measurements
t[s.lidar.distance] = (0.0, float("inf"))
for m in from_table(db, "float32"):
if m.topic == "/lidar/digital/range":
t[s.lidar.distance] = (m.timestamp, m.data)
return t
def complete(self, list_number):
situations_to_insert = []
for situation in self.situations_dict[list_number]:
list_number_2 = situation.index
if situation.point == len(situation.output):
for situation_2 in self.situations_dict[list_number_2]:
sit = Situation(situation_2.input, situation_2.output, situation_2.index, situation_2.point + 1)
situations_to_insert.append(sit)
for situation in situations_to_insert:
self.add_situation(situation, list_number)
def predict(self, list_number):
situations_to_insert = []
for situation in self.situations_dict[list_number]:
if situation.point < len(situation.output):
nonterminal = situation.output[situation.point]
for rule in self.rules_list:
if rule.input == nonterminal:
sit = Situation(nonterminal, rule.output, list_number, 0)
situations_to_insert.append(sit)
for situation in situations_to_insert:
self.add_situation(situation, list_number)
def __init__(self):
super().__init__(daemon=True)
self.name = 'EngineThread'
self.state = self.ENGINE_BOOT
self.sit = Situation()
self.stack = [self.sit]
self.index = 0
self.running = False
self.usemillisec = True
self.millisec = 1
def test_predict():
earley = Earley('a')
earley.rules_list = [Rule('U', 'S'), Rule('S', 'a')]
earley.predict(0)
is_added = False
to_add = Situation('S', 'a', 0, 0)
for sit in earley.situations_dict[0]:
if sit == to_add:
is_added = True
break
assert is_added
def find_situations(concerns):
# management of the first concern
intro_sit_file = open('data/fixed_texts/intro_situations.txt', "r")
sm.my_print_file(intro_sit_file, FLAG)
intro_sit_file.close()
# replacement of * in the questions with the concern LEO is facing at the moment
uncompleted_question = sm.choose_sentence("situations")
question = sm.replace_a_star(uncompleted_question,
concerns[0].get_concern())
sm.my_print_string(question, FLAG)
answer = input()
new_answer, keywords_list = analyze_answer(answer)
for keyword in keywords_list:
situation = sm.complete_keywords(new_answer, keyword[0])
concerns[0].add_situation(Situation(situation))
return concerns, answer
def makeCorpus(world, wordFile, objectFile):
# Input: World object, filenames of word and object files
# Output: Corpus as list of Situations
corpus = []
wf = open(wordFile, 'r')
of = open(objectFile, 'r')
wl = wf.readline()
ol = of.readline()
while (len(wl) != 0 and len(ol) != 0):
wl = processLine(wl)
ol = processLine(ol)
words = [world.words_key.index(x) for x in wl.split(' ') if x != '']
objects = [
world.objects_key.index(x) for x in ol.split(' ') if x != ''
]
corpus.append(Situation(words, objects))
wl = wf.readline()
ol = of.readline()
return corpus
def __init__(self, word):
situation = Situation('U', 'S', 0, 0)
self.situations_dict = {i: set() for i in range(len(word) + 1)}
self.situations_dict[0].add(situation)
self.word = word
def scan(self, list_number, symbol):
for situation in self.situations_dict[list_number]:
if situation.output[situation.point] == symbol:
sit = Situation(situation.input, situation.output, situation.index, situation.point + 1)
self.add_situation(sit, list_number + 1)
from situation import Situation
_S = Situation()
_COLLISION_FORCE_THRESHOLD = 100
# Any contact between the robot and operator
contact_occurs = _S.collision.occurs.eventually()
# Any hazardous contact between the robot and operator
collision_occurs = (_S.collision.occurs & _S.collision.force >
_COLLISION_FORCE_THRESHOLD).eventually()
# Any contact due to the operator moving into a stopped robot
operator_collides = (_S.collision.occurs & ~_S.robot.is_moving).eventually()
# The robot stops within 250ms when an obstacle gets too close
_SAFETY_STOP_THRESHOLD = 0.75
robot_safety_stops = ((_S.lidar.distance < _SAFETY_STOP_THRESHOLD).implies(
(~_S.robot.is_moving)
| (~_S.robot.is_moving).eventually(hi=0.250))).always()
monitored_conditions = {
"Contact between operator and robot": contact_occurs,
"Collision between operator and robot": collision_occurs,
"Operator collides with robot": operator_collides,
"Robot performs safety stop": robot_safety_stops,
}
def __init__(self) -> None:
self.callback = None
self.sit = Situation()
def main():
# コマンドライン引数を読む
args = read_args()
# confファイルの読み込み
conf = configparser.ConfigParser()
conf.read('./setting.ini', 'UTF-8')
# log設定
logging.basicConfig(level=logging.DEBUG,
format="%(relativeCreated)08d[ms] - %(name)s - %(levelname)s - %(processName)-10s - %(threadName)s -\n*** %(message)s",
filename="log/app.log",
)
# 歩きスマホ人数
arukisumaho_num = 0
# 人数
person_num = 0
if args.img:
img = abspath(args.img)
else:
# 画像ファイル名を設定
basename = datetime.now().strftime("%Y%m%d-%H%M%S")
img = "./media/"+basename+".jpg"
# 画像を撮影
wc = Webcam(conf.get("webcam", 'ip') ,
conf.get("webcam", 'port') ,
conf.get("webcam", 'user') ,
conf.get("webcam", 'password')
)
wc.setup_basic_auth()
wc.take_picture(img)
# 写真が存在しない場合終了する
if not exists(img):
print('画像がないよ')
return
# インスタンス作成
situation = Situation()
pm = ProjectionMapping(
conf.get("vpt", 'ip') ,
conf.getint("vpt", 'port')
)
# 最初に車の動画を流す
pm.waiting()
while True:
if not args.img:
wc.take_picture(img)
situation.recognize(img)
person_num = situation.get_person_num()
arukisumaho_num = situation.get_arukisumaho_num()
print('検出された人数： %d' % person_num)
print('歩きスマホ人数： %d' % arukisumaho_num)
try :
# 歩きスマホいる場合
if arukisumaho_num >= 1:
pm.arukisumaho()
sleep(10) # TODO : 動画時間入れる
pm.waiting()
else:
# 歩行者が大勢いる場合
if person_num > 1:
pm.wide_cross()
sleep(10) # TODO : 動画時間入れる
pm.waiting()
# 歩行者ひとりいる場合
elif person_num == 1:
pm.normal_cross()
sleep(10) # TODO : 動画時間入れる
pm.waiting()
# 歩行者がいない場合
else:
sleep(3)
pass
except KeyboardInterrupt:
pm.waiting()