A TCN-BiLSTM and ANR-IEKF Hybrid Framework for Sustained Vehicle Positioning During GNSS Outages

Abstract

Highlights: What are the main findings? An Adaptive Noise-Regulated Iterated Extended Kalman Filter (ANR-IEKF) has been developed as a robust estimation kernel that dynamically adjusts noise characteristics in real time. This approach effectively addresses the uncertainties associated with deep learning predictions while ensuring numerical stability. An integration of the Temporal Convolutional Network and Bidirectional Long Short-Term Memory (TCN-BiLSTM) is developed to leverage the TCN’s capability to extract local temporal features from INS data while utilizing BiLSTM’s ability to model long-range, bidirectional trajectory dependencies. A hybrid ANR-IEKF + TCN-BiLSTM architecture is proposed. Within this framework, the TCN-BiLSTM generates precise pseudo-GNSS measurements from raw INS data, which are then adaptively fused by the ANR-IEKF. What are the implications of the main finding? The hybrid TCN-BiLSTM architecture establishes a transformative framework for compensating positioning errors by decoding complex nonlinear relationships between raw INS data and vehicle displacement. This method consistently outperforms traditional models, thereby enabling reliable navigation in environments with degraded GNSS signals. Comprehensive real-vehicle experiments conducted across various driving scenarios demonstrate that the integrated ANR-IEKF+TCN-BiLSTM algorithm exhibits superior accuracy and robustness during GNSS outages. This finding establishes it as a reliable solution to the inherent limitations of conventional GNSS/INS systems. The performance of integrated Global Navigation Satellite System and Inertial Navigation System (GNSS/INS) navigation often declines in complex urban environments due to frequent GNSS signal blockages. This poses a significant challenge for autonomous driving applications that require continuous and reliable positioning. To address this limitation, this paper presents a novel hybrid framework that combines a deep learning architecture with an adaptive Kalman Filter. At the core of this framework is a Temporal Convolutional Network and Bidirectional Long Short-Term Memory (TCN-BiLSTM) model, which generates accurate pseudo-GNSS measurements from raw INS data during GNSS outages. These measurements are then fused with the INS data stream using an Adaptive Noise-Regulated Iterated Extended Kalman Filter (ANR-IEKF), which enhances robustness by dynamically estimating and adjusting the process and observation noise statistics in real time. The proposed ANR-IEKF + TCN-BiLSTM framework was validated using a real-world vehicle dataset that encompasses both straight-line and turning scenarios. The results demonstrate its superior performance in positioning accuracy and robustness compared to several baseline models, thereby confirming its effectiveness as a reliable solution for maintaining high-precision navigation in GNSS-denied environments. Validated in 70 s GNSS outage environments, our approach enhances positioning accuracy by over 50% against strong deep learning baselines with errors reduced to roughly 3.4 m.