eess.SPMar 6, 2026

KD-EKF: Knowledge-Distilled Adaptive Covariance EKF for Robust UWB/PDR Indoor Localization

Authors: Kyeonghyun Yoo, Wooyong Jung, Namkyung Yoon, Sangmin Lee, Sanghong Kim, Hwangnam Kim

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Ultra-wideband (UWB) indoor localization provides centimeter-level accuracy and low latency, but its measurement reliability degrades severely under Non-Line-of-Sight (NLOS) conditions, leading to meter-scale ranging errors and inconsistent uncertainty characteristics. Inertial Measurement Unit (IMU)-based Pedestrian Dead Reckoning (PDR) complements UWB by providing infrastructure-free motion estimation; however, its error accumulates nonlinearly over time due to bias and noise propagation. Fusion methods based on Extended Kalman Filters (EKF) and Particle Filters (PF) can improve average localization accuracy through probabilistic state estimation. However, these approaches typically rely on manually tuned measurement covariances. Such fixed or heuristically tuned parameters are hard to sustain across varying indoor layouts, NLOS ratios, and motion patterns, leading to limited robustness and poor generalization of measurement uncertainty modeling in heterogeneous environments. To address this limitation, this work proposes an adaptive measurement covariance scaling framework in which reliability cues are learned from historical UWB/PDR trajectories. A large teacher model is employed offline to generate temporally consistent next-position predictions from structured UWB/PDR sequences, and this behavior is distilled into a lightweight student model suitable for real-time deployment. The student model continuously regulates EKF measurement covariances based on prediction residuals, enabling environment-aware fusion without manual re-tuning. Experimental results demonstrate that the proposed KD-EKF framework significantly reduces localization error, suppresses error spikes during Line-of-Sight (LOS)/NLOS transitions, and mitigates long-term drift compared to fixed-parameter EKF, thereby improving measurement robustness across diverse indoor environments.

Figures & tables

Figure 1: Two-stage system overview of the proposed KD–EKF hybrid localization framework.
Figure 2: Floor plans and experimental trajectories used in the evaluation. Green areas denote LOS areas covered by UWB anchors, while orange areas indicate NLOS areas without anchor coverage. Experiment 1 represents a short-range environment with separated LOS/NLOS segments, whereas Experiment 2 corresponds to a closed rectangular corridor with repeated LOS/NLOS transitions.
Figure 3: Experimental hardware configuration. (a) Mobile tag device integrating a Raspberry Pi, a DWM1000 UWB module, and an IMU sensor for PDR preprocessing. (b) Fixed UWB anchor device based on the DWM1000 module deployed in LOS areas.
Figure 4: CDF of Absolute Trajectory Error (ATE)
ModelATE RMSE (m)
Teacher model (LLaMA-3.1-8B)0.3067
Student model (LLaMA-3.2-1B, KD)1.2311
Base student model(LLaMA-3.2-1B)1.8713
Table 1: Comparison of Position Regression Performance (ATE RMSE )
ModelLatency (ms)Throughput (Hz)
Teacher model (8B)283.13.5
Student model (1B, KD)129.27.7
Base student model (1B)131.57.6
Table 2: Inference Latency and Throughput
Figure 5: Comparison of estimated trajectories for Experiment 1 in a short-range environment with mixed LOS/NLOS conditions.
MethodMeanRMSEMax
UWB1.101.352.50
PDR0.851.051.40
EKF (Fixed)0.700.901.80
EKF (Base student model)0.460.61.45
KD-EKF (Proposed)0.350.450.85
Table 3: Experiment 1: ATE performance comparison (Unit: m)
Figure 6: Trajectory comparison for Experiment 2 in a closed rectangular corridor with repeated LOS/NLOS transitions.
MethodMeanRMSEMax
UWB0.961.426.43
PDR1.722.274.34
EKF (Fixed)1.091.344.05
EKF (Base student model)0.510.632.50
KD-EKF (Proposed)0.360.491.97
Table 4: Experiment 2: ATE performance comparison in a closed rectangular corridor environment (Unit: m)
ConfigurationKDStudentAdaptive Cov.RMSE (m)Max (m)
EKF (Fixed)–––1.344.05
EKF (Base student model)–✓✓0.632.50
KD-EKF (Proposed)✓✓✓0.491.97
Table 5: Ablation study on the contribution of individual components in the proposed KD-EKF framework.
Figure 7: Boxplot of ATE for the compared configurations.
More in this topic

Wireless Localization

Wireless localization uses radio signals and sensor fusion to estimate the positions and trajectories of people, devices, and transmitters.

