# Speeding Up Iterative Closest Point Using Stochastic Gradient Descent

## TL;DR

This paper introduces a stochastic gradient descent-based approach to accelerate the iterative closest point algorithm for 3D point cloud alignment, achieving faster convergence while maintaining accuracy and robustness across sensors.

## Contribution

The paper presents a novel ICP optimization method using SGD, significantly improving convergence speed and robustness compared to traditional ICP methods.

## Key findings

- Faster convergence than existing ICP methods.
- Solutions comparable in quality to standard ICP.
- Robustness to sensor parameter variations.

## Abstract

Sensors producing 3D point clouds such as 3D laser scanners and RGB-D cameras are widely used in robotics, be it for autonomous driving or manipulation. Aligning point clouds produced by these sensors is a vital component in such applications to perform tasks such as model registration, pose estimation, and SLAM. Iterative closest point (ICP) is the most widely used method for this task, due to its simplicity and efficiency. In this paper we propose a novel method which solves the optimisation problem posed by ICP using stochastic gradient descent (SGD). Using SGD allows us to improve the convergence speed of ICP without sacrificing solution quality. Experiments using Kinect as well as Velodyne data show that, our proposed method is faster than existing methods, while obtaining solutions comparable to standard ICP. An additional benefit is robustness to parameters when processing data from different sensors.