Multi-Source Data Integration for Navigation in GPS-Denied Autonomous Driving Environments
Shaman Bhat, Ashwin Kavasseri
Abstract
Autonomous driving is making rapid advances, and the future of driverless cars is close to fruition. The biggest hurdle for autonomous driving currently is the reliability and dependability of navigation systems. Navigation systems are predominantly based on GPS signals and despite it being highly available there are scenarios where GPS is either not present or unavailable such as in tunnels, indoor environments, and urban areas with high signal interference. This paper proposes an adaptive decision-making algorithm that leverages multi source data source integration for navigation in GPS-denied environments. The algorithm enables seamless switching between the different data sources such as LTE or 5G for autonomous driving systems to maintain accurate navigation even when GPS signals are unavailable. Overall, this approach represents a sound methodology for developing navigation systems that can reliably support autonomous driving applications in real-world conditions.