Q-NAV: AI-Assisted Quantum Navigation For GPS-Denied Autonomous Systems

Systems for global navigation such as GPS are of great importance in enabling autonomous vehicles, drones, and robotic systems to navigate. Yet because these systems rely on external satellite signals, autonomous systems can be susceptible to signal obstruction, jamming, spoofing, and operation in areas where GPS is unavailable. Although conventional Inertial Navigation Systems (INS) can offer navigation independent of external signals, they suffer from sensor noise accumulation and drift, which leads to a gradual increase in localisation errors over time. This study presents a hybrid navigation approach that combines quantum inertial sensing with an Artificial Intelligence (AI)-based error-correction and sensor-fusion method to enhance localisation accuracy in GPS-denied situations. The quantum inertial sensors, based on atominterferometric principles, provide highly sensitive measurements of both acceleration and rotation, and the AI component is intended to learn to compensate for sensor noise, bias, and accumulated navigation drift. The framework will be tested through a simulation study that compares conventional inertial navigation, quantumassisted navigation, and AI-assisted quantum inertial navigation under different noise levels and in GPS-denied conditions. Performance will be evaluated using position error, velocity error, orientation error, and navigation drift as the main indicators. The objective of the proposed method is to show how combining emerging quantum sensing with intelligent computational methods can enable more reliable, highprecision autonomous navigation. This research has potential applications in autonomous drones, intelligent transportation, robotics, and other safety-critical systems that need reliable positioning without relying continuously on satellite navigation.