Skip to content

Commit

Permalink
Finished_simulation_result_updating
Browse files Browse the repository at this point in the history
  • Loading branch information
@YaolinGe
YaolinGe committed Sep 24, 2023
1 parent a4eca36 commit ffeb414
Show file tree
Hide file tree
Showing 2 changed files with 219 additions and 2 deletions.
219 changes: 218 additions & 1 deletion Publication/src/Simulators/EDA_SIM.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,7 @@
Email: [email protected]
Date: 2023-09-05
"""
import datetime

from WGS import WGS
from Config import Config
Expand All @@ -18,10 +19,13 @@
from joblib import Parallel, delayed, dump, load
from concurrent.futures import ThreadPoolExecutor
import seaborn as sns
from matplotlib.patches import PathPatch
from matplotlib.path import Path
import pandas as pd
from scipy.stats import gaussian_kde
from matplotlib.cm import get_cmap
from sklearn.neighbors import KernelDensity
from statsmodels.nonparametric.smoothers_lowess import lowess
from time import time
from matplotlib import tri
from shapely.geometry import Polygon, Point
Expand Down Expand Up @@ -90,7 +94,9 @@ def __init__(self) -> None:
self.load_data()
# self.plot_metrics_total()
# self.plot_temporal_traffic_density_map()
self.plot_ground_truth()
# self.plot_temporal_traffic_flow_density_map4paper()
self.plot_metrics4paper()
# self.plot_ground_truth()
# self.plot_es()
self.trajectory

Expand Down Expand Up @@ -255,6 +261,57 @@ def make_subplot_overset_underset(planner, step, is_over, ax):
# plt.savefig(fpath + "P_{:03d}.png".format(k))
# plt.close("all")

def plot_temporal_traffic_flow_density_map4paper(self) -> None:
"""
Plot the traffic flow density map for each specific case. For the paper. No need to save all the figures.
"""
num_steps = [29, 59, 89, 119]
time_start = datetime.datetime(2022, 5, 11, 9, 52)
fpath = figpath + "TrafficFlow/"
checkfolder(fpath)

def make_subplot(planner, item):
fig = plt.figure(figsize=(48, 10))
gs = GridSpec(nrows=1, ncols=4, figure=fig)
for i in range(len(num_steps)):
num_step = num_steps[i]
traj = self.trajectory[planner][item][:, :num_step, :]
lat, lon = WGS.xy2latlon(traj[:, :, 0], traj[:, :, 1])
df = pd.DataFrame(np.stack((lat.flatten(), lon.flatten()), axis=1), columns=['lat', 'lon'])
ax = fig.add_subplot(gs[i])
sns.kdeplot(df, x='lon', y='lat', fill=True, cmap="Reds", levels=25, thresh=.1)
plt.plot(self.polygon_border_wgs[:, 1], self.polygon_border_wgs[:, 0], 'k-.')
plg = plt.Polygon(np.fliplr(self.polygon_obstacle_wgs), facecolor='w', edgecolor='k', fill=True,
linestyle='-.')
plt.gca().add_patch(plg)

# Plot border and create a mask
border_path = Path(self.polygon_border_wgs[:, ::-1])
border_patch = PathPatch(border_path, facecolor='none', edgecolor='none')
ax.add_patch(border_patch)

# Mask KDE using border path
for collection in ax.collections:
collection.set_clip_path(border_patch)

if i == 0:
plt.ylabel("Latitude")
else:
plt.ylabel("")
plt.xlabel("Longitude")
date_string = time_start + datetime.timedelta(hours=(num_step + 1)/30 * 2)
plt.title(f"Density map at " + date_string.strftime("%H:%M"))
plt.xticks(self.lon_ticks)
plt.xlim([self.lon_min, self.lon_max])
plt.ylim([self.lat_min, self.lat_max])

plt.savefig(fpath + "TF_{:s}_{:s}.png".format(planner, item))
plt.close("all")

for planner in self.planners:
for item in self.cv:
make_subplot(planner, item)

def plot_temporal_traffic_density_map(self, step: 'int', row_ind, col_ind, fig, gs) -> None:
"""
Plot the traffic density map for each specific case.
Expand Down Expand Up @@ -296,6 +353,166 @@ def make_subplot(planner, item, step: int = 0):
ax = fig.add_subplot(gs[row_ind + 2, col_ind + 1])
make_subplot('rrt', 'equal', step=step)

def plot_metrics4paper(self) -> None:
"""
Plot the metrics for each specific case. For the paper. No need to save all the figures.
"""
fpath = figpath + "Metrics/"
checkfolder(fpath)
def make_subplot_metric(data, metric, title, ax):
# Define a mapping for the legend labels
legend_map = {
'eibv': 'EIBV dominant',
'ivr': 'IVR dominant',
'equal': 'Equal weighted'
}

# Preparing data for Seaborn
df_list = []
for key in ['eibv', 'ivr', 'equal']:
temp_df = pd.DataFrame({
'Time Step': np.tile(np.arange(self.num_steps), self.num_replicates),
metric: data[key].flatten(),
'Type': [legend_map[key]] * self.num_steps * self.num_replicates # Use the mapping here
})
df_list.append(temp_df)

df = pd.concat(df_list, axis=0)

# Using Seaborn's lineplot with uncertainty envelopes
sns.lineplot(data=df, x='Time Step', y=metric, hue='Type', ax=ax, ci="sd", err_style="band")

