async def announce_movement(self, curr_ind, next_ind, curr, nxt, arriving):
curr_channel = self.bot.working_guild.get_channel(curr.cat_id)
nxt_channel = self.bot.working_guild.get_channel(nxt.cat_id)
s_map = util.draw_map(
curr_ind,
next_ind,
self.direction,
len(self.bot.metro_state.metro.stations),
not arriving,
self.bot.metro_state.metro.looped,
)
msg = f"Arriving at: **{curr_channel.name}**\nNext stop: **{nxt_channel.name}**"
if not arriving:
msg = (
f"Departing: **{curr_channel.name}**\nNext stop: **{nxt_channel.name}**"
)
msg = f"{s_map}\n\n{msg}"
msg = discord.utils.escape_mentions(msg)
await self.announce(msg)
async def announce_movement(self, curr_ind, next_ind, curr, nxt, arriving):
curr_channel = self.bot.working_guild.get_channel(curr.cat_id)
nxt_channel = self.bot.working_guild.get_channel(nxt.cat_id)
s_map = util.draw_map(
curr_ind,
next_ind,
self.direction,
len(self.bot.metro_state.metro.stations),
not arriving,
self.bot.metro_state.metro.looped,
)
curr_name = clean_emoji(curr_channel.name)
nxt_name = clean_emoji(nxt_channel.name)
msg = f"Arriving at: **{curr_name}**\nNext stop: **{nxt_name}**"
if not arriving:
msg = f"Departing: **{curr_name}**\nNext stop: **{nxt_name}**"
msg = f"{s_map}\n\n{msg}\n\nReact below to disembark at the {'current' if arriving else 'next'} stop."
msg = discord.utils.escape_mentions(msg)
await self.announce(msg)
print_help_command()
print('cmd>', end=' ')
instruction = raw_input()
argv = instruction.strip(' ').split(' ')
if len(argv) == 1:
if argv[0] == 'q':
sys.exit(0)
elif argv[0] == 'a':
plot_all(graph)
elif argv[0] == 'a_id':
plot_all(graph, 'l')
elif argv[0] == 'a_rid':
plot_all(graph, 'r')
else:
print('[ERROR] wrong command')
continue
if len(argv) == 2:
if argv[0] == 'i':
plot_id(graph, argv[1])
elif argv[0] == 'i_map':
plot_id(graph, argv[1])
util.draw_map(plt.gca(), base_map)
else:
print('[ERROR] wrong command')
continue
else:
print('[ERROR] wrong arguments')
continue
def test(self,
with_animation=False,
with_map=False,
with_curve=True,
with_output=False):
# 输入搜索区域矩阵数据
# df = pd.read_csv('/home/qin/code/experiments/cooperative_search/input2.csv', header=None)
# matrix_sr = df.values
# 随机初始化目标位置
matrix_sr = np.zeros((self.row, self.column))
target_position_index = np.random.choice(self.row * self.column,
self.num_target,
replace=False)
target_position_x = target_position_index // self.column
target_position_y = target_position_index % self.column
for x, y in zip(target_position_x, target_position_y):
matrix_sr[x][y] = 1
# 参数定义
map_init = np.zeros([self.row, self.column])
p = 0.9
q = 0.3
radius_search = 4
radius_communicate = 25
radius_consider = 6
bound_speed = 2
bound_Q = 10000000000000000
# 随机初始化agent位置
center_x = np.random.randint(radius_communicate // 2,
self.row - radius_communicate // 2)
center_y = np.random.randint(radius_communicate // 2,
self.column - radius_communicate // 2)
center = (center_x, center_y)
coords_in = []
for x in range(center[0] - radius_communicate // 2,
center[0] + radius_communicate // 2 + 1):
for y in range(center[1] - radius_communicate // 2,
center[1] + radius_communicate // 2 + 1):
if (pow(x - center[0], 2) + pow(y - center[1], 2)) <= pow(
radius_communicate // 2, 2):
coords_in.append((x, y))
id_coords = list(
np.random.choice(len(coords_in), size=6, replace=False))
coords_in = np.array(coords_in)
coords = coords_in[id_coords]
coords = list(coords.astype(float))
print(center, coords)
# agent初始化
agent0 = Agent(0, coords[0], map_init, p, q, radius_search,
radius_communicate, radius_consider, bound_speed,
