Do Spinning Radar Doppler Velocity Measurements Improve Vehicle Detection and Tracking?

Eric Xie; Daniil Lisus; Timothy D. Barfoot

cs.ROSep 17, 2026

Do Spinning Radar Doppler Velocity Measurements Improve Vehicle Detection and Tracking?

Authors: Eric Xie, Daniil Lisus, Timothy D. Barfoot

Organizations: Robotics Institute, University of Toronto, Canada

Abstract

Spinning frequency-modulated continuous-wave (FMCW) radars have been gaining popularity in autonomous vehicle perception on account of their robustness to adverse weather conditions and 360° field of view. Recently, scanning radars have also been shown capable of generating per-azimuth Doppler velocity. In this paper, we investigate whether these Doppler velocity measurements improve spinning radar vehicle detection and tracking performance. For detection, we estimate the ego motion and use it to undo the Doppler range distortion of the radar image before passing it to a network. For tracking, we propose a new way to estimate a per-vehicle velocity and use it as a prior for the tracker's motion model. Since Doppler-enabled spinning radar data is not available in any dataset with ground-truth dynamic object labels, our first contribution is an automatic labelling pipeline that uses an ensemble of fine-tuned off-the-shelf lidar detectors to label all 643 km of the Boreas Road Trip dataset. We then transfer detections to radar, and use over 250 km of vehicle-dense sequences as ground-truth training data. By training and evaluating two state-of-the-art detectors, we show that Doppler undistortion can improve detection accuracy by up to $2.37$ points on mean average precision. Furthermore, we show that the Doppler velocity prior can improve tracking accuracy by $13.68$ points on multi-object tracking accuracy (MOTA) versus the zero-velocity initialization baseline, while achieving $99.7\%$ of the MOTA obtained using ground-truth velocities as the prior.

Figures & tables

Fig. 1: Tracking with the baseline’s zero-velocity initialization (left) and with our Doppler velocity prior (right) on two Boreas-RT sequences. Box colour denotes track identity, and magenta arrows show the estimated velocity. The reference car in (a) moves to the right at $35\text{\,}\mathrm{m}$ , and the two in (b) to the left at $25.9\text{\,}\mathrm{m}$ and $29.8\text{\,}\mathrm{m}$ , respectively. The baseline tracker is unable to associate the radar labels across frames, causing track fragmentation. With our Doppler prior, the detections are associated into single continuous tracks.

Panel of Figure 2 — Fig. 2: Automatic lidar-to-radar labelling on an Urban sequence of Boreas-RT. Starting from the $10\text{\,}\mathrm{Hz}$ lidar input (a), stage (b) detects and fuses proposals (cyan, dashed then solid) for three vehicles with ticks indicating box heading. Stage (c) generates tracks, where purple, orange, and green identify a stationary, straight-moving, and turning vehicle, respectively. Velocity arrows span the distance travelled in one second. Stage (d) refines tracks with a continuous-time $\mathrm{SE}(3)$ trajectory, and stage (e) samples the trajectory at the radar azimuth timestamp and transfers the resulting boxes to the $4\text{\,}\mathrm{Hz}$ radar scans.

Panel of Figure 3 — Fig. 3: Skyway scene of Boreas-RT with a vehicle travelling at $32.1\text{\,}\mathrm{m}$ , (a) raw and (b) after Doppler undistortion with the ground-truth ego velocity. Orange dashed boxes transfer the nearest lidar frame at its midpoint, and teal solid boxes sample the vehicle’s continuous trajectory at the estimated azimuth-crossing time. The two boxes are separated by $8.6\text{\,}\mathrm{m}$ . The brackets mark the spread of the vehicle’s return. Note how the vehicle’s return in the undistorted scan fit tighter within the teal box, with the remaining offset due to the target’s own motion. Intensity is normalized per panel.

Fig. 4: Single-scan Doppler velocity initialization for tracking, following one vehicle through three stages. (a–c) Doppler undistortion using the estimated ego velocity and an annular sector (purple) around a detection box (cyan). (d–g) Cross-correlation within the polar rectangle and consensus selection of per-azimuth radial velocities (red). (h–i) Projection onto the box heading, giving the velocity prior (teal), and track initialization. This Boreas-RT vehicle is located $12.8\text{\,}\mathrm{m}$ from the radar, with an estimated speed of $7.5\text{\,}\mathrm{m}$ against a ground-truth $8.8\text{\,}\mathrm{m}$ .

