ICT is embedded and pervasive into our daily lives. The notion of Cyber Physical Systems (CPS) has emerged: embedded computational collaborating devices, capable of controlling physical elements and responding to humans. The Cross-layer modEl-based fRamework for multi-oBjective dEsign of Reconfigurable systems in unceRtain hybRid envirOnments (CERBERO) project aims at developing a design environment for CPS based of two pillars: a cross-layer model based approach to describe, optimize, and analyze the system and all its different views concurrently; an advanced adaptivity support based on a multi-layer autonomous engine. To overcome the limit of current tools, CERBERO provides: libraries of generic Key Performance Indicators for reconfigurable CPSs in hybrid/uncertain environments; novel formal and simulation-based methods; a continuous design environment guaranteeing early-stage analysis and optimization of functional and non-functional requirements, including energy, reliability and security. CERBERO effectiveness will be assessed in challenging and diverse scenarios, brought by industrial leaders: an embedded CPS with self-healing capabilities for planetary explorations (TASE-S&T), an ocean monitoring CPSoS (AS), and a Smart Travelling CPSoS for Electric Vehicle (TNO-CRF-S&T). CERBERO will automate multi-objective decisions to meet requirements and correct/optimized–by–construction designs. Interoperable components (i.e. DynAA by TNO, AOW by IBM, PREESM by INSA, PAPI-ARTICo3 by UPM, MDC by UniCA-UniSS) will be enhanced with additional features (as security, USI), mostly released as open-source to foster open innovation and a real path to standardisation, and integrated (IBM- AI) into a unique framework. Design speed up (one order of magnitude), increased performance (30% less energy) and reduced costs of deployment (by rapid prototyping and system in the loop incremental design) and maintenance (by runtime verification and adaptivity) of CPSoS are expected.