Add Bayesian search examples #1100
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # To introduce Bayesian search in Stone Soup, we will start with a simple example. We will simulate | ||
| # a bearings-only sensor searching for a single stationary target (we will visualise this later). | ||
| # We assume that the sensor never locates the target, but record the probability that the target | ||
| # should have been found by a given timestep. We will control the sensor using using a few different |
There was a problem hiding this comment.
Suggested change
| # should have been found by a given timestep. We will control the sensor using using a few different | |
| # should have been found by a given timestep. We will control the sensor using a few different |
| # Bayesian Search | ||
| # --------------- | ||
| # The implementation of sensor management in many of Stone Soup's existing tutorials and examples | ||
| # relies on the assumption that we have perfect knowledge of a target's prior location. However in |
There was a problem hiding this comment.
Suggested change
| # relies on the assumption that we have perfect knowledge of a target's prior location. However in | |
| # relies on the assumption that we have perfect knowledge of a target's prior location. However, in |
| # if running sequential search, perform this now | ||
| if seq_flag: | ||
| sensor.timestamp = timestep | ||
| sensor.dwell_centre = sensor.dwell_centre + sensor.fov_angle/2. |
There was a problem hiding this comment.
Suggested change
| sensor.dwell_centre = sensor.dwell_centre + sensor.fov_angle/2. | |
| sensor.dwell_centre = sensor.dwell_centre + sensor.fov_angle/3. |
Does it make more sense to rotate the sensor by one bin per step?
Comment on lines
+91
to
+99
| from stonesoup.plotter import AnimatedPlotterly | ||
|
|
||
| import copy | ||
| from datetime import datetime, timedelta | ||
| import numpy as np | ||
| import pandas as pd | ||
| import plotly.express as px | ||
| import plotly.graph_objects as go | ||
| import time |
There was a problem hiding this comment.
Suggested change
| from stonesoup.plotter import AnimatedPlotterly | |
| import copy | |
| from datetime import datetime, timedelta | |
| import numpy as np | |
| import pandas as pd | |
| import plotly.express as px | |
| import plotly.graph_objects as go | |
| import time | |
| import copy | |
| from datetime import datetime, timedelta | |
| import numpy as np | |
| import plotly.graph_objects as go | |
| import time |
Unused imports
|
|
||
| from stonesoup.types.state import StateVector | ||
| from stonesoup.sensor.radar.radar import RadarRotatingBearing | ||
| from stonesoup.types.angle import Angle |
There was a problem hiding this comment.
Suggested change
| from stonesoup.types.angle import Angle |
Unused Import
| covar=np.diag([400*weights[j][i], 0, | ||
| 400*weights[j][i], 0]), | ||
| timestamp=timesteps[j]) | ||
| for j in range(simulation_length)]) |
There was a problem hiding this comment.
Suggested change
| for j in range(simulation_length)]) | |
| for j in range(sim_length)]) |
Comment on lines
+380
to
+387
| sensor_history_b, search_cell_history_b, probs_b = search_loop(prior, sensor1, | ||
| optbrutesensormanager, timesteps, | ||
| prob_det) | ||
| sensor_history_r, search_cell_history_r, probs_r = search_loop(prior, sensor2, | ||
| randomsensormanager, timesteps, | ||
| prob_det) | ||
| sensor_history_s, search_cell_history_s, probs_s = search_loop(prior, sensor3, None, timesteps, | ||
| prob_det, seq_flag=True) |
There was a problem hiding this comment.
Suggested change
| sensor_history_b, search_cell_history_b, probs_b = search_loop(prior, sensor1, | |
| optbrutesensormanager, timesteps, | |
| prob_det) | |
| sensor_history_r, search_cell_history_r, probs_r = search_loop(prior, sensor2, | |
| randomsensormanager, timesteps, | |
| prob_det) | |
| sensor_history_s, search_cell_history_s, probs_s = search_loop(prior, sensor3, None, timesteps, | |
| prob_det, seq_flag=True) | |
| sensor_history_b, search_cell_history_b, probs_b = search_loop(prior, sensor1, | |
| optbrutesensormanager, timesteps, | |
| prob_det) | |
| sensor_history_r, search_cell_history_r, probs_r = search_loop(prior, sensor2, | |
| randomsensormanager, timesteps, | |
| prob_det) | |
| sensor_history_s, search_cell_history_s, probs_s = search_loop(prior, sensor3, None, timesteps, | |
| prob_det, seq_flag=True) |
Comment on lines
+439
to
+468
| # from plotly import graph_objects as go | ||
| # def plot_search_heatmap(plt, x_pos, y_pos, search_cell_history, **kwargs): | ||
| # | ||
| # # initialise heatmap trace | ||
| # trace_base = len(plt.fig.data) | ||
| # | ||
| # heatmap_kwargs = dict(x=[], y=[], z=[], colorscale="YlOrRd", opacity=0.6, | ||
| # showlegend=True, showscale=False, name="heatmap", | ||
| # legendgroup="heatmap") | ||
| # heatmap_kwargs.update(kwargs) | ||
| # plt.fig.add_trace(go.Heatmap(heatmap_kwargs)) | ||
| # | ||
| # # get number of traces already in plt | ||
| # # add data for each frame | ||
| # for frame in plt.fig.frames: | ||
| # | ||
| # data_ = list(frame.data) | ||
| # traces_ = list(frame.traces) | ||
| # | ||
| # frame_time = datetime.fromisoformat(frame.name) | ||
| # | ||
| # for particle_state in search_cell_history: | ||
| # if frame_time == particle_state.timestamp: | ||
| # weights = [float(w) for w in particle_state.weight] | ||
| # data_.append(go.Heatmap(x=x_pos, y=y_pos, z=weights)) | ||
| # traces_.append(trace_base) | ||
| # frame.traces = traces_ | ||
| # frame.data = data_ | ||
| # | ||
| # return plt |
There was a problem hiding this comment.
This function isn't used for plotting in this example
| sensor_info = [copy(sensor)] | ||
| prob_found_list = [0] | ||
|
|
||
| for i, timestep in enumerate(timesteps[1:]): |
There was a problem hiding this comment.
Suggested change
| for i, timestep in enumerate(timesteps[1:]): | |
| for timestep in timesteps[1:]: |
i is unused
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Add two examples showcasing the use of Bayesian search in Stone Soup. Note that much of length of the two files comes from plotting functions which are hidden by default in the rendered examples.