RAMBA: 4D Radar Mapping by Bundle Adjustment

2026-05-24

AI Summary

This work addresses the lack of effective methods for offline global map optimization with 4D radar by proposing RAMBA, a novel framework that introduces bundle adjustment to 4D radar mapping for the first time. RAMBA jointly optimizes multi-frame radar states within a unified optimization framework by integrating covariance-weighted voxelized multi-frame geometric residuals, IMU preintegration factors, radar ego-velocity constraints, and loop closure detections. Experimental results demonstrate that RAMBA significantly improves map consistency and trajectory accuracy on the ColoRadar and SNAIL Radar datasets, outperforming existing radar-inertial odometry and pose graph optimization approaches.

Abstract

4D radar is increasingly attractive for robotic mapping because it provides range, azimuth, elevation, and Doppler measurements while remaining robust in adverse visibility conditions. Although recent radar and radar--inertial odometry methods have achieved promising online state estimation performance, offline global map refinement for 4D radar remains underexplored. This paper presents RAMBA, a radar bundle-adjustment framework for globally consistent 4D radar mapping. Given initial poses and radar frames from a radar--inertial odometry front-end, RAMBA jointly refines radar frame states using covariance-weighted geometric residuals, IMU preintegration factors, and radar ego-velocity constraints. The geometric residuals extend pairwise GICP to a multi-frame optimization by forming voxel-based correspondences across selected frames and weighting each residual with point covariances. To improve robustness against drift and revisits, RAMBA enforces temporal consistency during correspondence formation while explicitly supporting loop-closure constraints. Experiments on the ColoRadar and SNAIL Radar datasets show that RAMBA improves map consistency and usually enhances trajectory accuracy over radar--inertial odometry and pose-graph optimization baselines.

Problem

• 4D radar
• global map refinement
• offline optimization
• radar mapping

Innovation

• 4D radar
• bundle adjustment
• multi-frame optimization
• covariance-weighted residuals
• loop closure