RaNDT SLAM: Radar SLAM Based on Intensity-Augmented Normal Distributions Transform

TL;DR

RaNDT SLAM introduces a fast, accurate radar-based SLAM framework that leverages intensity-augmented Normal Distributions Transform and sensor fusion, suitable for unstructured and vision-denied environments.

Contribution

It presents a novel radar SLAM method combining intensity data with NDT, enabling efficient and precise localization in challenging environments.

Findings

Operates faster than existing radar SLAM methods.

Achieves high accuracy in indoor and outdoor environments.

Validated on a new benchmark and Oxford Radar RobotCar datasets.

Abstract

Rescue robotics sets high requirements to perception algorithms due to the unstructured and potentially vision-denied environments. Pivoting Frequency-Modulated Continuous Wave radars are an emerging sensing modality for SLAM in this kind of environment. However, the complex noise characteristics of radar SLAM makes, particularly indoor, applications computationally demanding and slow. In this work, we introduce a novel radar SLAM framework, RaNDT SLAM, that operates fast and generates accurate robot trajectories. The method is based on the Normal Distributions Transform augmented by radar intensity measures. Motion estimation is based on fusion of motion model, IMU data, and registration of the intensity-augmented Normal Distributions Transform. We evaluate RaNDT SLAM in a new benchmark dataset and the Oxford Radar RobotCar dataset. The new dataset contains indoor and outdoor environments besides multiple sensing modalities (LiDAR, radar, and IMU).