# physical-superintelligence-lab/simple architecture

> Architecture diagram of the GitHub repository physical-superintelligence-lab/simple, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-03). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/physical-superintelligence-lab/simple
- Repository: https://github.com/physical-superintelligence-lab/simple

## Overview

SIMPLE is a simulation platform for robot policy learning and evaluation, especially humanoid loco-manipulation. A typical workflow defines a task and robot, builds a randomized scene, runs it in a simulator, and uses teleoperation or a policy to produce actions and demonstrations; evaluation runs policies against simulation and records results. The graph also shows planning, dataset handling, and the documented client-server evaluation mode. Edges reflect the excerpts, SOURCE INDEX references, or README; unsampled internals are represented only where the tree or README establishes a capability.

Main entry points: [`src/simple/core/task.py`](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/core/task.py) (Task definition), [`src/simple/robots/g1_inspire_wholebody.py`](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/robots/g1_inspire_wholebody.py) (Robot models), [`src/simple/cli/teleop_wbc.py`](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/cli/teleop_wbc.py) (Teleoperation), [`src/simple/evals/env_runner.py`](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/evals/env_runner.py) (Evaluation runner)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_world["Task and world"]
  node_task["Task definition<br/>[task.py]"]
  node_randomization["Scene randomization<br/>[manager.py]"]
  node_layout["World layout<br/>[layout.py]"]
  node_scene["Scene assets<br/>[scene_manager.py]"]
  node_robot["Robot models"]
  node_sensors["Sensor configuration<br/>[config.py]"]
end

subgraph group_runtime["Simulation runtime"]
  node_simulator["Simulation environment<br/>[base_dual_env.py]"]
  node_isaac["Isaac Sim backend<br/>[isaacsim.py]"]
  node_mujoco["MuJoCo backend<br/>[mujoco.py]"]
end

subgraph group_control["Control and planning"]
  node_agents["Policy and agents<br/>[base_agent.py]"]
  node_teleop["Teleoperation<br/>[teleop_wbc.py]"]
  node_controllers["Robot controllers<br/>[eef.py]"]
  node_planner["Motion planning<br/>[curobo.py]"]
  node_grasps["Grasp estimation<br/>[bodex.py]"]
end

subgraph group_dataeval["Data and evaluation"]
  node_collection["Data collection<br/>[datagen.py]"]
  node_datasets[("Dataset formats<br/>[lerobot.py]")]
  node_processing["Dataset processing"]
  node_evaluation["Evaluation runner<br/>[env_runner.py]"]
  node_metrics["Evaluation metrics<br/>[metrics.py]"]
end

node_user(("Operator"))
node_policy_server["Policy server"]

node_user -->|"operates"| node_teleop
node_user -.->|"starts collection"| node_collection
node_task -->|"resets with"| node_randomization
node_task -->|"builds"| node_layout
node_task -->|"configures"| node_robot
node_task -->|"configures"| node_sensors
node_randomization -->|"randomizes"| node_layout
node_isaac -->|"runs"| node_task
node_mujoco -->|"runs"| node_task
node_teleop -->|"uses"| node_agents
node_teleop -->|"commands"| node_controllers
node_teleop -->|"operates"| node_simulator
node_robot -->|"uses"| node_controllers
node_robot -->|"uses policy model"| node_mujoco
node_planner -->|"plans for"| node_robot
node_planner -->|"reads task"| node_task
node_planner -->|"reads layout"| node_layout
node_grasps -->|"uses robot interface"| node_robot
node_collection -.->|"collects from"| node_simulator
node_collection -.->|"writes episodes"| node_datasets
node_processing -.->|"processes data"| node_datasets
node_evaluation -->|"runs episodes"| node_simulator
node_evaluation -->|"reads and writes"| node_datasets
node_evaluation -.->|"reports results"| node_metrics
node_policy_server -.->|"serves inference"| node_evaluation

