UNRIO: Uncertainty-Aware Velocity Learning for Radar-Inertial Odometry

Jui-Te Huang; Tianshu Huang; Anthony Rowe; Michael Kaess

cs.ROApr 15, 2026

UNRIO: Uncertainty-Aware Velocity Learning for Radar-Inertial Odometry

Authors: Jui-Te Huang, Tianshu Huang, Anthony Rowe, Michael Kaess

Organizations: Carnegie Mellon University, Pittsburgh, PA 15213, USA

Abstract

mmWave radars are robust to darkness and occlusions such as dust and smoke, and can directly constrain ego-velocity from a single frame via Doppler measurements, making them attractive sensors for odometry in visually denied conditions. However, almost all existing radar-inertial odometry systems rely on lossy radar point clouds that are highly sparse, generally concentrated in a narrow angular band, and aliased at high speeds. We propose to instead estimate ego-velocity directly from unfiltered mmWave I/Q signals. Taking advantage of a foundation model for 4D radar spectrum, we develop a system that integrates uncertainty-aware velocity predictions with IMU measurements using an uncertainty-weighted sliding-window pose graph to accurately compute odometry even when provided radar data with aliasing or an unfavorable field of view. Evaluated on public benchmark datasets, our system, UNRIO, attains the lowest relative pose error on the majority of sequences across held-out environments, despite differing chirp configurations, motion patterns, and platforms with its strongest gains coming where point clouds fail most: lateral motion in IQ1M and heavily aliased Doppler in ColoRadar.

Figures & tables

Fig. 1: Our radar-inertial odometry system uses raw mmWave radar spectrum as input. Our system predicts velocity and its uncertainty from the radar spectrum using a Transformer neural network trained on a large dataset with diverse motion patterns. ColoRadar longboard ride at 4.6 m/s, past the waveform’s $\pm 2.50$ m/s unambiguous interval: the static-scene return folds across the Doppler axis.

Panel of Figure 2 — Fig. 2: Two ways radar point clouds lose information. Top: at $1.45$ m/s, the platform outruns the $\pm 1.22$ m/s unambiguous interval, so the return folds across the Doppler axis. Bottom: green marks the $1.8\%$ of bins CFAR keeps, which are all the body-velocity optimization sees.

Trace	Dist (m)	APE					RPE
		PC [ 6 ]	EKF RIO [ 2 ]	GRT [ 16 ]	UNRIO Doppler	UNRIO velocity	PC [ 6 ]	EKF RIO [ 2 ]	GRT [ 16 ]	UNRIO Doppler	UNRIO velocity
cfa.1.fwd	64	0.289	0.685	0.758	0.316	0.267	0.408	0.886	0.934	0.212	0.097
cfa.3.fwd	65	0.214	0.251	0.626	0.211	0.100	0.444	0.543	0.611	0.207	0.151
cfa.a.fwd (1)	112	0.281	1.626	0.622	0.237	0.221	0.589	0.950	0.458	0.199	0.181
cfa.a.fwd (2)	127	0.148	5.597	0.998	0.903	0.279	0.482	1.258	0.757	0.460	0.196
morrison.1.fwd (1)	199	3.193	3.930	1.433	3.917	1.680	0.449	1.074	0.450	0.212	0.148

TABLE I: APE / RPE RMSE (m) — IQ1M sequences with forward/lateral motion. Color background indicate best , 2nd best performance.

Fig. 3: Qualitative trajectory comparison on six representative sequences. Ground truth is shown as a dashed black line; estimated trajectories from PC, GRT, Doppler, and Velocity are shown as solid colored lines, aligned to the ground truth via SE(3) Umeyama alignment. Insets highlight regions of interest where method differences are most pronounced. Lateral-motion sequences demonstrate the clear advantage of learning-based methods over the classical DSP point-cloud baseline.

Trace	Dist (m)	APE				RPE
		PC [ 6 ]	EKF RIO [ 2 ]	UNRIO Doppler	UNRIO velocity	PC [ 6 ]	EKF RIO [ 2 ]	UNRIO Doppler	UNRIO velocity
12_21_2020_arpg_lab_run4	78	0.400	0.468	1.368	0.780	0.908	0.871	1.400	0.809
12_21_2020_ec_hallways_run4	93	0.464	0.978	1.038	0.867	1.234	1.349	0.449	0.345
2_22_2021_longboard_run6	846	102.078	528.885	4.835	1.260	36.308	11.912	0.525	0.377
2_22_2021_longboard_run7	1201	146.366	6181.662	9.317	6.452	33.939	11.321	0.754	0.656
2_23_2021_edgar_army_run5	103	0.399	1.977	0.515	0.338	1.211	1.919	0.501	0.336

TABLE II: APE / RPE RMSE (m). — ColoRadar test sequences evaluation. Color background indicate best , 2nd best performance.

