DETB

DETB is a simulation-first design evaluation toolkit for quadruped systems engineering. It is designed to turn Isaac Lab experiments into reviewable engineering evidence, not deployment claims or sim-to-real guarantees.

What DETB Is

• A reproducible experiment runner for quadruped design studies.
• A contract-first pipeline for manifests, metrics, and artifact bundles.
• A framework for terrain, sensor, fault, and tuning studies.
• A public-facing GitHub repository intended for technical review.

What DETB Is Not

• A field deployment stack.
• A certification tool.
• A claim of validated sim-to-real transfer.
• A generic reinforcement learning benchmark zoo.

Current Status

The repository supports a complete mock backend for fast local validation and verified Isaac Lab execution paths for train, evaluate, visualize, and train-gui. Simulator-native task and robot ownership now live under source/detb_lab, while detb/ remains the orchestration and evidence layer. On March 17, 2026, minimal real train and evaluate smoke runs completed through the baseline DETB ANYmal-C family, the DETB stability-focused family, and the first divergent DETB robot/profile family using the pinned isaaclab51 Python runtime.

Pinned Baseline

• Isaac Sim: 5.1.0
• Isaac Lab checkout: ..\IsaacLab-5.1
• Isaac Lab version file: 2.3.0
• Conda environment: isaaclab51
• Default device: cuda:0
• Default GPU index: 0

On this machine, GPU 0 resolves to NVIDIA GeForce RTX 5080 Laptop GPU with driver 581.29. DETB records the detected GPU and driver automatically in the run manifest.

Session Setup

For low-friction sessions, see docs/permissions.md. The practical target is one persistent approval for python -m detb.cli plus one for python -m pytest.

Quick Start

cd "C:\Dev Projects\NVIDIA Omniverse\IsaacLab Testbed"
python -m pip install -e .[dev]
python -m pip install -e source/detb_lab
detb train
detb evaluate
detb tune --set study=sweep
detb visualize --set visualization.execute=true
detb train-gui --set visualization.train_execute=true

GUI And Tuning Workflows

Use dry-run launch specs first so the exact Isaac Lab command is reviewable before anything opens:

detb visualize
detb train-gui

Then execute the pinned GUI path when you are ready to inspect simulator behavior:

detb visualize --set visualization.execute=true
detb train-gui --set visualization.train_execute=true

Tune against the design sweep and adjust objective weights or thresholds directly from the CLI:

detb tune --set study=sweep
detb tune --set study=sweep --set objective.terrain_weight=0.5 --set objective.target_tgs=0.8

Select a DETB-owned task variant directly from Hydra when you want a different Isaac registry family:

detb train --set task=flat_walk_stability
detb evaluate --set task=flat_walk_stability
detb visualize --set task=flat_walk_stability --set visualization.execute=true

Select the simple-actuator robot/profile path by pairing the matching task and robot groups:

detb train --set task=flat_walk_simple_actuator --set robot=anymal_c_simple_actuator
detb evaluate --set task=flat_walk_simple_actuator --set robot=anymal_c_simple_actuator

Commands

Command	Purpose
detb train	Produce a manifest, training summary, reward curve artifacts, and either a synthetic checkpoint (mock) or copied real checkpoint/config snapshots (isaaclab).
detb evaluate	Produce episode metrics, aggregate metrics, overview plot, summary markdown, and Isaac run metadata when execution.backend=isaaclab.
detb sweep	Run staged design-point screening using the configured study points.
detb tune	Rank design points using objective weights and target thresholds.
detb sensor-eval	Compare configured sensor profiles under matched conditions.
detb terrain-eval	Run the configured terrain battery and compute TGS.
detb failure-eval	Sweep fault severity and detect the first critical threshold.
detb visualize	Launch the pinned Isaac Lab GUI playback path or emit a launch spec when visualization.execute=false.
detb train-gui	Launch the pinned Isaac Lab GUI training path or emit a launch spec when visualization.train_execute=false.
detb generate-requirements --source-dir ...	Generate a candidate requirement ledger from stored artifacts.

Artifact Contract

Each major run writes to outputs/<command>/<run_id>/ and includes:

• resolved_config.yaml
• run_manifest.json
• artifact_registry.json
• machine-readable CSV or JSON outputs
• reviewable markdown summary
• lightweight SVG plots where applicable

GUI commands also emit launch-spec JSON files so the exact Isaac Lab invocation is reviewable before execution.

Repository Layout

configs/      Hydra configs and pinned defaults
detb/         Core package, runtime, backends, and contracts
source/detb_lab/  External Isaac Lab extension package for custom tasks and robot configs
docs/         Bootstrap, architecture, research, and workflow notes
scripts/      Convenience launchers
tests/        Unit and integration tests
outputs/      Generated artifacts (gitignored)

Documentation

• docs/bootstrap.md
• docs/architecture.md
• docs/data-contracts.md
• docs/cli.md
• docs/modules.md
• docs/reproducibility.md
• docs/terrain-research.md
• docs/terrain-design.md

Roadmap

1. Keep the mock-backed pipeline stable as the contract baseline.
2. Expand DETB-owned task families and robot configs under source/detb_lab.
3. Use GUI launch specs and Isaac Lab playback/training to inspect progress visually.
4. Tune objective weights and target thresholds until studies produce useful outputs.
5. Complete terrain taxonomy and generator mapping from Isaac Lab best practices.
6. Expand the verified real Isaac Lab backend from smoke runs into broader study coverage and requirement generation workflows.