click node_task "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/core/task.py"
click node_randomization "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/dr/manager.py"
click node_layout "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/core/layout.py"
click node_scene "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/scenes/scene_manager.py"
click node_robot "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/robots/g1_inspire_wholebody.py"
click node_sensors "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/sensors/config.py"
click node_simulator "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/envs/base_dual_env.py"
click node_isaac "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/engines/isaacsim.py"
click node_mujoco "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/engines/mujoco.py"
click node_agents "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/agents/base_agent.py"
click node_teleop "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/cli/teleop_wbc.py"
click node_controllers "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/robots/controllers/eef.py"
click node_planner "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/mp/curobo.py"
click node_grasps "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/grasps/bodex.py"
click node_collection "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/cli/datagen.py"
click node_datasets "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/datasets/lerobot.py"
click node_processing "https://github.com/physical-superintelligence-lab/simple/blob/main/scripts/postprocess_psi0.py"
click node_evaluation "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/evals/env_runner.py"
click node_metrics "https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/evals/metrics.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_task,node_randomization,node_layout,node_scene,node_robot,node_sensors toneBlue
class node_simulator,node_isaac,node_mujoco toneAmber
class node_agents,node_teleop,node_controllers,node_planner,node_grasps,node_policy_server toneMint
class node_collection,node_datasets,node_processing,node_evaluation,node_metrics toneRose
class node_user toneIndigo
```

## Components

### Task and world

- [Task definition](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/core/task.py): `src/simple/core/task.py`
- [Scene randomization](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/dr/manager.py): `src/simple/dr/manager.py`
- [World layout](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/core/layout.py): `src/simple/core/layout.py`
- [Scene assets](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/scenes/scene_manager.py): `src/simple/scenes/scene_manager.py`
- [Robot models](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/robots/g1_inspire_wholebody.py): `src/simple/robots/g1_inspire_wholebody.py`
- [Sensor configuration](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/sensors/config.py): `src/simple/sensors/config.py`

### Simulation runtime

- [Simulation environment](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/envs/base_dual_env.py): `src/simple/envs/base_dual_env.py`
- [Isaac Sim backend](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/engines/isaacsim.py): `src/simple/engines/isaacsim.py`
- [MuJoCo backend](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/engines/mujoco.py): `src/simple/engines/mujoco.py`

### Control and planning

- [Policy and agents](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/agents/base_agent.py): `src/simple/agents/base_agent.py`
- [Teleoperation](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/cli/teleop_wbc.py): `src/simple/cli/teleop_wbc.py`
- [Robot controllers](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/robots/controllers/eef.py): `src/simple/robots/controllers/eef.py`
- [Motion planning](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/mp/curobo.py): `src/simple/mp/curobo.py`
- [Grasp estimation](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/grasps/bodex.py): `src/simple/grasps/bodex.py`

### Data and evaluation

- [Data collection](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/cli/datagen.py): `src/simple/cli/datagen.py`
- [Dataset formats](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/datasets/lerobot.py): `src/simple/datasets/lerobot.py`
- [Dataset processing](https://github.com/physical-superintelligence-lab/simple/blob/main/scripts/postprocess_psi0.py): `scripts/postprocess_psi0.py`
- [Evaluation runner](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/evals/env_runner.py): `src/simple/evals/env_runner.py`
- [Evaluation metrics](https://github.com/physical-superintelligence-lab/simple/blob/main/src/simple/evals/metrics.py): `src/simple/evals/metrics.py`

### Other components

- **Operator**
- **Policy server**

## Connections

- Operator → Teleoperation: operates (evidence: `src/simple/cli/teleop_wbc.py`)
- Operator → Data collection: starts collection
- Task definition → Scene randomization: resets with (evidence: `src/simple/core/task.py`)
- Task definition → World layout: builds (evidence: `src/simple/core/task.py`)
- Task definition → Robot models: configures (evidence: `src/simple/core/task.py`)
- Task definition → Sensor configuration: configures (evidence: `src/simple/core/task.py`)
- Scene randomization → World layout: randomizes (evidence: `src/simple/dr/manager.py`)
- Isaac Sim backend → Task definition: runs (evidence: `src/simple/engines/isaacsim.py`)
- MuJoCo backend → Task definition: runs (evidence: `src/simple/engines/mujoco.py`)
- Teleoperation → Policy and agents: uses (evidence: `src/simple/cli/teleop_wbc.py`)
- Teleoperation → Robot controllers: commands (evidence: `src/simple/cli/teleop_wbc.py`)
- Teleoperation → Simulation environment: operates (evidence: `src/simple/cli/teleop_wbc.py`)
- Robot models → Robot controllers: uses (evidence: `src/simple/robots/g1_inspire_wholebody.py`)
- Robot models → MuJoCo backend: uses policy model (evidence: `src/simple/robots/g1_inspire_wholebody.py`)
- Motion planning → Robot models: plans for (evidence: `src/simple/mp/curobo.py`)
- Motion planning → Task definition: reads task (evidence: `src/simple/mp/curobo.py`)
- Motion planning → World layout: reads layout (evidence: `src/simple/mp/curobo.py`)
- Grasp estimation → Robot models: uses robot interface (evidence: `src/simple/grasps/bodex.py`)
- Data collection → Simulation environment: collects from
- Data collection → Dataset formats: writes episodes
- Dataset processing → Dataset formats: processes data
- Evaluation runner → Simulation environment: runs episodes (evidence: `src/simple/evals/env_runner.py`)
- Evaluation runner → Dataset formats: reads and writes (evidence: `src/simple/evals/env_runner.py`)
- Evaluation runner → Evaluation metrics: reports results
- Policy server → Evaluation runner: serves inference (evidence: `README.md`)
