Performance Analysis of Automotive SAR With Radar Based Motion Estimation

Oded Bialer, Tom Tirer

Abstract

Automotive synthetic aperture radar (SAR) can achieve a significant angular resolution enhancement for detecting static objects, which is essential for automated driving. Obtaining high resolution SAR images requires precise ego vehicle velocity estimation. A small velocity estimation error can result in a focused SAR image with objects at offset angles. In this paper, we consider an automotive SAR system that produces SAR images of static objects based on ego vehicle velocity estimation from the radar return signal without the overhead in complexity and cost of using an auxiliary global navigation satellite system (GNSS) and inertial measurement unit (IMU). We derive a novel analytical approximation for the automotive SAR angle estimation error variance when the velocity is estimated by the radar. The developed analytical analysis closely predicts the true SAR angle estimation variance, and also provides insights on the effects of the radar parameters and the environment condition on the automotive SAR angle estimation error. We evaluate via the analytical analysis and simulation tests the radar settings and environment condition in which the automotive SAR attains a significant performance gain over the angular resolution of the short aperture physical antenna array. We show that, perhaps surprisingly, when the velocity is estimated by the radar the performance advantage of automotive SAR is realized only in limited conditions. Hence since its implementation comes with an increase in computation and system complexity as well as an increase in the detection delay it should be used carefully and selectively.

Introduction

Standard automotive radars typically suffer from poor angular resolution, often limited to approximately $1^\circ$. This limitation hinders the ability of autonomous systems to distinguish between closely spaced static objects such as parked vehicles, utility poles, or curbs. Synthetic Aperture Radar (SAR) technology offers a potential solution by utilizing the vehicle's own motion to synthesize a larger virtual antenna aperture, thereby significantly enhancing angular resolution.

However, the efficacy of SAR is highly dependent on precise motion estimation. Traditional SAR systems rely on high-precision Global Navigation Satellite Systems (GNSS) and Inertial Measurement Units (IMU) to track the radar's position with millimeter-level accuracy. To reduce system cost and complexity, there is significant interest in estimating vehicle velocity directly from the radar return signals. This approach, however, introduces a "velocity-angle ambiguity," where small errors in velocity estimation lead to spatially displaced but seemingly focused images.

Geometry of automotive SAR illustrating the host vehicle trajectory and the resulting displacement of detected reflection points due to velocity estimation errors.

Methodology

This research develops a novel analytical framework to predict the error variance in automotive SAR angle estimation when velocity is derived from radar data. The authors employ first-order Taylor expansions to establish a mathematical link between errors in initial radar measurements—specifically Doppler and angle—and the resulting SAR output.

A critical component of this methodology is the formal modeling of the velocity-angle ambiguity. This phenomenon explains why images can remain sharp even when spatially inaccurate, as velocity errors are "masked" by corresponding angle offsets. The researchers derived a closed-form expression to quantify how SAR estimation errors scale with parameters such as vehicle speed, the number of static targets in the environment, and the number of radar frames processed.

Block diagram of the SAR processing chain, showing the integration of matched filter outputs from multiple frames to generate the final SAR output.

The analytical model was validated through extensive synthetic simulations of a 77GHz FMCW MIMO radar system. These simulations allowed for precise control over noise levels and ground-truth velocities, testing the system across a range of speeds (3 m/s to 25 m/s) and target densities (2 to 15 reflection points).

Key Insights and Results

The study reveals that the performance benefits of automotive SAR are highly conditional:

  • The "Focused but Wrong" Phenomenon: Radar-based motion estimation often results in images that are sharp but spatially shifted. Because these images are not blurry, standard auto-focusing algorithms fail to identify the spatial error.
  • Speed and Performance Plateau: While increased vehicle speed extends the synthetic aperture, it also amplifies the impact of measurement errors from the physical antenna array. This leads to a performance plateau where further speed increases do not yield proportional resolution gains.
  • Angular Sensitivity: SAR performance is found to be lowest at the boresight (directly in front of the vehicle) and improves as the squint angle increases toward the sides—a behavior opposite to that of traditional physical radar arrays.
  • Operational Constraints: Significant performance gains over standard physical arrays are typically only realized at speeds exceeding 10 m/s and in environments with a high density of static reflection points. In sparse environments, SAR may underperform compared to traditional radar.

Conclusion

While SAR offers a path to high-resolution automotive imaging, its implementation using radar-only motion estimation requires careful consideration. The identified limitations regarding speed, target density, and spatial accuracy suggest that SAR should be deployed selectively. Future work will investigate hybrid systems that integrate radar-based estimation with lower-cost GNSS/IMU sensors to overcome these constraints and provide more reliable performance in diverse driving environments.

Summary updated

Abstract

Automotive synthetic aperture radar (SAR) can achieve a significant angular resolution enhancement for detecting static objects, which is essential for automated driving. Obtaining high resolution SAR images requires precise ego vehicle velocity estimation. A small velocity estimation error can result in a focused SAR image with objects at offset angles. In this paper, we consider an automotive SAR system that produces SAR images of static objects based on ego vehicle velocity estimation from the radar return signal without the overhead in complexity and cost of using an auxiliary global navigation satellite system (GNSS) and inertial measurement unit (IMU). We derive a novel analytical approximation for the automotive SAR angle estimation error variance when the velocity is estimated by the radar. The developed analytical analysis closely predicts the true SAR angle estimation variance, and also provides insights on the effects of the radar parameters and the environment condition on the automotive SAR angle estimation error. We evaluate via the analytical analysis and simulation tests the radar settings and environment condition in which the automotive SAR attains a significant performance gain over the angular resolution of the short aperture physical antenna array. We show that, perhaps surprisingly, when the velocity is estimated by the radar the performance advantage of automotive SAR is realized only in limited conditions. Hence since its implementation comes with an increase in computation and system complexity as well as an increase in the detection delay it should be used carefully and selectively.
BibTeX
@article{bialer2023performance,
  author = {Oded Bialer and Tom Tirer},
  title = {{Performance Analysis of Automotive SAR With Radar Based Motion Estimation}},
  journal = {IEEE Trans. Vehicular Technology},
  year = {2023},
  url = {https://openreview.net/forum?id=saJSL-zsIHQ},
}
Figure