End-to-End LiDAR optimization for 3D point cloud registration

TL;DR

This paper introduces an adaptive LiDAR sensing framework that dynamically optimizes sensor parameters and registration hyperparameters, enhancing 3D point cloud registration accuracy and efficiency by integrating feedback into the sensing process.

Contribution

It presents a novel joint optimization approach for LiDAR data acquisition and registration, which is adaptive and feedback-driven, unlike traditional fixed-parameter methods.

Findings

Outperforms fixed-parameter baselines in simulation

Improves registration accuracy and efficiency

Retains generalization across different scenarios

Abstract

LiDAR sensors are a key modality for 3D perception, yet they are typically designed independently of downstream tasks such as point cloud registration. Conventional registration operates on pre-acquired datasets with fixed LiDAR configurations, leading to suboptimal data collection and significant computational overhead for sampling, noise filtering, and parameter tuning. In this work, we propose an adaptive LiDAR sensing framework that dynamically adjusts sensor parameters, jointly optimizing LiDAR acquisition and registration hyperparameters. By integrating registration feedback into the sensing loop, our approach optimally balances point density, noise, and sparsity, improving registration accuracy and efficiency. Evaluations in the CARLA simulation demonstrate that our method outperforms fixed-parameter baselines while retaining generalization abilities, highlighting the potential of adaptive LiDAR for autonomous perception and robotic applications.