ActiveScale: teaching robots active perception via model-data-hardware co-design
Haoyu_Xiong_ · x · 2026-09-19
A CMU + HKUST(GZ) team introduced ActiveScale, a framework advancing robotic active perception through coordinated model, data, and hardware design:
- Model: augments a VLA with historical video observations and explicit camera-pose supervision, using per-frame pose tokens and a lightweight prediction head to associate observations across viewpoints.
- Data: a scalable human–robot mid-training recipe using 1,000 hours of egocentric and robotic data, exploiting natural camera motion in human activity.
- Hardware: the Active Perception Mobile Manipulation Platform (AMP) supports single-operator teleoperation for scalable demonstration collection coordinating viewpoint changes with manipulation.
Experiments show improved success rates on active-perception tasks, with ablations confirming camera-pose-aware modeling and egocentric data. Code, dataset, and models are released.
Related event: ActiveScale: Co-Designed Framework Teaches Robots Active Perception(2 posts)→
More from Embodied
- AM-Bench: A Simulation Suite and Benchmark for Aerial Manipulation — GuanyaShi · 2026-09-19
- Northwestern's Phantom Twist drone spins its body 25 times a second to vanish mid-flight — churchkey · 2026-09-19
- Neuralink Helps Man With ALS, a Heartwarming BCI Case Going Viral — justin_hart · 2026-09-19
- Neuralink VOICE trial: patient speaks again in his own voice via brain implant — flyme2mars · 2026-09-19
- Viral 'robot unplugs itself and stands up' video is a real AGIBOT A3 — jonstephens85 · 2026-09-19
- Robotics AI can't copy the LLM playbook: 5% error rates kill the product — carlosdponx · 2026-09-19