Fulvio Mastrogiovanni

Embodied agents interacting with the same physical process may maintain heterogeneous, asynchronous, and observer-relative representations. Rather than assuming that such representations should always be globally aligned, we investigate which distinctions among them must actually be resolved for coherent interaction. We formalize this question through relational sufficiency: an interaction-specific relational-evidence map defines the representational ambiguity left unresolved by comparison, and sufficiency holds when each residual ambiguity fiber lies within an interaction-equivalence class. When this condition fails, additional sensorimotor anchors must separate precisely the ambiguities that can change the interaction-relevant outcome. For smooth systems, we derive a local first-order lower bound on the number of independent scalar anchors required, given by the dimension of directions that are invisible to relational evidence but visible to the interaction-relevant map. We instantiate the theory in two variants of an asynchronous hand-off scenario that share the same observer agents, representations, transports, and relational evidence. Relative synchronization requires no additional anchor, whereas hand-off at an absolute time requires exactly one. The analysis therefore provides a task-relative stopping condition for representational alignment: residual disagreement need not be eliminated when it is invariant with respect to the interaction.

Human-aware robot navigation, which enables robots to move through human-populated environments safely, comfortably, and trust-worthily, remains a core challenge in robotics despite advances in sensing, perception, and motion control throughout the past decade. The field appears to be fragmented by the absence of universally accepted navigation benchmarks. This hinders a direct comparison of methods, also in view of the inherent differences characterizing robot platforms, and makes the identification of paths to progress more difficult.Such a fragmentation becomes apparent when considering the heterogeneity of the employed principles to implement social-aware navigation approaches, that is, from classical model-based techniques and Reinforcement Learning, to emerging approaches leveraging Large Language Models and Vision-Language (-Action) Models. Motivated by these critical research gaps, this paper presents a comprehensive review of social-aware navigation research activities, analyzing methods and approaches set forth in the literature in the past 25 years. Our analysis identifies important emerging trends. Among them: (i) a shift from classical, reactive controllers toward data-driven and learning-based methods, particularly Deep Reinforcement Learning; (ii) the use of multi-modal approaches to sensor fusion, combining vision and depth information with inertial data; (iii) the increasing adoption of foundation models; (iv) the evolution from individual to group-aware navigation models able to respect social normative prescription; and (v) the design of hybrid approaches architectures that combine optimization-based safety guarantees with learned social policies. However, key challenges persist, including for example such diverse aspects like the generalization across diverse cultural contexts, real-time performance in dense crowds, and a critical lack of standardized metrics to assess performance. In this paper, we provide a systematic taxonomy of methodological approaches, pinpoint persistent research gaps, and outline promising future directions, whereas the integration of Generative Artificial Intelligence techniques and multi-modal learning stand out as key frontiers for enabling a natural -- yet implicit -- human-robot communication in navigation tasks. We hope that this review could serve as a foundational resource to align the field, and -- in perspective -- to develop robots able to navigate effectively while adhering to complex social norms.

Background/Objectives: Falls represent a major health concern for stroke survivors, necessitating effective risk assessment tools. This study proposes the Instrumented Fall Risk Assessment (IFRA) scale, a novel screening tool derived from Instrumented Timed Up and Go (ITUG) test data, designed to capture mobility measures often missed by traditional scales. Methods: We employed a two-step machine learning approach to develop the IFRA scale: first, identifying predictive mobility features from ITUG data and, second, creating a stratification strategy to classify patients into low-, medium-, or high-fall-risk categories. This study included 142 participants, who were divided into training (including synthetic cases), validation, and testing sets (comprising 22 non-fallers and 10 fallers). IFRA’s performance was compared against traditional clinical scales (e.g., standard TUG and Mini-BESTest) using Fisher’s Exact test. Results: Machine learning analysis identified specific features as key predictors, namely vertical and medio-lateral acceleration, and angular velocity during walking and sit-to-walk transitions. IFRA demonstrated a statistically significant association with fall status (Fisher’s Exact test p = 0.004) and was the only scale to assign more than half of the actual fallers to the high-risk category, outperforming the comparative clinical scales in this dataset. Conclusions: This proof-of-concept study demonstrates IFRA’s potential as an automated, complementary approach for fall risk stratification in post-stroke patients. While IFRA shows promising discriminative capability, particularly for identifying high-risk individuals, these preliminary findings require validation in larger cohorts before clinical implementation.