Axis Weekly
This week was about trust and transfer: making community data cleaner, generated tasks broader, and trained policies more robust as they move from simulation to real robots.
Key updates:
- Data quality: We completed a suspicious-user audit script to detect abnormal collection behavior using user statistics and replay/verify failure reasons.
- Webapp and simulation: We improved key gripper and asset interactions, including penetration, heavy-object grasping, and IK flexibility.
- Recover-from-failure: We tested Failure Task 892 and collected 300 failure initial states, with the next round moving to repaired and more randomized tasks.
- TaskGen: Articulated-object generation is now merged into the automatic generation pipeline, covering cabinets, dishwashers, drawers, and existing randomization workflows.
- Model and real-world stack: We completed the first round of fine-tuning, evaluation, and benchmarking, merged the π0.5 evaluation pipeline into the real-world stack, and started bringing a new embodiment into the loop.
A closer look at this week’s progress 🧵
Axis Weekly
This week, we focused on making the robotics data loop more measurable and reproducible: separating real user signals from bot traffic, expanding TaskGen into articulated-object tasks, and turning data-to-model workflows into repeatable services.
Key updates:
- Data quality: Task 805’s high failure rate was driven by bots, not real players.
- TaskGen: Codebase delivered for an upcoming update that will support end-to-end generation of articulated-object tasks from prompts.
- Simulation and data infra: Asset bugs fixed, and the automated recover-from-failure pipeline is nearing full deployment.
- Model training: Achieved a ~40% success rate in cross-simulation evaluation (IsaacLab to MuJoCo).
- Sim-to-real: Updated the domain randomization roadmap to heavily boost physical parameter diversity.
A closer look at this week’s progress 🧵