PulseAugur
EN
LIVE 12:43:45

New ML frameworks enhance vehicle localization in GPS-denied environments

By PulseAugur Editorial · [4 sources] ·

Researchers have developed two novel machine learning frameworks to improve vehicle localization, particularly in environments where GPS signals are unreliable. The first, PRML2, combines Kalman filtering with physics-regularized machine learning to enhance accuracy and generalization using onboard sensors. The second, EVC-Mamba, utilizes an evidential Mamba model to create a virtual velocity sensor for correcting IMU drift, offering uncertainty quantification and real-time deployment capabilities. Both approaches aim to provide robust and cost-effective localization solutions for autonomous systems. AI

IMPACT These advancements could lead to more reliable and cost-effective autonomous navigation systems, especially in challenging environments.

RANK_REASON Two research papers published on arXiv detailing new machine learning frameworks for vehicle localization.

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 4 sources. How we write summaries →

New ML frameworks enhance vehicle localization in GPS-denied environments

COVERAGE [4]

  1. arXiv cs.AI TIER_1 English(EN) · Abinav Kalyanasundaram, Karthikeyan Chandra Sekaran, Wolfgang Utschick, Michael Botsch ·

    Physics-Regularized Machine Learning for Proprioceptive Vehicle Localization Using Onboard Sensors

    arXiv:2607.05663v1 Announce Type: cross Abstract: Accurate and robust localization is essential for autonomous mobility systems in real-world environments. While fusing Inertial Measurement Unit (IMU) data with satellite-based correction signals provides precise vehicle pose esti…

  2. arXiv cs.LG TIER_1 English(EN) · Abinav Kalyanasundaram, Karthikeyan Chandra Sekaran, Wolfgang Utschick, Michael Botsch ·

    Uncertainty-Aware Velocity Correction for Proprioceptive Vehicle Localization using Evidential Mamba

    arXiv:2607.05669v1 Announce Type: cross Abstract: Reliable localization in GNSS-denied environments remains a fundamental challenge for intelligent vehicles, as inertial navigation systems accumulate unbounded drift without external correction. Existing approaches provide drift c…

  3. arXiv cs.LG TIER_1 English(EN) · Michael Botsch ·

    Uncertainty-Aware Velocity Correction for Proprioceptive Vehicle Localization using Evidential Mamba

    Reliable localization in GNSS-denied environments remains a fundamental challenge for intelligent vehicles, as inertial navigation systems accumulate unbounded drift without external correction. Existing approaches provide drift correction through dedicated infrastructure, expens…

  4. arXiv cs.LG TIER_1 English(EN) · Michael Botsch ·

    Physics-Regularized Machine Learning for Proprioceptive Vehicle Localization Using Onboard Sensors

    Accurate and robust localization is essential for autonomous mobility systems in real-world environments. While fusing Inertial Measurement Unit (IMU) data with satellite-based correction signals provides precise vehicle pose estimates, performance degrades substantially during o…

RELATED ENTITIES

RELATED TOPICS