# divaksh06/fast-trained-latent-adaptive-quadrotor-control architecture

> Architecture diagram of the GitHub repository divaksh06/fast-trained-latent-adaptive-quadrotor-control, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-04). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control
- Repository: https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control

## Overview

This repository is an early single-platform quadrotor hover-control RL baseline. Training creates or resumes a TD3 policy in a custom PyBullet hover environment; separate simulation and evaluation scripts run saved policies and report performance. The README also describes future latent adaptation and uncertainty-gated safety, but those modules are not present in the tree, so they are not modeled as implemented components. Arrows reflect only calls and data flow visible in the excerpts or README.

Main entry points: [`train/simulate_hover.py`](https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/train/simulate_hover.py) (Hover diagnostics), [`envs/paper_hover_env.py`](https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/envs/paper_hover_env.py) (Paper hover environment), [`train/train_td3_paper.py`](https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/train/train_td3_paper.py) (TD3 training), [`train/eval_td3_paper.py`](https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/train/eval_td3_paper.py) (Interactive evaluation)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_learning["Policy learning"]
  node_trainer["TD3 training<br/>[train_td3_paper.py]"]
  node_policy["TD3 policy<br/>[train_td3_paper.py]"]
end

subgraph group_environment["Hover simulation"]
  node_hover_env["Paper hover environment<br/>[paper_hover_env.py]"]
  node_pybullet["PyBullet drone simulator"]
end

subgraph group_assessment["Policy assessment"]
  node_simulator["Hover diagnostics<br/>[simulate_hover.py]"]
  node_report["Simulation reports<br/>[simulate_hover.py]"]
  node_evaluator["Interactive evaluation<br/>[eval_td3_paper.py]"]
end

node_engineer(("Researcher"))

node_engineer -->|"starts training"| node_trainer
node_trainer -->|"creates environment"| node_hover_env
node_hover_env -->|"steps physics"| node_pybullet
node_trainer -->|"trains and saves"| node_policy
node_engineer -->|"runs diagnostics"| node_simulator
node_simulator -->|"loads checkpoint"| node_policy
node_simulator -->|"runs episodes"| node_hover_env
node_simulator -->|"generates"| node_report
node_engineer -->|"runs evaluation"| node_evaluator
node_evaluator -->|"loads checkpoint"| node_policy
node_evaluator -->|"runs episodes"| node_hover_env

click node_trainer "https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/train/train_td3_paper.py"
click node_policy "https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/train/train_td3_paper.py"
click node_hover_env "https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/envs/paper_hover_env.py"
click node_simulator "https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/train/simulate_hover.py"
click node_report "https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/train/simulate_hover.py"
click node_evaluator "https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/train/eval_td3_paper.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_trainer,node_policy toneBlue
class node_hover_env,node_pybullet toneAmber
class node_simulator,node_report,node_evaluator toneMint
class node_engineer toneIndigo
```

## Components

### Policy learning

- [TD3 training](https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/train/train_td3_paper.py): `train/train_td3_paper.py`
- [TD3 policy](https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/train/train_td3_paper.py): `train/train_td3_paper.py`

### Hover simulation

- [Paper hover environment](https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/envs/paper_hover_env.py): `envs/paper_hover_env.py`
- **PyBullet drone simulator**

### Policy assessment

- [Hover diagnostics](https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/train/simulate_hover.py): `train/simulate_hover.py`
- [Simulation reports](https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/train/simulate_hover.py): `train/simulate_hover.py`
- [Interactive evaluation](https://github.com/divaksh06/fast-trained-latent-adaptive-quadrotor-control/blob/main/train/eval_td3_paper.py): `train/eval_td3_paper.py`

### Other components

- **Researcher**

## Connections

- Researcher → TD3 training: starts training (evidence: `README.md`)
- TD3 training → Paper hover environment: creates environment (evidence: `train/train_td3_paper.py`)
- Paper hover environment → PyBullet drone simulator: steps physics (evidence: `envs/paper_hover_env.py`)
- TD3 training → TD3 policy: trains and saves (evidence: `train/train_td3_paper.py`)
- Researcher → Hover diagnostics: runs diagnostics (evidence: `README.md`)
- Hover diagnostics → TD3 policy: loads checkpoint (evidence: `train/simulate_hover.py`)
- Hover diagnostics → Paper hover environment: runs episodes (evidence: `train/simulate_hover.py`)
- Hover diagnostics → Simulation reports: generates (evidence: `train/simulate_hover.py`)
- Researcher → Interactive evaluation: runs evaluation (evidence: `train/eval_td3_paper.py`)
- Interactive evaluation → TD3 policy: loads checkpoint (evidence: `train/eval_td3_paper.py`)
- Interactive evaluation → Paper hover environment: runs episodes (evidence: `train/eval_td3_paper.py`)
