Explainable Collaborative Sensing for Critical Decision Systems
ID：147
View protection：Participant Only
Updated time：2026-07-27 12:57:52 Views：9
Online
Abstract
Abstract—The collaborative operation of multiple heterogeneous sensors to enable accurate and reliable situational awareness is found in critical decision systems such as autonomous vehicles, disaster response platforms, industrial automation, and smart healthcare. While collaborative sensing allows for better coverage, robustness and fault tolerance, it also creates problems with the understanding of the contribution of individual sensors to the decisions. In this paper, we provide a detailed review of Explainable Collaborative Sensing (ECS) and argue the need for embedding explainability in sensor fusion and decision-making instead of as a post-processing step. To improve the transparency, trust and reliability of decisions, an explainable collaborative sensing framework, which consists of six layers, is proposed to achieve explainable sensor fusion, trust aware reasoning and edge-cloud intelligence. The review also presents mathematical models for weighted sensor fusion, explanation quality assessment and dynamic trust updating and demonstrates the effectiveness of the proposed framework by employing metrics like decision accuracy, explanation quality, trust correlation, and inference latency. Finally, the challenges of current research, new opportunities, and directions for the future of creating transparent, trusted, and human-centric collaborative sensing systems with the goal of supporting safety-critical applications in the real world are discussed.
Keywords
Collaborative Sensing, Explainable Artificial Intelligence (XAI), Sensor Fusion, Trust Modelling, Edge–Cloud Computing, Critical Decision Systems, Human-Centric AI
Post comments