Event
ActECC: An Action Chunk Scheduling System for Edge-Cloud Collaborative VLA Inference
1 reports · 1 sources
Zihao Zheng and other authors propose ActECC, a system for edge-cloud collaborative VLA inference that uses kinematics-guided usable-range assessment and latency-aware replanning scheduling to address the difficulties of determining the effective range of action chunks and scheduling replanning requests. The authors state that experiments show that, compared with existing edge-cloud collaboration schemes, ActECC reduces per-action latency by up to 27.4%, cuts cloud invocations by up to 45.1%, and lowers the average execution error.
Report timeline
- ActECC: An Action Chunk Temporal Scheduling System for Edge-Cloud Collaborative VLA Inference
The authors propose ActECC, which addresses the difficulty of determining the valid range of action chunks and scheduling replanning requests in edge-cloud collaborative VLA inference through kinematics-guided usable range assessment and delay-aware replanning scheduling. Experiments show it reduces per-action latency by up to 27.4%, cuts cloud calls by up to 45.1%, and lowers average execution error.
arXiv cs.RO ·