def site_interpolation(edges, selected_points):
global travel_range
edges_fine = []
selected_points_fine = copy.copy(selected_points)
for edge in edges:
start = selected_points[edge[0]]
end = selected_points[edge[1]]
st.reset()
length, path = pa.find(start, end, 0, [])
if min(length, st.global_min) <= travel_range:
edges_fine.append(edge)
while min(length, st.global_min) > travel_range:
print(start, '\n', end)
reach = get_P(point_coverage(start, mode='travel'))
print('covered')
line = ge.LineString([start, end])
new_site = reach.exterior.intersection(line)
selected_points_fine.append(new_site)
edges_fine.append([edge[0], len(selected_points_fine) - 1])
edges_fine.append([len(selected_points_fine) - 1, edge[1]])
start = new_site
st.reset()
length, path = pa.find(start, end, 0, [])
print('interpolation')
print('interpolation done')
return edges_fine, selected_points_fine
def point_coverage(point, mode='cover'):
global cover_range
global travel_range
global obs
# global rdif
points = [None for _ in range(65)]
if mode == 'cover':
nowr = cover_range
else:
nowr = travel_range
rdif = (nowr * 3) / 100
while not all(points):
# print(nowr)
if nowr <= 0:
break
circle = get_circle(point, nowr)
# print(circle)
for i, circle_point in enumerate(circle):
if not points[i]:
# print('input',point,circle_point)
# print(nowr)
st.reset()
# print(point,'\n',circle_point)
dis, path = pa.find(point, circle_point, 0, [])
# print(dis,st.global_min)
# if i==65:
# stack.append([point,circle_point])
if min(dis, st.global_min) <= cover_range:
points[i] = circle_point
nowr -= rdif
return points
def ob_dis():
stack = [0 for _ in range(len(st.obs))]
for i, ob in enumerate(st.obs):
# print(i)
st.reset()
lenght, path = pa.find(st.pstart, ob.centroid, 0, [])
stack[i] = min(lenght, st.global_min)
with open("D:/UAV project/obs_distance.csv", "w+") as my_csv:
csvWriter = csv.writer(my_csv, delimiter=',')
csvWriter.writerow(stack)
print('obs_distance to start generated')
return
def picked_sites_dis(selected_points_this):
# global ref
# ref=selected_points_this
dis_mat = [[-1 for _ in range(len(selected_points_this))]
for _ in range(len(selected_points_this))]
i = 0
while i < len(selected_points_this) - 1:
# print('i',i,len(selected_points_this))
for j in range(i + 1, len(selected_points_this)):
st.reset()
# print(i,j)
# print(type(i),type(j))
# print(selected_points_this[i],'\n',selected_points_this[j])
# print(type(selected_points_this[i]),type(selected_points_this[j]))
lenght, path = pa.find(selected_points_this[i],
selected_points_this[j], 0, [])
# print('min',st.global_min,'lenght',lenght,'\n')
dis_mat[i][j] = min(st.global_min, lenght)
# print('pre',i,len(selected_points_this))
i += 1
# csv_write(dis_mat,"picked_sites_dis")
print('sites dis calculated')
return dis_mat
def branch_and_cut(
dis_mat,
task_number,
solver='exact',
plot=1): #generate task permutation find pareto optimality solution
task = task_generator(len(dis_mat), task_number)
def cal_fitness(dis, en, a=0.1):
return a * en + (1 - a) * dis
def exact_solver(dis_mat, task):
min_dis_all = float('inf')
min_en_all = float('inf')
max_fitness = float('inf')
solution = [[], [], []]
for this_task in itertools.permutations(task):
this_task = [0] + list(this_task) + [0]
dis_all = 0
en_all = 0
for i in range(0, len(this_task) - 1):
mi, ma = min(this_task[i],
this_task[i + 1]), max(this_task[i],
this_task[i + 1])
dis_all += dis_mat[mi][ma]
en_all += cal_travel_cost(
dis_mat[mi][ma],
10 * ((len(this_task) - 2 - i) / (len(this_task) - 2)))
fitness = cal_fitness(dis=dis_all, en=en_all)
if dis_all < min_dis_all:
min_dis_all = dis_all
solution[0] = [this_task, dis_all * 100, en_all, fitness]
if en_all < min_en_all:
min_en_all = en_all
solution[1] = [this_task, dis_all * 100, en_all, fitness]
if fitness < max_fitness:
max_fitness = fitness
solution[2] = [this_task, dis_all * 100, en_all, fitness]
return solution
if solver == 'exact':
solution = exact_solver(dis_mat, task)
print(solution)
with open("D:/UAV project/solution.txt", "w+") as f:
f.write('total_dis(km),total_energy_consumption(battery),fitness\n')
f.write('min time solution\n')
f.write(str(solution[0]))
f.write('min energy consumption_solution\n')
f.write(str(solution[1]))
f.write('max fitness solution\n')
f.write(str(solution[2]))
print('task planning done')
if plot:
edges = []
task_route = solution[plot - 1][0]
for i in range(len(task_route) - 1):
st.reset()
lenght, path = pa.find(selected_points_raw[task_route[i]],
selected_points_raw[task_route[i + 1]], 0,
[])
if st.global_path:
edges.extend(st.global_path)
else:
edges.extend(path)
p([st.obs,
pan.GeoSeries(selected_points_raw),
pan.GeoSeries(edges)], ['red', 'black'])
print('route plotted')
return