INESC-ID projects approved across various FCT Calls

Recently, fourteen INESC-ID projects were approved by the Foundation for Science and Technology (FCT) in the Calls SR&TD/PEX, SAÚDE+, DEFESA+, CMU and UTA. Accounting for a total budget of €1,460,302.61 for INESC-ID, our institution will coordinate eleven of these projects, while in PIXEL, SHAPE-W and SAVER-U it will be a participating institution.

The approved projects and respective calls are as follows:

SAÚDE+ (HEALTH + SCIENCE: Call for Exploratory Research Projects 2025)

PIXEL (Play-based Inference and eXploration in AI-Powered Serious Games for Measuring Health-Related Quality of Life in Children)

PI: Rui Prada (AIPS)

The current methods to analyse health-related quality of life in children relies in questionnaires and information reported by adults, which can compromise the accuracy and continuity of the assessments. To address this, the PIXEL project aims to create an AI-powered serious game that relies on children’s naturally playful behaviour to create an assessment of their well-being in real time. This is expected to provide a more accessible, engaging and child-centred way to monitor their health while also supporting the early detection of psychosocial difficulties.

SR&TD / PEX (R&D Projects in All Scientific Domains 2025)

MEDIAR (MEDiate AI in Architecture)

PI: Inês Caetano (GI)

The growing use of Generative Artificial Intelligence (AI) in architecture raises concerns about its impact on creative thinking, design processes, authorship, and professional responsibility. MEDIAR aims to investigate how Generative AI can be responsibly integrated into architectural practice, not merely as a visualisation tool but as a partner in the creative design process. By addressing issues such as transparency, bias, and design control, the project seeks to develop guidelines that support ethical, inclusive, and responsible AI-assisted architectural design.

GoVerify (Mechanised Semantics and Verification for Go)

PI: Bernardo Toninho (IDSS)

The Go programming language is widely used in critical infrastructure and distributed systems, but it lacks formal foundations and comprehensive verification tools to guarantee program correctness. The GoVerify project aims to address this gap by developing the first mechanised verification framework for Go, including formal language foundations, automated verification tools, and session-typed communication channels. By providing machine-verified foundations, the project seeks to improve the reliability and security of Go programs, particularly in critical infrastructure, with all developments to be made available as open-source Rocq libraries.

DSIABI (Smart, Secure and Reliable Aggregators for Interoperability between Blockchains)

PI: Miguel Pupo Correia (DPSS)

The growing diversity of blockchain networks creates significant interoperability challenges, making it difficult to securely and efficiently transfer assets across different platforms. This project aims to develop and validate trustworthy, secure, and intelligent cross-chain aggregators that coordinate transactions across multiple blockchain networks while optimising costs, speed, and liquidity. By building on emerging interoperability standards and European regulatory frameworks, the project seeks to enable reliable and scalable interoperability within the evolving Web3 ecosystem.

Shape-W (Human-centered stochastic planning and explainability in warehouses)

PI: Sandra Gama (GI)

Warehouse operations face increasing challenges due to unpredictable demand, high service-level expectations, and labour restrictions, while existing optimisation methods have limited integration between stochastic optimization, human behaviour modelling, and explainable decision support. The SHAPE-W project aims to develop an integrated framework by using operational data to incorporate human performance into warehouse planning, which in turn seeks to enable more efficient, reliable, and transparent real-time decision-making.

Quantum Cryptography Overview: Primitives and Protocols Without Classical Difficulty

PI: Manuel Goulão (HPCAS)

The development of quantum computing challenges existing cryptographic assumptions and raises fundamental questions about the foundations of secure communication. This project aims to investigate the minimal computational assumptions required to build secure cryptographic systems in a quantum world. By clarifying the hierarchy and security properties of certain primitives, the project seeks to advance the theoretical foundations of quantum cryptography and support the development of secure systems resilient to quantum computing threats.

SAT-OCPA (SAT-based portfolio for object-centric process analysis)

PI: Alessandro Gianola (ARSR)

Modern business processes involve complex interactions with different types of interconnected data objects, making them difficult to model, analyse, and understand. SAT-OCPA aims to develop innovative methods for analysing object-centric processes, enabling more accurate process modelling and advanced analysis to improve operational efficiency and decision-making.

