Learning to Localise Automated Vehicles in Challenging Environments using Inertial Navigation Systems (INS)

TL;DR

This paper introduces an Artificial Neural Network-based algorithm to enhance INS/GNSS integrated navigation accuracy during GNSS signal outages in complex urban driving scenarios, demonstrating significant improvements in displacement and orientation estimation.

Contribution

The paper presents a novel Input Delay Neural Network approach that learns error drift patterns in INS, outperforming traditional neural networks in computational efficiency and accuracy during challenging conditions.

Findings

Up to 89.55% improvement in displacement estimation.

Up to 93.35% improvement in orientation rate estimation.

Effective in complex driving scenarios like sharp turns and quick accelerations.

Abstract

An algorithm based on Artificial Neural Networks is proposed in this paper to improve the accuracy of Inertial Navigation System (INS)/ Global Navigation Satellite System (GNSS) integrated navigation during the absence of GNSS signals. The INS which can be used to continuously position autonomous vehicles during GNSS signal losses around urban canyons, bridges, tunnels and trees, suffers from unbounded exponential error drifts cascaded over time during the integration of the gyroscope and double integration of the accelerometer to displacement. More so, the error drift is characterised by a pattern dependent on time. The Input Delay Neural Network (IDNN) has the ability to learn the error drift over time [1] and possesses the quality of being more computationally efficient than the Recurrent Neural Network (RNN), Long Short-Term Memory, and the Gated Recurrent Unit Network. Furthermore published literatures focus on travel routes which do not take complex driving scenarios into consideration, we therefore investigate in this paper the performance of the proposed algorithm on challenging scenarios, such as hard brake, roundabouts, sharp cornering, successive left and right turns and quick changes in vehicular acceleration across numerous test sequences. The results obtained show that the Neural Network-based approaches are able to provide up to 89.55 % improvement on the INS displacement estimation and 93.35 % on the INS orientation rate estimation.