bound_Q)
agent1 = Agent(1, coords[1], map_init, p, q, radius_search,
radius_communicate, radius_consider, bound_speed,
bound_Q)
agent2 = Agent(2, coords[2], map_init, p, q, radius_search,
radius_communicate, radius_consider, bound_speed,
bound_Q)
agent3 = Agent(3, coords[3], map_init, p, q, radius_search,
radius_communicate, radius_consider, bound_speed,
bound_Q)
agent4 = Agent(4, coords[4], map_init, p, q, radius_search,
radius_communicate, radius_consider, bound_speed,
bound_Q)
agent5 = Agent(5, coords[5], map_init, p, q, radius_search,
radius_communicate, radius_consider, bound_speed,
bound_Q)
# 初始化一些后面用到的列表
list_agent = [agent0, agent1, agent2, agent3, agent4,
agent5] # agent列表
N = len(list_agent) # agent数量
count = 0 # 循环次数
unsures = []
# 热图初始化
if with_animation:
fig = plt.figure(figsize=(8, 8))
ax = fig.add_subplot(111)
map_im = float(1) / (np.exp(map_init) + 1)
im = ax.imshow(map_im, cmap='bone_r', vmin=0, vmax=1)
plt.colorbar(im, shrink=0.5)
coord_pls = [[[], []], [[], []], [[], []], [[], []], [[], []],
[[], []]]
colors = ['red', 'yellow', 'blue', 'c', 'green', 'm']
for i in range(N):
coord_pls[i][0].append(coords[i][0])
coord_pls[i][1].append(coords[i][1])
for i in range(N):
ax.plot(coord_pls[i][1],
coord_pls[i][0],
color=colors[i],
marker='>',
linewidth=0.3)
plt.pause(1)
# 主循环
while True:
# 循环退出条件
unsure = 0
for agent in list_agent:
for i in range(self.row):
for j in range(self.column):
unsure += np.exp(-2 * np.linalg.norm(agent.map[i][j]))
unsure_final = unsure / (N * self.column * self.row)
unsures.append(unsure_final)
if unsure_final <= 0.001:
if with_map:
for agent in list_agent:
maps = []
maps.append(float(1) / (np.exp(agent.map) + 1))
#print(maps)
draw_map(maps[0])
if with_curve:
draw_curve(unsures)
if with_output:
output = pd.DataFrame(maps[0])
output.to_csv('output.csv', index=False)
break
# 单个agent概率图更新
for agent in list_agent:
cell_search = agent.cellsearch()
agent.update(matrix_sr, cell_search)
# 概率图融合
coord_all = [] # 所有agent坐标的列表
list_map = [] # 所有agent概率图的列表
for agent in list_agent:
coord_all.append(agent.coord)
list_map.append(agent.map)
array_map = np.array(list_map) # 转array
for agent in list_agent:
agent_neighbor = agent.neighbor(coord_all)
agent.fuse(agent_neighbor, N, array_map)
# agent移动
for agent in list_agent:
cell_consider = agent.cellconsider()
agent.move(cell_consider)
count += 1
print(count)
if with_animation:
# 绘制热图
map_im0 = np.zeros([self.row, self.column])
for agent in list_agent:
map_im0 += float(1) / (np.exp(agent.map) + 1)
map_im = map_im0 / N
im.set_array(map_im)
# 绘制折线图
for i, agent in zip(range(N), list_agent):
coord_pls[i][0].append(agent.coord[0])
coord_pls[i][1].append(agent.coord[1])
if count >= 1:
coord_pls[i][0].pop(0)
coord_pls[i][1].pop(0)
ax.plot(coord_pls[i][1],
coord_pls[i][0],
color=colors[i],
marker='>',
linewidth=0.3)
ax.lines.pop(0)
plt.pause(0.3)
map_dir = util.get_map_dir(sys.argv)
graph = util.get_topodata(map_dir)
base_map = util.get_mapdata(map_dir)
route = util.get_routingdata()
central_curves = {}
for nd in graph.node:
central_curves[nd.lane_id] = nd.central_curve
plt.ion()
while 1:
print_help_command()
print 'cmd>',
instruction = raw_input()
argv = instruction.strip(' ').split(' ')
if len(argv) == 1:
if argv[0] == 'q':
sys.exit(0)
elif argv[0] == 'r':
plot_route(route, central_curves)
elif argv[0] == 'r_map':
plot_route(route, central_curves)
util.draw_map(plt.gca(), base_map)
else:
print '[ERROR] wrong command'
continue
else:
print '[ERROR] wrong arguments'
continue