AUTO-POWER (Impact of autonomous electric vehicles on electric energy systems)

PI: Hugo Morais (SPS)

The increasing adoption of electric and autonomous vehicles poses new challenges for electricity grids, particularly as renewable energy generation becomes more widespread. The Auto-POWER project aims to investigate the impact of Autonomous Electric Vehicles (AEVs) on power systems with high penetration of renewable resources. It will use a multi-agent microservices platform to test and validate different use cases, contributing to more efficient, flexible, and sustainable energy systems.

SAVER-U (Shared Control Architecture for Virtual eFAST Robotic Ultrasound)

PI: Catarina Moreira (GI) / Ezequiel Zorzal (GI)

Emergency trauma assessment is often hindered by a shortage of skilled ultrasound operators, delaying diagnosis and increasing reliance on unnecessary CT scans. The SAVER-U project aims to address this challenge by developing an immersive robotic ultrasound system that combines Artificial Intelligence (AI), collaborative robotics (“cobot”), and Virtual Reality (VR). Through shared control and AI-assisted diagnosis, the system seeks to enable faster, more accessible, and reliable remote ultrasound examinations.

DEAR (Repairing programs using automatic reasoning)

PI: Vasco Manquinho (ARSR)

Program debugging is a time-consuming and costly task, particularly for inexperienced developers, while existing automated techniques often struggle to identify and fix errors effectively. This project aims to improve automated program debugging by combining Large Language Models (LLMs) with program analysis and automated reasoning techniques to identify bugs more efficiently and generate targeted code repairs. It will also develop tools that explain program behaviour, errors, and code modifications in natural language, supporting both developers and programming education.

UTA (Call for Exploratory Research Projects under the UT Austin Portugal Program)

EXPLOR-HODOR (EXPLanation and ObseRvation for HumanOid DOmestic RObots)

PI: Bruno Martins (HLT)

Current robotic Artificial Intelligence (AI) systems still struggle to perform complex, long-term tasks that require contextual understanding, adaptive decision-making, and natural collaboration with humans. Building on advances in Large Language Models (LLMs) and the previous FOMO-HODOR project, EXPLOR-HODOR aims to develop a computational framework for robotic cognition and interaction.

DEFENSE + SCIENCE: Call for Exploratory Projects 2025

ESCUDO (Establishment of Secure Communications Using Dual Orchestration: QKD and PQC)

PI: Manuel Goulão (HPCAS)

Current cryptographic systems are vulnerable to future attacks from powerful quantum computers, threatening the security of sensitive communications. This project aims to address this challenge by developing a hybrid key establishment system combining Quantum Key Distribution (QKD) and Post-Quantum Cryptography (PQC) to provide robust protection against threats. A proof-of-concept implementation will demonstrate its security, performance, and integration with existing network infrastructures, supporting the transition towards quantum-resistant communications, particularly in the defence and critical infrastructure sectors.

LLM2C2 (LLMs for Automated Configuration of Command and Control Systems)

PI: José Fragoso Santos (ARSR)

Modern military operations rely on multiple interconnected Command and Control (C2) systems, but configuring these systems is currently a manual and complex process that requires specialised expertise and can lead to delays and interoperability issues. The LLM2C2 aims to automate C2 system configuration by translating operational instructions written in natural language into configurations that comply with NATO interoperability standards, by combining Large Language Models (LLMs) with formal verification techniques.

CMU (Call for Exploratory Research Projects under the CMU Portugal Program 2025)

QSHIELD (Quantitative Program Analysis for Improving LLM Code Generation)

PI: José Fragoso Santos (ARSR)

Large Language Models (LLMs) frequently generate code containing subtle bugs and functional errors, making their use in software development potentially unreliable. This project aims to address this problem by developing a novel approach called Reinforcement Learning from Static Analysis Feedback (RLSAF), which integrates program analysis into the training and code generation processes of LLMs. Through this, QSHIELD seeks to improve the reliability, quality, and trustworthiness of AI-generated code, and, consequently, AI-assisted software development.