Position filtering 2

[szeyoong-low]

Position refiners noise filtering

• Implemented a FIR filter (utama_core/run/refiners/filters.py)
• Wrote unit tests to verify that mean squared error is reduced across the board (utama_core/tests/refiners/noise_filter_test.py)
• Integrated filters into the position refiners (utama_core/run/refiners/position.py)
• Added utilities for exporting vision data and real time tracking of robot coordinates during live testing (utama_core/run/refiners/position.py)
• Modified other unit tests affected by PositionRefiner taking more arguments necessary for the filters (utama_core/tests/refiners/position_unit_test.py and position_refiner_integration_test.py)
• Added a folder (vision_data) with various datasets generated from simulations for use in unit testing and running analytics
• Analysed the filters to optimise parameters and verify effectiveness in a Jupyter notebook (vision_data/filter_results_summary.ipynb)
• Added multiple dependencies to Pixi: scipy, seaborn, pyqtgraph, pyside6 (pixi.toml)

Completed by @szeyoong-low and @AndrewAha

[Copilot]

Pull request overview

This PR implements FIR (Finite Impulse Response) filtering for noise reduction in robot position tracking from vision data. The implementation includes a new filter module, comprehensive unit tests validating MSE reduction, and integration into the position refinement pipeline with support for both friendly and enemy robot tracking.

Key Changes:

• New FIR filter implementation with configurable parameters for position smoothing
• Unit tests demonstrating filter effectiveness across 6 robots using real simulation data
• Updated PositionRefiner API to accept team color and robot count parameters for filter initialization
• Added visualization and analysis dependencies (scipy, seaborn, pyqtgraph, pyside6)

Reviewed changes

Copilot reviewed 17 out of 24 changed files in this pull request and generated 11 comments.

Show a summary per file

File	Description
utama_core/run/refiners/filters.py	New FIR filter implementation with position and orientation filtering support
utama_core/tests/refiners/noise_filter_test.py	Comprehensive unit tests validating MSE reduction for x, y, and vector positions
utama_core/run/refiners/position.py	Integration of filters into position refiners (not shown in diff)
utama_core/tests/refiners/position_unit_test.py	Updated test signatures to match new PositionRefiner constructor
utama_core/tests/refiners/position_refiner_integration_test.py	Updated integration test with new PositionRefiner parameters
utama_core/run/strategy_runner.py	Updated PositionRefiner instantiation with team and robot count parameters
utama_core/entities/data/vision.py	Added Gaussian noise injection method for testing
pixi.toml	Added scipy, seaborn, pyqtgraph, and pyside6 dependencies
vision_data/*.csv	Test data files for filter validation
vision_data/.jupyter/*	Jupyter workspace configuration (should not be committed)
main.py	Changed mode from "rsim" to "grsim"

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

This comment appears to contain commented-out code.

Suggested change

# robot.add_gaussian_noise()

[Copilot]

This comment appears to contain commented-out code.

Suggested change

# self.data_collected += 1

[Copilot]

This comment appears to contain commented-out code.

Suggested change

	# # for b_robot in combined_vision_data.blue_robots:
	# # writer.writerow({
	# # TS_COL: combined_vision_data.ts,
	# # ID_COL: b_robot.id,
	# # COLOR_COL: "blue",
	# # X_COL: b_robot.x,
	# # Y_COL: b_robot.y,
	# # TH_COL: b_robot.orientation
	# # })
	# self.data_collected += 1
	# For live testing:
	# try:
	# current_coord = np.array((
	# filtered_vision_data.yellow_robots[2].x,
	# filtered_vision_data.yellow_robots[2].y
	# ))
	# if not np.isnan(self._last_coord).any():
	# self._delta_stream.append(np.linalg.norm(current_coord - self._last_coord))
	# self._last_coord = current_coord
	# except:
	# pass
	# finally:
	# self._curve.setData(self._delta_stream)
	# try:
	# self._x_stream.append(filtered_vision_data.yellow_robots[0].x)
	# self._y_stream.append(filtered_vision_data.yellow_robots[0].y)
	# except:
	# pass
	# finally:
	# pass
	# finally:

[Copilot]

This comment appears to contain commented-out code.

[Copilot]

This comment appears to contain commented-out code.

Suggested change

# self._curve.setData(self._x_stream, self._y_stream)

[Copilot]

Normal methods should have 'self', rather than 'b1', as their first parameter.

Suggested change

	def ball_merge_predicate(b1: RawBallData, b2: RawBallData) -> bool:
	return abs(b1.x - b2.x) + abs(b1.y - b2.y) < CameraCombiner.BALL_MERGE_THRESHOLD
	@staticmethod
	def ball_merge_predicate(b1: RawBallData, b2: RawBallData) -> bool:
	return abs(b1.x - b2.x) + abs(b1.y - b2.y) < CameraCombiner.BALL_MERGE_THRESHOLD
	@staticmethod

[Copilot]

Normal methods should have 'self', rather than 'b1', as their first parameter.

Suggested change

	def ball_merge_predicate(b1: RawBallData, b2: RawBallData) -> bool:
	return abs(b1.x - b2.x) + abs(b1.y - b2.y) < CameraCombiner.BALL_MERGE_THRESHOLD
	@staticmethod
	def ball_merge_predicate(b1: RawBallData, b2: RawBallData) -> bool:
	return abs(b1.x - b2.x) + abs(b1.y - b2.y) < CameraCombiner.BALL_MERGE_THRESHOLD
	@staticmethod

[Copilot]

Import of 'plt' is not used.

Suggested change

import matplotlib.pyplot as plt

[Copilot]

Import of 'pg' is not used.

Suggested change

import pyqtgraph as pg # type: ignore

[Copilot]

Import of 'QtWidgets' is not used.

Suggested change

from pyqtgraph.Qt import QtWidgets # type: ignore