Tracking Large-scale Shared Bikes with Inertial Motion Learning in GNSS Blocked Environments

2026-05-08

AI Summary

This work addresses the challenge of accurate positioning for shared bicycles in GNSS-denied environments such as urban canyons, where low-cost inertial sensors suffer from cumulative drift and insufficient robustness. The authors propose a novel approach that integrates bicycle kinematic constraints with an uncertainty-aware Mixture-of-Experts (MoE) model. By employing a gating mechanism to dynamically weight multiple expert modules, the method enhances multi-task learning performance. Notably, it leverages, for the first time, the intrinsic relationship between cadence and acceleration to dynamically calibrate wheel speed. Evaluated on real-world ride data from DiDi, the proposed method achieves at least a 12% improvement in positioning accuracy over baseline approaches, with the 95th percentile wheel speed error below 0.5 m/s, significantly enhancing both accuracy and robustness of inertial navigation.

Abstract

Although Global Navigation Satellite Systems (GNSS) provide a general solution for bike tracking outdoors, there still exist complex riding environments where only inertial navigation systems work, such as urban canyons. Despite decades of research, localization using only low-cost inertial sensors still faces challenges such as cumulative drifts and poor robustness caused by filtering methods. Furthermore, sensors such as visual and LiDAR could provide reliable measurements, but they are not suitable for large-scale deployment. In this paper, we propose an inertial tracking framework that integrates bicycle mechanical constraints with a mixture-of-experts model. Specifically, we leverage multiple expert modules to capture shared representations and weight them through the gating mechanism, thus improving multi-task learning performance and enabling uncertainty-aware trajectory estimation. Furthermore, based on the mechanical transmission between the pedal and the rear wheel of a bike, we explore the intrinsic relationship between the rider's periodic pedalling behaviors and acceleration variations, and convert such patterns into bike's wheel speed for dynamic calibration. Experiments with real-world riding data from shared bikes of the DiDi ride-hailing platform demonstrate that our system improves the accuracy of baselines by at least 12%, with wheel speed errors below 0.5 m/s at 95-percentile.

Problem

• GNSS-denied localization
• inertial navigation
• shared bike tracking
• cumulative drift
• large-scale deployment

Innovation

• inertial navigation
• mixture-of-experts
• bicycle mechanical constraints
• dynamic calibration
• GNSS-denied localization