See all 65 papers

Explore similar work

CardsList
Aug 10, 2026

WRAP: Wasserstein-Robust Adaptive Plug-in for Robot Localization

Robotic localization under changing sensing conditions can suffer from biased errors and miscalibrated covariances. We present WRAP, an adapter-agnostic Wasserstein-robust plug-in for nonlinear extended Kalman filter (EKF) and error-state Kalman filter (ESKF) stacks. A causal module supplies time-varying effective process and measurement statistics; a mean-preserving Wasserstein local update then computes least-favorable covariances and a robust gain without changing the propagation model, residual, or retraction. This separates mean adaptation from covariance robustification and uses distinct radii for propagation and sensing. On 18 UWB--IMU sequences held out from adapter training, adapter-only and WRAP reduce mean 3-D position RMSE by 19.8%19.8\% and 27.4%27.4\% relative to the nominal ESKF; an isotropic ablation reaches 19.5%19.5\%, linking the incremental gain to directional process-covariance redistribution. An in-sample GNSS--INS study shows that mean adaptation provides most of the accuracy gain, while DR improves consistency and mitigates over-tightened classical covariance estimates. The robust solve takes 0.05 ms for UWB and 2.92 ms for GNSS on a Jetson Orin Nano.

Minhyuk Jang, Astghik Hakobyan, Jungjin Lee +2

Department of Mechanical Science and Engineering, Grainger College of Engineering, University of Illinois Urbana-Champaign, Urbana, IL, USA · Center for Scientific Innovation and Education and the National Polytechnic University of Armenia, Yerevan, Armenia · Department of Electrical and Computer Engineering and Automation and Systems Research Institute, Seoul National University, Seoul, South Korea

Kalman FilteringRobot LocalizationInertial NavigationExtended Kalman FilteringRobust EstimationError-State Kalman Filtering
AskPDF
Mar 13, 2026

AoI-FusionNet: Age-Aware Tightly Coupled Fusion of UWB-IMU under Sparse Ranging Conditions

Accurate motion tracking of snow particles in avalanche events requires robust localization in global navigation satellite system (GNSS)-denied outdoor environments. This paper introduces AoI-FusionNet, a tightly coupled deep learning-based fusion framework that directly combines raw ultra-wideband (UWB) time-of-flight (ToF) measurements with inertial measurement unit (IMU) data for 3D trajectory estimation. Unlike loose-coupled pipelines based on intermediate trilateration, the proposed approach operates directly on heterogeneous sensor inputs, enabling localization even under insufficient ranging availability. The framework integrates an Age-of-Information (AoI)-aware decay module to reduce the influence of stale UWB ranging measurements and a learned attention gating mechanism that adaptively balances the contribution of UWB and IMU modalities based on measurement availability and temporal freshness. To evaluate robustness under limited data and measurement variability, we apply a diffusion-based residual augmentation strategy during training, producing an augmented variant termed AoI-FusionNet-DGAN. We assess the performance of the proposed model using offline post-processing of real-world measurement data collected in an alpine environment and benchmark it against UWB multilateration and loose-coupled fusion baselines. The results demonstrate that AoI-FusionNet substantially reduces mean and tail localization errors under intermittent and degraded sensing conditions.

Tehmina Bibi, Anselm Köhler, Jan-Thomas Fischer +1

TU Berlin, Germany · School of Electrical Engineering and Computer Science, TU Berlin, Germany · Austrian Research Centre for Forests, Austria

GNSS-Denied LocalizationMultimodal Sensor Fusion
AskPDF
Jul 17, 2026

Deployment-Ready UWB Localization for Industrial Ground Robots with Automatic Anchor Calibration and Terrain-Aware Fusion

Ultra-Wideband (UWB) ranging has become a viable option for industrial Autonomous Mobile Robot (AMR) localization due to improved accuracy and low cost. However, real-world deployments remain limited by two recurring challenges: calibrating static anchors can be time-consuming and error-prone, and integrating UWB with existing onboard sensors requires careful design to ensure robust and consistent pose estimation. Addressing these challenges, this paper presents an end-to-end pipeline that combines automatic anchor calibration with a generic multi-sensor estimator tailored to surface-bound vehicle motion. It targets existing AMR stacks in scenarios where robot pose priors are available for initialization. The calibration stage estimates anchor positions and range biases, while the localization stage fuses UWB with proprioceptive sensing in a bias-aware Extended Kalman Filter to improve consistency without extensive parameter tuning. Experiments on a commercial logistics AMR in a warehouse setting demonstrate accurate positioning indoors and across outdoor transitions, with improved consistency compared to an earlier estimator formulation. Evaluation on an independent forklift dataset further indicates transferability to other platforms. The method remains effective in test cases with limited line-of-sight and sparse anchor coverage. These results show that UWB localization can be deployed with substantially reduced manual effort while preserving the accuracy required for industrial AMRs. The collected warehouse dataset is made publicly available.

Alexander Raab, Giulio Delama, Roland Jung +1

AGILOX Services GmbH, Austria · Control of Networked Systems Group, University of Klagenfurt, Austria

Wireless LocalizationIndustrial RoboticsRobot Localization
AskPDF