DNOI-4DRO: Deep 4D Radar Odometry with Differentiable Neural-Optimization Iterations

Shouyi Lu1, Huanyu Zhou1, Guirong Zhuo1*, Xiao Tang1
1School of Automotive Studies, Tongji University
*Corresponding author
Paper Code - Coming Soon... arXiv

Abstract

A novel learning-optimization-combined 4D radar odometry model, named DNOI-4DRO, is proposed in this paper. The proposed model seamlessly integrates traditional geometric optimization with end-to-end neural network training, leveraging an innovative differentiable neural-optimization iteration operator. In this framework, point-wise motion flow is first estimated using a neural network, followed by the construction of a cost function based on the relationship between point motion and pose in 3D space. The radar pose is then refined using Gauss-Newton updates. Additionally, we design a dual-stream 4D radar backbone that integrates multi-scale geometric features and clustering-based class-aware features to enhance the representation of sparse 4D radar point clouds. Extensive experiments on the VoD and Snail-Radar datasets demonstrate the superior performance of our model, which outperforms recent classical and learning-based approaches. Notably, our method even achieves results comparable to A-LOAM with mapping optimization using LiDAR point clouds as input.

Method

Overview of our backbone. (1) The feature and context extractors encode the point and context features of the input point cloud, respectively. (2) The feature correlation module constructs an all-pair correlation volume by calculating the matrix dot product of two-point features. (3) In each iteration, the differentiable neural-optimization iteration operator uses the pose estimated in the previous iteration to look up correlation features from the correlation volume, which are then processed through a GRU to generate a point motion field. The point motion field is fed into a least-squares-based optimization layer, where the pose is updated based on geometric constraints. After multiple iterations, the network outputs the predicted pose.

Overview of the proposed MultiEditor framework

Qualitative Visualization

BibTeX

@article{lu2025dnoi,
  title={DNOI-4DRO: Deep 4D Radar Odometry with Differentiable Neural-Optimization Iterations},
  author={Lu, Shouyi and Zhou, Huanyu and Zhuo, Guirong},
  journal={arXiv preprint arXiv:2505.12310},
  year={2025}
}