TECHNICAL TUTORIALS
Technical Tutorials
18 bilingual tutorials on learning implementation, robot simulation, navigation scenes, VLA, and engineering toolchains.
Repeated topics are consolidated into general explanations and synthetic examples. Personal information, private environment snapshots, and chat indexes are omitted. The original archive and historical artifact catalog are not published.
Official repository sources checked on 2026-10-09. References pin commits; documentation review and example code do not establish target GPU, robot, or complete simulator validation.
This collection references 28 official documents and code examples. Official source and checksum manifest (.json) →
Learning & Implementation3
Tensors, gradients, and RL implementation.
- PyTorch: Axes, Broadcasting, and Storage Layout
Define shape contracts and distinguish axis changes, reshaping, views, and copies.
- Autograd, Gradient Accumulation, and Training Recovery
Understand graph lifetime, inference modes, gradient accumulation, and checkpoint contracts.
- RL Implementation: Objectives, Masks, and Gradient Paths
Turn PPO/GAE, TD, centralized training, and representations into inspectable interfaces.
Simulation & Robot Modeling4
Assets, contact, timing, and humanoid control.
- Robot Assets: CAD, USD, URDF, and Inertial Validation
Validate units, composition layers, joint frames, and physical properties in stages.
- Minimal Experiments for Contact and Mecanum Wheels
Inspect collision proxies, roller axes, support envelopes, and wheel-speed conventions.
- Simulation Timing, NavMesh Updates, and Profiling
Separate physics, control, sensing, and rendering clocks and isolate startup costs.
- Humanoid Robots: Control Dimensions, Symmetry, and Curriculum Recovery
Separate model coordinates from policy actions and validate contacts, resets, and recovery.
VLA & Imitation Learning3
Action chunks, model serving, and paired experiments.
- VLA Execution: Action Chunks, Latency, and Feedback
Separate action representation, generation, consumption, and scheduling; measure observation age.
- OpenPI × LIBERO: Serving, Observations, and Action Contracts
Use pinned upstream code to check fields, image orientation, action queues, and a single request.
- Paired Direct/Agent Evaluation and Progress Verification
Fix episode identity and budgets, separating task success, tool completion, and ablations.
Python & GPU Toolchains3
Builds, timing, versions, and reproducible debugging.
- Python Engineering: pyproject, uv Locking, and Build Isolation
Separate runtime and build dependencies; diagnose installation through wheel compatibility and backend errors.
- GPU Software Stacks and Correct PyTorch/JAX Timing
Separate drivers, toolkits, and framework runtimes; synchronize asynchronous benchmarks.
- Reproducible Debugging: Git, Downloads, ABI, and Evidence States
Locate installation, network, extension, and runtime failures by layer and sanitize evidence.
Suggested practice routes
- Learning: tensor layout → autograd → RL masks and gradients → evaluation protocol.
- Robotics: assets → contact → simulation timing → ROS 2 → crowds and navigation evaluation.
- VLA: action chunks → OpenPI/LIBERO → paired evaluation → GPU timing and debugging.