SS1: Advances in Maritime Tracking, Sensor Technologies, and Navigational Safety

The maritime domain is experiencing rapid digital transformation, with an increasing reliance on automated monitoring, advanced analytics to support safe, secure, and efficient operations. Maritime traffic contributes significantly to global trade, yet incidents such as collisions, groundings, and security breaches remain persistent challenges, as reported by EMSA and other maritime authorities. These occurrences highlight the limitations of current vessel tracking and monitoring infrastructures, which often depend on cooperative signaling, limited coverage, or static observation points. To address these challenges, emerging technologies in sensor networks, data fusion, and real-time analytics present opportunities to enhance situational awareness, improve decision-making, and hence maritime safety and security. By integrating diverse sensing modalities and leveraging robust methods for vessel tracking and detection, the maritime domain can move toward more resilient and adaptive operational capabilities, supporting both routine traffic management and risk mitigation in complex environments.

SS2: Advanced Nonlinear Filtering

This special session focuses on recent advances in nonlinear state estimation (filters, smoothers, and predictors) for both discrete and continuous time system models, with areas such as:

• Nonlinear and/or Non-Gaussian Estimation
  • Density-specific estimators (e.g., Gaussian, Student, transformed Gaussian, Rayleigh, Laplace) including nested, sigma-point, or stochastic integration-based design,
  • Global estimators such as point-mass, Gaussian mixture, or sequential Monte Carlo methods, a.k.a. particle filters, and Monte Carlo sampling methods,
  • Particle flow, homotopy-based, and progressive estimators,
  • Performance evaluation of estimation methods,
  • Joint and dual estimation for state and model parameters estimation.
• Robust Estimation
  • Robust techniques with partially unknown system models (system functions or noise statistics),
  • Robust techniques for measurements corrupted by outliers or unexpected model behaviours,
  • Linearly/nonlinearly constrained estimation.
• Efficient Estimator Design and Applications
  • State estimation in high-dimensional spaces,
  • Performance analysis of existing nonlinear filtering methods,
  • Applications of nonlinear state estimation methods.
• State Estimator Design aided by Machine Learning.

SS3: Applications of Stone Soup

The Stone Soup framework is a flexible, modular, open-source framework for developing and proving a wide variety of tracking and information-fusion-based solutions. Since its inception in 2017, it has aimed to provide an open, easy-to-deploy framework to develop and assess the performance of different types of track/state estimation algorithms. Now, through repeated application in many use-cases, implementation of a wide variety of algorithms, multiple releases, and contributions from the community, the framework has reached a mature point and is proving to be an essential tool in evaluation and characterisation of tracking and state estimation approaches. This special session highlights recent research contributions within the Stone Soup framework and emphasises the evaluation and comparison capabilities. Discussions in this session will typically draw upon Stone Soup’s evaluation features to include evaluation of a candidate approach against a number of other approaches in a variety of use cases.

SS4: The Convergence of AI and Information Fusion

Advances in artificial intelligence (AI) and machine learning are fundamentally reshaping the landscape of information fusion. From deep neural networks and probabilistic generative models to large language models (LLMs) and multi-modal architectures utilizing vision, language, and sensor data, AI now enables new forms of perception, representation, and contextual reasoning. Simultaneously, the information fusion community is expanding beyond classical sensor-level fusion paradigms to incorporate semantic, contextual, and human-centric information streams under uncertainty. This shift is driven by complex real-world requirements where decisions must combine physics-based sensors (radar, LiDAR, EO/IR), soft information (text, language), environmental priors, and situational context.

This special session focuses on the convergence of these fields, with emphasis on hybrid model-based and data-driven fusion architectures, uncertainty-aware deep learning, large models for semantic fusion, and joint inference frameworks combining probabilistic models with modern AI components. Particular interest is placed on challenges such as mathematical foundations for trustworthy AI fusion, robust uncertainty modeling in deep networks, environments. Application domains include autonomous vehicles, distributed sensing networks, unmanned systems, cyber-physical security, remote sensing, environmental monitoring, healthcare, and human–machine teaming.

Bringing together researchers from information fusion, AI, and statistical signal processing will foster an integrated research agenda. The session aims to identify foundational principles, discuss benchmark methodologies, and highlight open problems that will define next-generation multi-modal fusion systems powered by AI.