		Frames (k)		Boxes (k)		Tracks (k)		Averages
Route	# of seqs.	Lidar	Radar	Lidar	Radar	Lidar	Radar	Seq. length (km)	Target speed (m/s)
Regional	6	49.3	20.5	930.4	388.3	8.4	8.4	9.3	4.3
Skyway	5	28.8	12.0	370.6	153.9	5.3	5.2	11.1	9.7
Suburbs	10	103.3	43.1	2,195	913.2	18.0	18.0	7.9	2.1
Urban	7	164.5	68.4	2,886	1,203	14.2	14.2	8.6	2.4
Industrial	5	43.4	18.1	803.7	335.9	5.5	5.5	5.4	1.8

TABLE I: Final auto-generated labels by route. The grey routes are used to train the RaFD and SIRA detectors. Target speed is the absolute speed of surrounding vehicles, including stationary vehicles.

Panel of Figure 6 — Fig. 5: Four vehicle tracks from the Urban and Regional routes of Boreas-RT, before refinement (left) and with our refiner (right). Boxes are per-frame poses, coloured by track identity, and magenta arrows are the estimated velocity, suppressed below $0.5\text{\,}\mathrm{m}$ . The four failure modes are (a) static jitter, (b) heading flips, (c) motion jitter, and (d) fragmentation.

	Good weather			Good + bad weather
Method	$\mathrm{mAP}@0.3$	$\mathrm{mAP}@0.5$	$\mathrm{mAP}@0.7$	$\mathrm{mAP}@0.3$	$\mathrm{mAP}@0.5$	$\mathrm{mAP}@0.7$
RaFD	$\mathbf{69.4\pm 1.5}$	$\mathbf{59.5\pm 1.9}$	$\mathbf{23.9\pm 1.1}$	$68.8\pm 1.8$	$54.2\pm 1.7$	$18.8\pm 1.6$
Our RaFD	$67.8\pm 2.0$	$56.9\pm 1.5$	$20.0\pm 1.6$	$\mathbf{72.0\pm 2.8}$	$\mathbf{59.7\pm 2.6}$	$\mathbf{20.3\pm 0.9}$
SIRA	$68.7\pm 1.1$	$58.1\pm 1.4$	$22.8\pm 0.9$	$66.1\pm 0.8$	$53.8\pm 1.1$	$19.9\pm 1.0$
Our SIRA	$\mathbf{69.4\pm 1.8}$	$\mathbf{58.7\pm 2.3}$	$\mathbf{23.9\pm 1.7}$	$\mathbf{68.6\pm 0.6}$	$\mathbf{55.0\pm 0.2}$	$\mathbf{20.4\pm 0.5}$

TABLE II: RaFD and SIRA reproduction on RADIATE. We report mean

\pm

sample standard deviation over three seeds. Bold marks the better of the published [ 39 , 41 ] and reproduced scores.

Input	$\mathrm{mAP}@0.3$	$\mathrm{mAP}@0.5$	$\mathrm{mAP}@0.7$
RaFD, raw	$83.82\pm 0.10$	$80.21\pm 0.10$	$65.13\pm 0.20$
+ Estimated ego	$\mathbf{84.85\pm 0.05}$	$81.45\pm 0.13$	$67.50\pm 0.10$
+ GT ego	$84.75\pm 0.13$	$\mathbf{81.50\pm 0.20}$	$\mathbf{67.65\pm 0.17}$
SIRA, raw	$76.46\pm 0.35$	$73.00\pm 0.35$	$59.44\pm 0.56$
+ Estimated ego	$76.28\pm 0.86$	$73.15\pm 0.88$	$60.44\pm 1.21$
+ GT ego	$\mathbf{76.57\pm 0.38}$	$\mathbf{73.40\pm 0.35}$	$\mathbf{60.70\pm 0.18}$

TABLE III: Detection results on Boreas-RT. We report mean

\pm

sample standard deviation over three seeds. GT and estimated refer to the ego velocity used for Doppler undistortion. Bold marks the best input per detector.

Fig. 6: Longitudinal velocity errors on

199,253

estimated vehicles from Skyway , Suburbs , and Urban sequences. The shaded band marks errors within

2\text{\,}\mathrm{m}

, and errors outside

[-10,10]

\mathrm{m}

accumulate in the boundary bins. Mean absolute error (MAE) and root-mean-square error (RMSE) use the untruncated errors and differ from the planar

\mathrm{MAE}_{V}

used for tracking.

