A Generative AI-Driven Architecture for Whole-Home Intelligent Services: Edge Models, Memory Management, and Household Applications
Organizations: Nanjing Normal University
Abstract
Generative artificial intelligence is entering the home. Cloud-based assistants face governance limitations in data transmission, continuous availability, and the reliability of physical actions, which are especially salient in households where devices remain online, data accumulate over time, and commands trigger real-world actions. Using the home local server as a unifying organizing axis, this article develops a design-oriented review framework. Sources from arXiv, CNKI, ACL Anthology, and official documentation were searched through November 2026 and coded into five evidence categories: reviews, system prototypes, benchmarks, HCI and eldercare studies, and official documentation. System designs were mapped onto shared architectural modules, identifying nine recurrent responsibilities of the home local server. Generative models support perceptual interpretation, intent understanding, and candidate planning, while deterministic validation, knowledge-graph constraints, and authorization leases confine candidate actions to an inspectable space. Home control, emotional companionship, and intelligent guarding can be compared within a common framework: they share state and memory but differ in latency, false positive and negative rates, proactivity, and permissions. Existing benchmarks rely on virtual environments and cannot substitute for real-deployment evaluation. Local deployment is not inherently safer, and technical continuity alone does not establish the social or medical effects of companionship and guarding. The article proposes a five-dimensional evaluation framework covering architectural capability, system resources, privacy, safety and reliability, and human experience, together with six design principles and a future research agenda for local-first intelligent home systems.
Figures & tables
| Source | Evidence Level | Primary Architectural Responsibilities | Environment |
|---|---|---|---|
| HomeLLaMA | System prototype | Model serving, memory, cloud-edge collaboration | Local + optional remote |
| Harmony | System prototype | Model serving, Agent orchestration | Local virtual environment |
| On-Device LLMs for HA | System prototype | Model serving, safe execution | CPU edge hardware |
| AdaHome | System prototype | Agent orchestration, preference adaptation | Local evaluation |
| Trail | System prototype | Typed decisions, abstention | Fully local benchmark |
| DESAMO | Prototype device | Audio perception, event recognition | Embedded device |
| Module | Core Responsibility | Primary Risk | Recommended Metrics |
|---|---|---|---|
| Device Access & Capability Registry | Multi-protocol device management and capability registration | Capability misdescription, registry hijacking | Access scale, registration failure rate |
| Perception & Event Extraction | Signal and multimodal input to structured events | Event misextraction, missed detection | Event accuracy, miss rate |
| Home State | Registry, topology, and current state maintenance | State inconsistency, staleness | State consistency, update latency |
| Memory Management | Layered short-term context and long-term memory maintenance | Memory contamination, sensitive accumulation | Retrieval accuracy, pruning coverage |
| Model Serving | Local model running, quantization, scheduling | Quantization degradation, resource contention | TTFT, memory footprint |
| Agent Orchestration | Task decomposition, tool calls, result integration | Tool misuse, runaway loops | Task completion, tool compliance rate |