Bridging design and energy systems: An integrated framework for architectural agency in off-grid housing

Off-grid and autonomous housing is gaining global momentum as communities and households confront rising energy demand, accelerating climate risks and the search for more sustainable living models. Yet, despite rapid technical advances, research on self-sufficient dwellings remains fragmented, dominated by engineering-led modelling, and lacking integration with architectural, behavioural, and policy perspectives. This review systematically examines these dispersed bodies of work to clarify how technical, environmental, social and governance dimensions collectively shape autonomous housing. Using a PRISMA-based Systematic Quantitative Literature Review of 97 peer-reviewed studies (2009–2024) across Web of Science, Scopus, ScienceDirect and IEEE Xplore, the analysis identifies three defining patterns: (1) a technologically mature but engineering-centric evidence base; (2) limited incorporation of environmental, behavioural and socio-technical factors; and (3) minimal policy frameworks linking design decisions to regulatory and governance contexts. While off-grid technologies are well established, the architectural and human dimensions that underpin long-term performance, user adaptation and system resilience remain underdeveloped. Key challenges include the marginal treatment of architectural variables in modelling, limited behavioural and design-focused research, and poor translation across technical, social and policy domains. Aligned with the objectives of this review to map the field, identify cross-domain gaps, and establish an integrative, design-led framework, the synthesis foregrounds architectural agency as a mechanism connecting systems, space, behaviour and governance. This perspective advances a design-led approach for shaping resilient, human-centred pathways for autonomous housing. • Systematic review of 97 studies on off-grid and autonomous housing. • Modelling studies dominate but rarely integrate architectural variables. • Simplified load assumptions limit realism of system-sizing models. • Integrative framework linking modelling, design, behaviour and policy. • Design-led pathways for resilient autonomous housing development.