FormerReckoning: Physics Inspired Transformer for Accurate Inertial Navigation
Open Access
- 4 December 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 2066-2071
Abstract
Although modern localization methods have achieved remarkable accuracy with various sensors, there are still some circumstances where only proprioceptive sensing works (Inertial Navigation). However, localization and navigation using only IMU sensors (costing less than $1000) still face significant challenges such as low accuracy and large cumulative errors when using traditional filter methods. Furthermore, AI-based approaches, while promising, often yield unpredictable and unreliable outputs. This paper proposes FormerReckoning, an inertial localization estimation framework for wheeled robotics that incorporates physical prompts into a Transformer framework to enhance translation estimation accuracy. Our tests show that FormerReckoning not only reduces mean translation errors to 0.72% but also surpasses all baseline models in performance, demonstrating its potential to provide reliable and precise localization in a cost-effective manner.Keywords
This publication has 24 references indexed in Scilit:
- A platypus-inspired electro-mechanosensory finger for remote control and tactile sensingNano Energy, 2023
- Four-dimensional indoor visible light positioning: A deep-learning-based perspectiveJournal of the Franklin Institute, 2023
- A Transformer-Based Feature Segmentation and Region Alignment Method for UAV-View Geo-LocalizationIEEE Transactions on Circuits and Systems for Video Technology, 2021
- Adaptive Hybrid Model-Enabled Sensing System (HMSS) for Mobile Fine-Grained Air Pollution EstimationIEEE Transactions on Mobile Computing, 2020
- Robust IMU/GPS/VO Integration for Vehicle Navigation in GNSS Degraded Urban AreasIEEE Sensors Journal, 2020
- AI-IMU Dead-ReckoningIEEE Transactions on Intelligent Vehicles, 2020
- Analysis of State Estimation Drift on a MAV Using PX4 Autopilot and MEMS IMU During Dead-reckoning2020
- Neural Lander: Stable Drone Landing Control Using Learned Dynamics2019
- Neural network for black-box fusion of underwater robot localization under unmodeled noiseRobotics and Autonomous Systems, 2018
- Recent advances in convolutional neural networksPattern Recognition, 2018