Releases: bprobert97/accord
Releases · bprobert97/accord
v3.0
What's Changed
- Increase font sizes by @bprobert97 in #13
- Add monte carlo by @bprobert97 in #14
- Add web app by @bprobert97 in #15
- Make requirements.txt file linux by default by @bprobert97 in #16
- Get streamlit app deployed by @bprobert97 in #17
- Update sim data by @bprobert97 in #18
- Update mc metrics by @bprobert97 in #19
- Fix web app add convergence and release by @bprobert97 in #20
- New version release by @bprobert97 in #21
Full Changelog: v2.2...v3.0
v2.2
What's Changed
- Update median values, add median percentile check by @bprobert97 in #12
Full Changelog: v2.1...v2.2
v2.1
v2.0
What's Changed
- Fig update by @bprobert97 in #3
- Constellation plotting by @bprobert97 in #4
- Improve constellation sim by @bprobert97 in #5
- Increase sim size by @bprobert97 in #6
- Add isl ranges by @bprobert97 in #7
- Increase n to 200 by @bprobert97 in #8
- 400 sats with graphs by @bprobert97 in #9
- Update figures for paper by @bprobert97 in #10
Full Changelog: v1.1...v2.0
v1.1
What's Changed
- Update files for paper release with zenodo by @bprobert97 in #2
Full Changelog: v1.0...v1.1
v1.0
[v1.0] - 2025-11
Summary
This is the first full release of the PoISE consensus mechanism. This has been released to accompany a presentation at the 2026 SciTech Forum.
Added
- Pytest unit tests for all code.
- Moved plotting into its own file.
- Correctness score now accounts for average NIS, following chi-squared statistics
- Reputation is now based on chi-squared statistics for bounds, and previous behaviour
- Simulates faulty and malicious nodes in accord_demo.py
- Uses a simpler EKF implementation with Filterpy
Full Changelog: v0.2.1...v1.0
v0.2.1
[v0.2.1] - 2025-10
Summary
This is a patch release of the PoISE consensus mechanism, providing key bug fixes.
Added
- Bug fix in od_filter.py to allow LOS only measurements to be processed correctly.
- Additional plots for chi2 functions and consensus scores.
v0.2
[v0.2] - 2025-09
Summary
This is the second proof-of-concept release of the PoISE consensus mechanism, using MATLAB for generating satellite simulation data and a second order extended kalman filter with a state transition tensor for orbit determination calculations.
Added
- Replaced TLE data inputs with simulated sensor data created by a MATLAB script.
- Implemented a Second Order Extended Kalman Filter with State Tranisiotn Tensor for permising Orbit Determination Calculations.
- Added the Normalised Innovation Score (NIS) to the consensus mechanism calculations.
- Simulation of both good and malicious data submissions for a small network of satellites.
- Generation of plots showing:
- Chi Squared probability distributions
- Addition of logging.
- Additional configuration for linting and type checking tools.
- Created new diagrams describing the orbit determination process.
v0.1
This is the first proof of concept release of the PoISE consensus mechanism. It:
- Uses TLE data to propagate orbital data using sgp4 and the skyfield library
- For one satellite, simulates the submission of good and malicious data
- Creates plots of the DAG structure and the satellite reputation
- Uses asyncio for asynchronous message communications