README.md

---
license: apache-2.0
task_categories:
- robotics
tags:
- LeRobot
- ledrone
- uav
- drone
- sim
- pybullet
- hover
- imitation-learning
- goal-conditioned
- state-based
size_categories:
- 10K<n<100K
configs:
- config_name: default
  data_files: data/*/*.parquet
---

# ledrone_pybullet_hover

A **goal-conditioned, state-based** imitation-learning dataset of a quadrotor
performing the **hover** task (reach a fixed target position and station-keep), generated in **PyBullet** rigid-body physics
with a privileged analytic expert. Each frame pairs the drone's 13-dim proprioceptive
state and a relative goal with the expert's velocity + yaw-rate command — ready to
train goal-conditioned control policies
([Diffusion Policy](https://diffusion-policy.cs.columbia.edu/), ACT, …). A
first-person-view (FPV) camera is shipped as an **optional auxiliary** stream (the
task is solvable from state alone).

Built with [LeRobot](https://github.com/huggingface/lerobot) (format `v3.0`) as part
of the [ledrone](https://github.com/ahive-org/ledrone) project.

<a class="flex" href="https://huggingface.co/spaces/lerobot/visualize_dataset?path=ahive/ledrone_pybullet_hover">
<img class="block dark:hidden" src="https://huggingface.co/datasets/huggingface/badges/resolve/main/visualize-this-dataset-xl.svg"/>
<img class="hidden dark:block" src="https://huggingface.co/datasets/huggingface/badges/resolve/main/visualize-this-dataset-xl-dark.svg"/>
</a>

## At a glance

| | |
|---|---|
| Task | `hover` — reach a fixed target position and station-keep |
| Type | Goal-conditioned, **state-based** (FPV image is an optional auxiliary) |
| Episodes | 300 |
| Frames | 60,000 (200/episode, 6.7s at 30 Hz) |
| Robot type | `ledrone` (quadrotor) |
| Physics | PyBullet rigid-body dynamics via [gym-pybullet-drones](https://github.com/utiasDSL/gym-pybullet-drones) |
| Expert | Analytic PD + reference-velocity feed-forward + course-to-goal yaw, slew-limited startup (privileged, **state-only**) |
| Camera | FPV 96×96 RGB (AV1) — auxiliary; scene has a goal marker, landmarks, textured ground |
| Action | NED velocity + yaw rate |
| Format | LeRobot `v3.0` |

## How it was generated

A privileged analytic **state-only** expert (PD + reference-velocity feed-forward)
flies the `hover` reference in the `LeDrone` env on the **PyBullet** backend.

- **DART-style coverage** ([Laskey et al. 2017](https://arxiv.org/abs/1703.09327)): the
  drone is driven by a *noised* command (σ = 0.15 m/s on the velocity channels) but each
  frame is **labelled with the clean expert command** — so the policy sees
  off-distribution states *and* the optimal correction for them, mitigating
  behavior-cloning covariate shift.
- **Slew-rate-limited setpoint** ramps the command from rest, taming the low-level
  controller's start-of-episode transient (early-episode tilt/rate spikes).
- **Controlled heading** (course-to-goal yaw); **randomized** start + goal per episode;
  the goal is stored **relative to current position** (origin-invariant).
- **Convergence gate**: an episode is accepted only once the expert reaches / tracks the
  goal (else the start/goal is resampled).

Reproduce (needs the `[pybullet]` extra):

```bash
python tools/generate_expert_dataset.py --task hover --backend pybullet \
    --repo-id ahive/ledrone_pybullet_hover --episodes 300 --push-to-hub --public
```

## Features

| Key | Dtype | Shape | Description |
|---|---|---|---|
| `observation.state` | float32 | (13,) | `pos_n/e/d`, `vel_n/e/d`, attitude quat `q_w/x/y/z`, FRD body rates `rate_x/y/z` |
| `observation.environment_state` | float32 | (3,) | Goal relative to current position: `err_n`, `err_e`, `err_d` |
| `action` | float32 | (4,) | Command: `vx`, `vy`, `vz`, `yaw_rate` |
| `observation.images.fpv` | video | (96, 96, 3) | Forward-facing FPV camera, AV1 (**auxiliary**) |

State is the SOTA quadrotor policy observation (relative goal + velocity + orientation +
body rates); **battery is intentionally excluded** (handled by domain randomization, not
observed — Hwangbo 2017, Molchanov 2019, Kaufmann ICRA 2022 / Nature 2023). Attitude is
a unit quaternion; map to a continuous 6D rep
([Zhou et al., CVPR 2019](https://arxiv.org/abs/1812.07035)) if desired. `action` is a
NED velocity + yaw rate clipped to the envelope (≤ 2 m/s horizontal, ≤ 1 m/s climb,
≤ 45 °/s yaw).

## Usage

```python
from lerobot.datasets.lerobot_dataset import LeRobotDataset
ds = LeRobotDataset("ahive/ledrone_pybullet_hover")
frame = ds[0]  # observation.state (13,), observation.environment_state (3,), action (4,)
```

## Intended use & limitations

- **Intended use**: goal-conditioned, **state-based** imitation for quadrotor `hover`
  (state + relative goal → velocity + yaw-rate). 300 episodes, all in `train`.
- **State-based, not visuomotor.** The task is solvable from `observation.state` +
  `observation.environment_state`; the FPV goal marker is in only **~7% of frames**.
  The image is an **optional auxiliary** input, not the primary task signal.
- **DART coverage, but no hard failures.** The injected noise gives off-distribution
  recovery labels, but the data still contains no crash / large-perturbation / failure
  demonstrations.
- **Simulation only** (PyBullet / gym-pybullet-drones, Crazyflie 2.x). The FPV scene is a
  synthetic benchmark scene, not photorealistic / sim-to-real imagery.


## Data analysis

Analysis across all three `ledrone` tasks, computed from this DART-regenerated data.

**Signature trajectories** — takeoff climbs from the ground, hover converges & station-keeps, circle flies full loops (>= 360 deg).

![Trajectories](figures/trajectories.png)

**Convergence & attitude** — every task drives the goal error to near-zero (circle keeps a small steady tracking lag from following a moving target); body tilt stays gentle (median ~10 deg, 95% <= 35 deg).

![Convergence and attitude](figures/convergence_attitude.png)

**Command / action space** — takeoff & hover station-keep at ~0.1 m/s while circle orbits at ~1.3 m/s; the climb command is bimodal for takeoff; yaw-rate is centered on the goal bearing.

![Action space](figures/action_space.png)

## License & citation

Apache-2.0. Please cite the [ledrone](https://github.com/ahive-org/ledrone) project.