Embodied Foundation Models at the Edge: A Survey of Deployment Constraints and Mitigation Strategies
This addresses the challenge of making AI models practical for real-time control in resource-constrained edge devices, though it is incremental as it surveys and organizes existing constraints rather than introducing new methods.
The paper tackles the problem of deploying foundation models in embodied edge systems by identifying key deployment constraints, such as memory bandwidth and compute latency, and proposes system-level co-design strategies to ensure reliable operation under strict size, weight, and power limits.
Deploying foundation models in embodied edge systems is fundamentally a systems problem, not just a problem of model compression. Real-time control must operate within strict size, weight, and power constraints, where memory traffic, compute latency, timing variability, and safety margins interact directly. The Deployment Gauntlet organizes these constraints into eight coupled barriers that determine whether embodied foundation models can run reliably in practice. Across representative edge workloads, autoregressive Vision-Language-Action policies are constrained primarily by memory bandwidth, whereas diffusion-based controllers are limited more by compute latency and sustained execution cost. Reliable deployment therefore depends on system-level co-design across memory, scheduling, communication, and model architecture, including decompositions that separate fast control from slower semantic reasoning.