# ax.errorbar(np.arange(self.num_steps), y=np.mean(data['eibv'], axis=0),
# yerr=np.std(data['eibv'], axis=0) / np.sqrt(self.num_replicates) * 1.645, fmt="-o",
# capsize=5, label="EIBV dominant")
# ax.errorbar(np.arange(self.num_steps), y=np.mean(data['ivr'], axis=0),
# yerr=np.std(data['ivr'], axis=0) / np.sqrt(self.num_replicates) * 1.645, fmt="-o",
# capsize=5, label="IVR dominant")
# ax.errorbar(np.arange(self.num_steps), y=np.mean(data['equal'], axis=0),
# yerr=np.std(data['equal'], axis=0) / np.sqrt(self.num_replicates) * 1.645, fmt="-o",
# capsize=5, label="Equal weighted")
ax.set_xticks(self.xticks, self.xticklabels)
ax.set_xlim([0, self.num_steps])
ax.set_xlabel("Time Step")
ax.set_ylabel(metric)
ax.set_title(title)
plt.legend(loc="upper left")
if metric == "ibv":
ax.set_ylim([self.ibv_min, self.ibv_max])
elif metric == "rmse":
ax.set_ylim([self.rmse_min, self.rmse_max])
elif metric == "vr":
ax.set_ylim([self.vr_min, self.vr_max])
else:
pass
def make_subplot(metric):
if metric == "ibv":
data = self.ibv
elif metric == "rmse":
data = self.rmse
elif metric == "vr":
data = self.vr
else:
pass
fig = plt.figure(figsize=(24, 10))
gs = GridSpec(nrows=1, ncols=2, figure=fig)

ax1 = fig.add_subplot(gs[0])
make_subplot_metric(data['myopic'], metric.upper(), "Myopic", ax1)

ax2 = fig.add_subplot(gs[1])
make_subplot_metric(data['rrt'], metric.upper(), "RRT*", ax2)

# Determine the global y-limits
ylim_min = min(ax1.get_ylim()[0], ax2.get_ylim()[0])
ylim_max = max(ax1.get_ylim()[1], ax2.get_ylim()[1])

# Set y-limits for both subplots
ax1.set_ylim([ylim_min, ylim_max])
ax2.set_ylim([ylim_min, ylim_max])

plt.savefig(fpath + f"{metric}.png")
plt.close("all")



# def make_subplot_metric(data, metric, ax):
# # Define line styles for different methods
# linestyle_map = {
# 'myopic': '-',
# 'rrt': '--'
# }
#
# # Define colors for different keys
# color_map = {
# 'EIBV dominant': 'blue',
# 'IVR dominant': 'red',
# 'Equal weighted': 'green'
# }
#
# # Define a mapping for the legend labels
# legend_map = {
# 'eibv': 'EIBV dominant',
# 'ivr': 'IVR dominant',
# 'equal': 'Equal weighted'
# }
#
# # Preparing data for Seaborn
# # Preparing data for Seaborn
# df_list = []
# for method in ['myopic', 'rrt']:
# for key in ['eibv', 'ivr', 'equal']:
# temp_df = pd.DataFrame({
# 'Time Step': np.tile(np.arange(self.num_steps), self.num_replicates),
# metric: data[method][key].flatten(),
# 'Type': [legend_map[key]] * self.num_steps * self.num_replicates,
# 'LineStyle': [linestyle_map[method]] * self.num_steps * self.num_replicates
# # Additional column for linestyle
# })
# df_list.append(temp_df)
#
# df = pd.concat(df_list, axis=0)
#
# # Using Seaborn's lineplot with uncertainty envelopes
# sns.lineplot(data=df, x='Time Step', y=metric, hue='Type', style="LineStyle", palette=color_map, ax=ax,
# ci="sd", err_style="band")
#
# ax.set_xticks(self.xticks, self.xticklabels)
# ax.set_xlim([0, self.num_steps])
# ax.set_xlabel("Time Step")
# ax.set_ylabel(metric)
# ax.set_title(metric.upper())
# plt.legend(loc="upper left")
# if metric == "ibv":
# ax.set_ylim([self.ibv_min, self.ibv_max])
# elif metric == "rmse":
# ax.set_ylim([self.rmse_min, self.rmse_max])
# elif metric == "vr":
# ax.set_ylim([self.vr_min, self.vr_max])
# else:
# pass
#
# def make_subplot(metric):
# if metric == "ibv":
# data = self.ibv
# elif metric == "rmse":
# data = self.rmse
# elif metric == "vr":
# data = self.vr
# else:
# pass
# fig, ax = plt.subplots(figsize=(12, 8))
# make_subplot_metric(data, metric, ax)
# plt.savefig(fpath + f"{metric}.png")
# plt.close("all")
#
make_subplot("ibv")
make_subplot("rmse")
make_subplot("vr")

self.ibv_min


def plot_metrics(self, step, row_ind, col_ind, fig, gs) -> None:
"""
Plot the metrics for each specific case.
Expand Down
2 changes: 1 addition & 1 deletion Publication/src/Visualiser/AgentPlotRRTStar.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -246,7 +246,7 @@ def plot_waypoints():
lat, lon = WGS.xy2latlon(np.array([loc[0], loc_p[0]]),
np.array([loc[1], loc_p[1]]))
ax.plot([lon[0], lon[1]],
[lat[0], lat[1]], "-b", alpha=.3)
[lat[0], lat[1]], "-g", alpha=.5)
ax.plot(rrt_traj[:, 1], rrt_traj[:, 0], 'k-', linewidth=2)

plt.savefig(self.figpath + "P_{:03d}.png".format(self.cnt))
Expand Down

0 comments on commit ffeb414

Please sign in to comment.