Efficient sensing and autonomous decision-making in large-scale environments face challenges such as sensor noise, limited coverage, and dynamic environmental changes, often leading to suboptimal responses. This study proposes an AI-driven framework that integrates multi-sensor data fusion, adaptive learning, and utility-based decision-making to address these issues. The framework combines heterogeneous sensor data using weighted fusion, improving the accuracy and reliability of environmental state estimation. Adaptive learning mechanisms dynamically adjust the learning rate, optimizing system performance by reducing prediction errors and refining model parameters over time. The autonomous decision-making module selects optimal actions based on utility functions, ensuring timely and accurate decisions without human intervention. The system demonstrates significant improvements in sensor coverage efficiency and overall performance, effectively handling complex, data-intensive environments. This work highlights the framework's robustness and its capacity to deliver optimal decision-making and sensing capabilities, validating its applicability in real-world, dynamic scenarios. However, the study is based on simulation, and the results may not fully reflect real-world complexities or limitations. Future work should evaluate the framework's performance in practical, large-scale deployments and address potential scalability and real-time implementation challenges.