{"id":26129229,"url":"https://github.com/dgac/acropole","last_synced_at":"2026-03-16T20:32:55.957Z","repository":{"id":220510026,"uuid":"623391257","full_name":"DGAC/Acropole","owner":"DGAC","description":"This repository contains the Acropole model for aircraft fuel flow prediction and Python packages for aircraft trajectory processing and fuel flow enhancement.","archived":false,"fork":false,"pushed_at":"2024-08-29T11:32:37.000Z","size":3741,"stargazers_count":20,"open_issues_count":0,"forks_count":4,"subscribers_count":5,"default_branch":"main","last_synced_at":"2025-04-13T18:45:56.311Z","etag":null,"topics":["air-traffic-management","aircraft-performance","artificial-intelligence","environment","machine-learning"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"agpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/DGAC.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2023-04-04T09:25:21.000Z","updated_at":"2025-03-19T07:10:39.000Z","dependencies_parsed_at":"2024-05-16T01:45:42.727Z","dependency_job_id":"bfa3880d-e00f-481a-ac75-169c73e6733c","html_url":"https://github.com/DGAC/Acropole","commit_stats":null,"previous_names":["dgac/acropole"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/DGAC/Acropole","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DGAC%2FAcropole","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DGAC%2FAcropole/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DGAC%2FAcropole/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DGAC%2FAcropole/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/DGAC","download_url":"https://codeload.github.com/DGAC/Acropole/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DGAC%2FAcropole/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":262212985,"owners_count":23276024,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["air-traffic-management","aircraft-performance","artificial-intelligence","environment","machine-learning"],"created_at":"2025-03-10T19:40:06.025Z","updated_at":"2026-03-16T20:32:55.925Z","avatar_url":"https://github.com/DGAC.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Acropole \u003cimg src=\"https://github.com/DGAC/Acropole/blob/main/logo.png\" width=\"30\"\u003e\n\nThis repository contains the Acropole model for aircraft fuel flow prediction and Python packages for aircraft trajectory processing and fuel flow enhancement.\n\n## Easy Install\n\nFor a trouble-free installation, creating a dedicated anaconda environment is recommended :\n\n```sh\nconda create -n acropole python=3.12 -c conda-forge\n```\n\nActivate the conda environment :\n\n```sh\nconda activate acropole\n```\n\nInstall this library:\n\n```sh\ngit clone https://github.com/DGAC/Acropole.git\ncd Acropole\npip install .\n\n```\n\n## Example of use\n\nHere is a minimal working example:\n\n```python\nimport pandas as pd\nfrom acropole import FuelEstimator\n\nfe = FuelEstimator()\n\nflight = pd.DataFrame({\n    \"typecode\": [\"A320\", \"A320\", \"A320\", \"A320\"],\n    \"groundspeed\": [400, 410, 420, 430],\n    \"altitude\": [10000, 11000, 12000, 13000],\n    \"vertical_rate\": [2000, 1500, 1000, 500],\n\n    # optional features:\n    \"second\": [0.0, 1.0, 2.0, 3.0],\n    \"airspeed\": [400, 410, 420, 430],\n    \"mass\": [60000, 60000, 60000, 60000]\n})\n\nflight_fuel = fe.estimate(flight) #flight.data if traffic flight\n```\n\nNote:\n\n- When the `second` column is provided, the fuel estimation is more accurate,\n  especially due to **derivatives of speeds** (acceleration) used in the estimation.\n- `airspeed` is optional. If not provided, it is assumed to be equal\n  to groundspeed. However, accurate airspeed is recommended for better estimation.\n- Expected sampling rate is 4 seconds, higher or lower sampling rate might induce noisier fuel flow. Resampling data before estimating fuel flow is recommanded.\n\nFor a more complete example, refer to `examples/fuel_estimation.ipynb`\n\n## Aircraft data and estimation models\n\nAircraft parameters from open data to feed the model are available in `data/aircraft_params.csv` and loaded by default. Model data is available in `models/` and also loaded by default.\n\nYou can specify your own data and model file with the following initialization of `FuelEstimator`. You need to make sure the same column names are in your aircraft CSV file.\n\n```python\nfe = FuelEstimator(\n    aircraft_params_path=\"path/to/your/data.csv\",\n    model_path=\"path/to/your/SavedModel/\",\n)\n```\n\n## Model training and evaluation\n\nThe Acropole model is a neural network built using data from Quick Access Recorder (QAR) from different aircraft types. Evaluation of the model and list of aircraft is available at https://github.com/DGAC/Acropole/tree/main/evaluation/Dense_Acropole_FuelFlow_Scaling.\n\n\n## Comparison of Different Model Performances\n\nComparison of different model performances per phase for 1000 test flights of A320-214 aircraft using real mass and true airspeed.\n\n| Phase | Samples \\# | Metric | ACROPOLE | OpenAP | OpenAP V2 | BADA  | Poll-Schumann |\n|-------|------------|--------|--------------|--------------|------------|---------------|----------------|\n|       |            | **MAPE (%)**  | 2.13    | 30.35                     | 8.84       | 6.53                       | 6.85                                       |\n| CLIMB | 1,403,850   | **MAE (kg/min)** | 1.66          | 25.81                     | 6.92       | 5.53                       | 5.65                                       |\n|       |            | **ME (kg/min)**  | 0.85     | -25.66                    | -2.48      | -5.27                      | -4.62                                      |\n||||||||||\n|       |            | **MAPE (%)**  | 4.41   | 18.59                     | 10.69      | 7.01                       | 4.84                                       |\n| LEVEL | 4,017,801   | **MAE (kg/min)** | 1.82      | 7.82                      | 3.48       | 2.65                       | 2.03                                       |\n|       |            | **ME (kg/min)**  | 1.22     | -7.47                     | 2.64       | -1.43                      | -0.73                                      |\n||||||||||\n|       |            | **MAPE (%)**  | 12.63      | 51.69                     | 32.4       | 21.50                      | 21.55                                      |\n| DESCENT| 1,684,117  | **MAE (kg/min)** | 2.71         | 8.62                      | 5.58       | 3.71                       | 4.71                                       |\n|       |            | **ME (kg/min)**  | 1.88         | -1.75                     | -1.16      | -0.64                      | -3.67                                      |\n||||||||||\n|       |            | **MAPE (%)**  | 5.91       | 27.60                     | 14.71      | 9.84                       | 8.61                                       |\n| ALL   | 7,105,768   | **MAE (kg/min)** | 1.99        | 11.55                     | 4.58       | 3.44                       | 3.29                                       |\n|       |            | **ME (kg/min)**  | 1.30       | -9.92                     | 0.84       | -2.03                      | -2.09                                      |\n||||||||||\n|       |            | **Processing time (s)** | 3          | 284                      | 255        | 474                        | 28                                         |\n\n\n\n## Credits\n\n```bibtex\n@inproceedings{jarry2024towards,\n  title={Towards aircraft generic Quick Access Recorder fuel flow regression models for ADS-B data},\n  author={Jarry, Gabriel and Delahaye, Daniel and Hurter, Christophe},\n  booktitle={International Conference on Research in Air Transportation},\n  year={2024}\n}\n\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdgac%2Facropole","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fdgac%2Facropole","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdgac%2Facropole/lists"}