SS5: Aerospace Estimation and Fusion

Information fusion plays a vital role in many aerospace problems, dating back to the first practical application of the Kalman filter for translunar navigation in NASA’s Apollo program. Today, information fusion problems span the aerospace field, from space object tracking and space domain awareness, missile defense, navigation, satellite remote sensing, and aircraft uncertainty quantification. Problems in the aerospace field often must deal with nonlinearity, sparsity in observations, and tracking of multiple targets. Solutions of these problems must also work to overcome limitations to available computational resources (e.g., onboard applications) and network bandwidth. This special session features advances in information fusion motivated by current challenges in aerospace problems. Examples of advancement in methods contained in this special session include:

• Nonlinear estimation techniques, including Gaussian mixture, sequential Monte Carlo, and ensemble methods for navigation and tracking
• Solution methods for initial distributions, admissible sets, and track initialization
• Multi-object tracking frameworks, including random finite sets, joint probabilistic data association filtering multi-hypothesis tracking, and factor graph approaches
• Differential algebra, state transition tensors, and other polynomial approximation methods
• Directional statistics and coordinate transformations
• Multiple model and multi-fidelity approaches
• Simultaneous localization and mapping
• Sensor characterization and system identification

SS6: Special Session on Evaluation of Technologies for Uncertainty Reasoning

The ETUR Session is intended to report the contemporary analysis of topics related to the ISIF’s Evaluation of Technologies for Uncertainty Representation (ETUR) working group, which aims to bring together advances and developments in evaluation of uncertainty representation and reasoning for information fusion solutions. More information about the ISIF ETURWG can be found here: https://www.isif.org/evaluation-techniques-uncertainty-representation-working-group

The 2026 ETUR special session will focus on the concept of Information Fusion Theory and Experimentation for Decision-Making under Uncertainty, and its connections with uncertainty representation and reasoning within the Information Fusion context. Topics of discussion will include:

• Ontology-based evaluation of uncertainty
• Uncertainty considerations of Large Language Models (LLMs)
• Automated uncertainty evaluation
• Uncertainty provenance
• Explainability & interpretability
• Multi-modal fusion
• Decision making
• Assessment of High-Level Information Fusion Systems
• Use Case-based Uncertainty Evaluation and Experimentation
• Advances in AI for Uncertainty Evaluation
• Uncertainty Evaluation in AI systems

The discussion will not be limited to specific approaches and can cover a wide range of applications.

SS7: Special Session on Extended Object and Group Tracking

Traditional object tracking algorithms assume that the target object can be modeled as a single point without a spatial extent. However, there are many scenarios in which this assumption is not justified. For example, when the resolution of the sensor device is higher than the spatial extent of the object, a varying number of measurements can be received, originating from points on the entire surface or contour or from spatially distributed reflection centers. Furthermore, a collectively moving group of point objects can be seen as a single extended object because of the interdependency of the group members. This Special Session addresses fundamental techniques, recent developments, and future research directions in the field of extended object and group tracking.

SS8: LAFUSION

he aim of this special session is to extend the papers presented in the First Latin American Workshop on Information Fusion (LAFUSION 2025) - that focused on the latest research results on the Information Fusion in Latin America. The goal of this workshop was to create a community of Information Fusion researchers in Latin America that will be part of the FUSION community in the next years.

Information fusion is a multidisciplinary field that focuses on combining and integrating information from diverse sources to improve the accuracy, completeness, and reliability of the resulting information. It involves the process of merging data or knowledge from multiple sensors, databases, or information systems to generate a unified and coherent representation of the underlying reality.

The main goal of information fusion is to extract meaningful and actionable insights by leveraging the strengths of individual information sources while compensating for their limitations, uncertainties, or redundancies. It aims to provide a more comprehensive and accurate understanding of a given situation or phenomenon than what can be achieved by using individual sources in isolation.

Information fusion techniques typically involve various processes, including data preprocessing, feature extraction, data association, probabilistic modeling, decision-making, and knowledge representation. These processes may utilize methods from diverse disciplines such as statistics, signal processing, pattern recognition, artificial intelligence, machine learning, and cognitive science.

Applications of information fusion are widespread and can be found in fields such as surveillance and intelligence, remote sensing, robotics, autonomous systems, medical diagnosis, weather forecasting, transportation systems, and cybersecurity. By integrating and interpreting information from multiple sources, information fusion enables improved situational awareness, decision-making, and prediction capabilities, leading to enhanced performance, efficiency, and reliability in complex and uncertain environments. Several Latin American problems could be solved by Information Fusion. We are looking to form a Forum to debate the usage of Information Fusion to produce solutions for the challenges in the region.

