PDRNN: Modular AI-Assisted Pedestrian Dead Reckoning System
A recent study introduces PDRNN, an innovative modular hybrid AI-driven pedestrian dead reckoning (PDR) system that employs a recurrent neural network (RNN) to integrate multimodal sensor information from loosely connected radio and inertial signal sources. Conventional PDR techniques often face challenges due to differences in sensor sampling rates, unreliable data transmission, and dynamic movements characterized by high acceleration and rapidly changing orientations. PDRNN overcomes these issues by predicting asynchronous sensor data streams from various estimation methods along reference paths. Each element of the system is treated as a distinct ensemble of machine learning (ML) models. This research can be found on arXiv with the identifier 2605.15252.
Key facts
- PDRNN is a modular hybrid AI-assisted pedestrian dead reckoning system.
- It uses a recurrent neural network (RNN) architecture.
- The system fuses multimodal sensor data from loosely coupled radio and inertial signal streams.
- It addresses challenges in sampling rate discrepancies and unreliable transmission.
- PDRNN forecasts asynchronous sensor data streams along reference trajectories.
- Each component is handled as an independent ensemble of ML models.
- The paper is published on arXiv with identifier 2605.15252.
Entities
Institutions
- arXiv