Speculative Policy Orchestration: A Latency-Resilient Framework for Cloud-Robotic Manipulation

2026-03-19 · Chanh Nguyen, Shutong Jin, Florian T. Pokorny, Erik Elmroth arxiv

Cloud robotics enables robots to offload high-dimensional motion planning and reasoning to remote servers. However, for continuous manipulation tasks requiring high-frequency control, network latency and jitter can severely destabilize the system, causing command starvation and unsafe physical execution. To address this, we propose Speculative Policy Orchestration (SPO), a latency-resilient cloud-edge framework. SPO utilizes a cloud-hosted world model to pre-compute and stream future kinematic waypoints to a local edge buffer, decoupling execution frequency from network round-trip time. To mitigate unsafe execution caused by predictive drift, the edge node employs an $ε$-tube verifier that strictly bounds kinematic execution errors. The framework is coupled with an Adaptive Horizon Scaling mechanism that dynamically expands or shrinks the speculative pre-fetch depth based on real-time tracking error. We evaluate SPO on continuous RLBench manipulation tasks under emulated network delays. Results show that even when deployed with learned models of modest accuracy, SPO reduces network-induced idle time by over 60% compared to blocking remote inference. Furthermore, SPO discards approximately 60% fewer cloud predictions than static caching baselines. Ultimately, SPO enables fluid, real-time cloud-robotic control while maintaining bounded physical safety.

📄 PDF Abstract BibTeX arXiv:2603.19418

Code (0)

등록된 구현이 없습니다.

Tasks

Motion Planning

Adaptive Utility driven Resource Orchestration for Resilient AI (AURORA-AI)

2026-06-25 · Rahul Umesh Mhapsekar, Ilias Cherkaoui, Lizy Abraham, Indrakshi Dey arxiv

Modern AI systems are increasingly deployed under non-stationary computational, demographic, and operational conditions in which static resource allocation strategies degrade both predictive performance and human-centric…

AdaptiveSD A Stability-Aware, Runtime-Adaptive Speculative Decoding Framework with Multi-Policy Orchestration for CPU-Constrained LLM Inference

2026-07-04 · Sadra Saremi arxiv

With the rise of small quantized GGUF-based language models and their increasing use for on-device inference tasks, we have seen the growing need for an approach capable of reliably delivering these models at scale even …

Reinforcement Learning

StreamServe: Adaptive Speculative Flows for Low-Latency Disaggregated LLM Serving

2026-02-11 · Satyam Kumar, Arpit Singh Gautam, Kailash Talreja, Saurabh Jha arxiv

Efficient LLM serving must balance throughput and latency across diverse, bursty workloads. We introduce StreamServe, a disaggregated prefill decode serving architecture that combines metric aware routing across compute …

Hybrid Orchestration of Edge AI and Microservices via Graph-based Self-Imitation Learning

2026-03-03 · Chen Yang, Jin Zheng, Yang Zhuolin, Lai Pan 외 arxiv

Modern edge AI applications increasingly rely on microservice architectures that integrate both AI services and conventional microservices into complex request chains with stringent latency requirements. Effectively orch…

Reinforcement Learning

Efficient Tree-Structured Deep Research with Adaptive Resource Allocation

2025-10-02 · Lunyiu Nie, Nedim Lipka, Ryan A. Rossi, Swarat Chaudhuri arxiv

Deep research agents, which synthesize information across diverse sources, are significantly constrained by the sequential nature of reasoning. This bottleneck results in high latency, poor runtime adaptability, and inef…