SS9: Scalable Fusion for Autonomous Systems: Theory, Algorithms, Applications, and Trusted Architectures

Autonomous systems are rapidly evolving toward large-scale, distributed, and self-organizing architectures, operating under uncertainty, degraded communications, and adversarial conditions. These trends fundamentally challenge traditional fusion paradigms, which often assume centralized processing, static models, or trusted data pipelines. This special session focuses on scalable fusion methods for autonomous systems, emphasizing theoretical foundations, algorithmic advances, and architectural innovations that enable fusion to scale across agents, modalities, missions, and trust boundaries. Topics include distributed and decentralized fusion, algebraic and graph-based representations, uncertainty-aware learning, zero-trust and provenance-aware fusion, and explainable decision processes suitable for autonomous and human-machine teaming systems.

SS10: Context-Based Information Fusion

The goal of the proposed session is to discuss approaches to context-based information fusion (IF). It will cover the design and development of information fusion solutions integrating sensor data with contextual knowledge. With the advent of powerful LLMs, foundation models, and agent-based systems, context has recently emerged as the key factor for “initializing” complex IF architectures to the given situation, required for correctly performing their tasks.

Developing IF systems that incorporate contextual factors thus offers opportunities to ensure or improve the quality of fused outputs, provide solutions tailored to application requirements, and enhance responses to user queries. Challenges in context-based strategies include selecting appropriate representations, exploitations, and instantiations. Context may be represented as knowledge bases, ontologies, geographical maps, etc., forming a powerful tool to enhance adaptability and system performance. Example applications include context-aided tracking and classification, situational reasoning, and ontology building and updating.

The session will address both representation and exploitation mechanisms to ensure contextual knowledge is efficiently integrated into the fusion process, enabling adaptive mechanisms.

SS11: Information Fusion for Situation Understanding and Sense-Making

The exploitation of all relevant information originating from a growing mass of heterogeneous sources, both device-based (sensors, video, etc.) and human-generated (text, voice, etc.), is a key factor in producing a timely, comprehensive, and accurate description of a situation or phenomenon to make informed decisions. Even when exploiting multiple sources, most fusion systems are developed for combining just one type of data (e.g., positional data) to achieve a certain goal (e.g., accurate target tracking) without considering other relevant information (e.g., current situation status) from other abstraction levels.

The goal of seamlessly combining information from diverse sources, both human- and device- generated, exists only in a few narrowly specialized and limited areas. In other words, there is no unified, holistic solution to this problem.

Processes at different levels generally work on data and information of different natures. For example, low-level processes could deal with device-generated data (e.g., images, tracks, etc.), while high-level processes might exploit human-generated knowledge (e.g., text, ontologies, etc.). The overall objective is to enhance the sense-making of information collected from multiple heterogeneous sources and processes for improved situational awareness. This includes topics such as sense-making of patterns of behavior, global interactions, information quality, and integrating sources of data, information, and contextual knowledge.

The proposed special session will bring together researchers working on fusion techniques and algorithms often considered different and disjoint. The objective is to foster discussion and propose viable solutions to address challenging problems in relevant applications.

SS12: Knowing the Past of Information Fusion to Prepare Its Future
– Milestones Lessons Learned, and Emerging Paradigms

Information fusion has matured from early multisensor tracking and estimation concepts into a cor- nerstone of today’s autonomous, networked, and learning-enabled systems. Yet, as the field acceler- ates—driven by data-centric AI, distributed architectures, and contested environments—research risks losing sight of the hard-won lessons embedded in its own lineage: why certain formulations pre- vailed, which assumptions repeatedly failed in practice, and how foundational ideas re-emerge under new names and computational regimes.

This special session invites contributions that illustrate how historically informed awareness can strengthen future-oriented research in information fusion under the guiding theme Knowing the Past of Information Fusion for Preparing its Future. We seek papers that connect seminal concepts, archi- tectures, and benchmark problems to contemporary challenges such as trustworthy AI fusion, sen- sor/communication convergence, resilience in adversarial settings, and human-centered decision sup- port. Suitable submissions may include (but are not limited to): historical reconstructions of key breakthroughs; “lessons learned” from real deployments; comparative analyses of classical vs. mod- ern (learning-based) fusion; the evolution of uncertainty modeling and evaluation metrics; and for- ward-looking perspectives grounded in the discipline’s intellectual and institutional history.

By making the field’s origins explicit, the session aims to foster sharper problem formulations, better experimental practice, and more responsible innovation—so that the next generation of fusion sys- tems is not only more capable, but also more robust, interpretable, and aligned with human goals.