Fig. 4: ColoRadar trajectories vs. ground truth (dashed) The radar is Doppler-unambiguous to

\pm 2.50

m/s; the longboard rides run at a median

4.7

and

5.5

m/s, aliasing

80\%

of frames. The point-cloud baseline reads Doppler directly and underestimate the velocity, while both learned models stay bounded.

UNRIO	Ablated	ColoRadar	IQ1M
Covariance, velocity branch
per-frame $\Sigma$	$\bar{\Sigma}_{\text{net}}$	5 / 5	22 / 1
per-frame $\Sigma$	$\bar{\Sigma}_{\text{emp}}$	7 / 3	18 / 5
Covariance, Doppler branch
per-frame $\Sigma$	$\bar{\Sigma}_{\text{net}}$	1 / 9	10 / 13
per-frame $\Sigma$	$\bar{\Sigma}_{\text{emp}}$	1 / 9	14 / 9

TABLE III: Uncertainty ablation: sequences won on RPE (10 m).

			fp32		bf16
Dataset	FFT	Stage	Doppler	Vel	Doppler	Vel
ColoRadar	8.98	Network	15.85	15.67	5.78	5.58
		Total	25.41	25.22	15.34	15.14
IQ1M	9.19	Network	15.89	15.69	6.17	5.59
		Total	25.65	25.46	15.94	15.36

TABLE IV: End-to-end pipeline (FFT + network + PGO) runtime per frame in ms (RTX 4090, Intel i9-13900K, batch 1).

Radar-Inertial Odometry

Radar-inertial odometry fuses radar measurements and inertial sensing to estimate a robot’s motion when other sensors are unreliable.

See all 18 papers

Geometrically-Constrained Radar-Inertial Odometry via Continuous Point-Pose Uncertainty Modeling
Apr 3, 2026Wooseong Yang, Dongjae Lee, Minwoo Jung +1Simultaneous Localization and Mapping Inertial Odometry
Dept. of Mechanical Engineering, SNU, Seoul, S. Korea
Dense Soft Weighting for Radar Ego-Velocity Estimation
Jul 29, 2026Atar Babgei, Chenyu Zhao, Michael Breza +1Recursive Least Squares Millimeter-Wave Radar
Imperial College London
Pushing Radar Odometry Beyond the Pavement: Current Capabilities and Challenges
Apr 27, 2026Shaunak Kolhe, Peng Jiang, Maggie Wigness +4Mobile Robotics Radar-Inertial Odometry
Texas A&M University · DEVCOM Army Research Laboratory

Apr 3, 2026J/K move · Enter open · S save

Geometrically-Constrained Radar-Inertial Odometry via Continuous Point-Pose Uncertainty Modeling

Wooseong Yang, Dongjae Lee, Minwoo Jung, Ayoung Kim

Dept. of Mechanical Engineering, SNU, Seoul, S. Korea

Simultaneous Localization and Mapping Inertial Odometry Uncertainty Quantification Radar-Inertial Odometry Robot State Estimation

Open paper AskPDF arXiv

Radar odometry is crucial for robust localization in challenging environments; however, the sparsity of reliable returns and distinctive noise characteristics impede its performance. This paper introduces geometrically-constrained radar-inertial odometry and mapping that jointly consolidates point and pose uncertainty. We employ the continuous trajectory model to estimate the pose uncertainty at any arbitrary timestamp by propagating uncertainties of the control points. These pose uncertainties are continuously integrated with heteroscedastic measurement uncertainty during point projection, thereby enabling dynamic evaluation of observation confidence and adaptive down-weighting of uninformative radar points. By leveraging quantified uncertainties in radar mapping, we construct a high-fidelity map that improves odometry accuracy under imprecise radar measurements. Moreover, we reveal the effectiveness of explicit geometrical constraints in radar-inertial odometry when incorporated with the proposed uncertainty-aware mapping framework. Extensive experiments on diverse real-world datasets demonstrate the superiority of our method, yielding substantial performance improvements in both accuracy and efficiency compared to existing baselines. We release our open-source implementation at https://github.com/wooseongY/GeoRIO.

Abstract

Radar-Inertial Odometry

Explore similar work

Geometrically-Constrained Radar-Inertial Odometry via Continuous Point-Pose Uncertainty Modeling

Dense Soft Weighting for Radar Ego-Velocity Estimation

Pushing Radar Odometry Beyond the Pavement: Current Capabilities and Challenges