Fig. 7: Tracking with GT boxes on seven Boreas-RT test sequences, varying only the birth velocity. Frames are grouped by mean reference-vehicle speed. MOTA is multi-object tracking accuracy, IDF1 is the identity F1 score, IDSW is identity switches per

1,000

reference boxes, and

\mathrm{MAE}_{V}

is the mean planar velocity error. Zero init. diverges from both priors above

10\text{\,}\mathrm{m}

Fig. 8: Tracking RaFD [ 39 ] (solid) and SIRA [ 41 ] (dashed) detections on the same seven test sequences as Fig. 7 , with zero init. (circles) or the Doppler prior (squares). Speed bins and metrics match Fig. 7 . No GT-velocity prior is available for detector boxes.

Prior	HOTA $\uparrow$	DetA $\uparrow$	AssA $\uparrow$	MOTA $\uparrow$	IDF1 $\uparrow$	IDSW $\downarrow$	Frag. $\downarrow$	FP $\downarrow$	FN $\downarrow$	$\mathrm{MAE}_{V}\downarrow$
Ground-truth boxes
Zero init.	85.16	80.66	90.06	76.86	86.41	25,511	6,478	117,232	54,440	1.31
Doppler	91.39	89.40	93.48	90.54	94.12	1,001	2,424	40,235	39,384	0.25
GT vel.	91.57	89.57	93.68	90.79	94.29	232	2,336	39,441	38,837	0.18
RaFD detections
Zero init.	51.26	45.23	58.71	32.21	61.51	23,298	52,552	301,007	253,293	1.64

TABLE IV: Tracking on seven test sequences with

852,000

ground-truth reference boxes. Bold marks the best prior per source. HOTA, DetA, and AssA are higher order tracking accuracy and its detection and association components, MOTA multi-object tracking accuracy, IDF1 identity F1, IDSW identity switches, Frag. fragmentations, FP/FN false positives/negatives, and

\mathrm{MAE}_{V}

the mean planar velocity error in

\mathrm{m}

Autonomous Driving Perception

Autonomous driving perception enables vehicles to detect and interpret road users, drivable space, and hazards using onboard and shared sensors.

See all 379 papers

Automotive mmWave Spinning Radar Place Recognition with Spatially Gated Feature-Correlation Representation
Sep 23, 2026Saimunur Rahman, Sagun Singh Shrestha, Abdelwahed Khamis +1Millimeter-Wave Radar mmWave Radar Sensing
CSIRO Robotics, CSIRO, Australia · Queensland University of Technology, Australia
Doppler-aided Position-Velocity Tracking of Hidden Objects using mmWave Radar
Oct 25, 2026Yerin Cheon, Kush Garg, Akshat DaveWireless Localization mmWave Radar Sensing
Stony Brook University, 100 Nicolls Rd, Stony Brook, NY 11794, United States. · Delhi Technological University, 110042, New Delhi, India.
Real-Time Radar Semantic and Panoptic Scene Understanding
Oct 19, 2026Santosh Parit, Julia Hindel, Rohit Mohan +4Autonomous Driving Perception Panoptic Segmentation
University of Freiburg · University of Oxford

Sep 23, 2026J/K move · Enter open · S save

Automotive mmWave Spinning Radar Place Recognition with Spatially Gated Feature-Correlation Representation

Saimunur Rahman, Sagun Singh Shrestha, Abdelwahed Khamis, Peyman Moghadam

CSIRO Robotics, CSIRO, Australia · Queensland University of Technology, Australia

Millimeter-Wave Radar mmWave Radar Sensing

Open paper AskPDF arXiv

Automotive spinning FMCW radar provides dense,

360^\circ

sensing and remains reliable under poor illumination and adverse weather, making it well-suited to autonomous navigation. Place recognition uses these observations to identify previously visited locations for re-localization and long-term navigation. However, heading changes appear as circular shifts in the polar radar representation, and conventional global aggregation can lose relationships among radar responses that are important for distinguishing similar places. We propose SGCA-Net, a spinning radar place recognition framework that combines rotation-robust feature extraction with Spatially Gated Correlation Aggregation (SGCA). SGCA learns spatial weights to reduce the influence of unstable and ambiguous radar regions, while aggregating pairwise correlations among local responses to preserve informative feature relationships. Experiments on the MulRan dataset show that SGCA-Net consistently outperforms SOTA methods across urban, campus, and open-road environments, while remaining robust to substantial heading variation. Evaluation on the HeRCULES dataset further demonstrates that SGCA-Net generalizes to unseen environments and radar sensors without fine-tuning.

Abstract

Autonomous Driving Perception

Explore similar work

Automotive mmWave Spinning Radar Place Recognition with Spatially Gated Feature-Correlation Representation

Doppler-aided Position-Velocity Tracking of Hidden Objects using mmWave Radar

Real-Time Radar Semantic and Panoptic Scene Understanding