ChenglinLiuChris/commtr-fig6-comparison

COMMTR Figure 6 — Agent-Pilot vs Human Pilots Landing Comparison

This package contains the data used to produce Figure 6 of the manuscript "Agent-Pilot" (submitted to Communications in Transportation Research, manuscript ID COMMTR-2026-0064), specifically Section 4.3.3 "Comparison with Human Pilots".

Figure 6 compares Agent-Pilot against ten trained human pilots on attitude control (pitch and bank angles, Fig. 6a–6b) and lateral tracking accuracy (cross-track error, XTE, Fig. 6c) during the final approach (0.2 nm / last ~0.3 nm segment) at the urban vertiport under clear and fog conditions, in Microsoft Flight Simulator (MSFS) 2024.

Directory structure

commtr_fig6_data/
├── README.md
├── SHA256SUMS
├── human_pilots_landing/          # Figure 6 comparison package (agent + human + figures)
│   ├── city_reset_pitch_bank.csv  # Agent-Pilot, Urban Clear, per-trial pitch/bank
│   ├── city_fog_pitch_bank.csv    # Agent-Pilot, Urban Fog, per-trial pitch/bank
│   ├── city_reset_xte.csv         # Agent-Pilot, Urban Clear, per-trial XTE
│   ├── city_fog_xte.csv           # Agent-Pilot, Urban Fog, per-trial XTE
│   ├── time_raw_statistics_0.3nm.csv  # HUMAN pilots, last-0.3-nm segment statistics
│   │                               #   (filename suffix -1 = Clear, -2 = Fog)
│   ├── compare_landing_data.py    # script that generates the comparison plots
│   ├── calculate_time_raw_statistics.py
│   ├── pitch_bank_comparison.png/.pdf/.svg   # Figure 6a–6b
│   └── xte_comparison.png/.pdf/.svg          # Figure 6c
├── agent_pitch_bank_results/      # Agent-Pilot per-trial pitch/bank for ALL 16 scenarios
│   ├── city_reset.csv, city_fog.csv, ... (16 scenario CSVs)
│   └── summary_statistics.csv
└── agent_xte_result/              # Agent-Pilot per-trial XTE for ALL 16 scenarios
    ├── city_reset.csv, city_fog.csv, ... (16 scenario CSVs)
    └── summary_statistics.csv

1. Agent-Pilot data (per-trial metrics)

• agent_pitch_bank_results/ — per-trial pitch and bank statistics (pitch_abs_mean, pitch_abs_sd, bank_abs_mean, bank_abs_sd, total_rows) for the 16 vertiport–weather scenarios, plus summary_statistics.csv (scenario-level averages).
• agent_xte_result/ — per-trial cross-track-error statistics (MAE, RMSE, SD, Max_XTE, Min_XTE, Max_Left, Max_Right, baseline_length) for the 16 scenarios, plus summary_statistics.csv.

The Urban Clear (city_reset) and Urban Fog (city_fog) rows are the Agent-Pilot data plotted in Figure 6; they are also copied into human_pilots_landing/ for the comparison.

2. Human pilots data (n = 10 per condition)

human_pilots_landing/time_raw_statistics_0.3nm.csv — statistics over the last 0.3 nautical mile segment of each human landing:

• pitch_abs_mean, pitch_abs_sd, bank_abs_mean, bank_abs_sd
• baseline_length_m, xte_abs_mean_m, xte_abs_sd_m, xte_abs_rmse_m
• filename convention: YYYY-MM-DD-HH-MM<initials>-<n>.csv where -1 = Clear (Reset) condition and -2 = Fog condition.

compare_landing_data.py selects the first 10 trials of each condition (Clear, Fog) to match the ten human pilots reported in the manuscript and produces the two comparison figures.

3. Figures

• pitch_bank_comparison.png/.pdf/.svg — Figure 6a–6b (pitch & bank, four sub-panels).
• xte_comparison.png/.pdf/.svg — Figure 6c (XTE MAE / SD / RMSE).

4. Provenance

• Server: sc2i (10.30.58.120), directory /data2/Chenglin/agent/flight_recorder/.
• Raw per-trial flight telemetry (35 Hz recorder) for the full 173 Agent-Pilot trials is available at /data2/Chenglin/agent/flight_recorder/agent-pilot-recorder_All173/ (5.5 GB) and the successful subset at agent-pilot-recorder_successfull/ (4.7 GB); these are not included in this package but can be provided on request.

5. License / citation / contact

• License: to be confirmed before release.
• Citation: please cite the Agent-Pilot manuscript (COMMTR-2026-0064).
• Contact: [author email to be added].