Collecting Larg-Scale Robotic Datasets on a High-Speed Mobile Platform

Yuxin Lin,Jiaxuan Ma,Sizhe Gu,Jipeng Kong,Bowen Xu,Xiting Zhao,Dengji Zhao,Wenhan Cao,Sören Schwertfeger

2024-08-01

Abstract:Mobile robotics datasets are essential for research on robotics, for example for research on Simultaneous Localization and Mapping (SLAM). Therefore the ShanghaiTech Mapping Robot was constructed, that features a multitude high-performance sensors and a 16-node cluster to collect all this data. That robot is based on a Clearpath Husky mobile base with a maximum speed of 1 meter per second. This is fine for indoor datasets, but to collect large-scale outdoor datasets a faster platform is needed. This system paper introduces our high-speed mobile platform for data collection. The mapping robot is secured on the rear-steered flatbed car with maximum field of view. Additionally two encoders collect odometry data from two of the car wheels and an external sensor plate houses a downlooking RGB and event camera. With this setup a dataset of more than 10km in the underground parking garage and the outside of our campus was collected and is published with this paper.

Robotics

What problem does this paper attempt to address?

The paper primarily addresses the challenges faced by mobile robots in collecting large-scale datasets, especially when efficient data collection is required over larger areas. Specifically, the original ShanghaiTech Mapping Robot, based on the Clearpath Husky mobile platform, has a maximum speed of only 1 meter/second, which limits its ability to collect large-scale datasets outdoors. To solve this problem, the research team designed a new high-speed mobile platform—a rear-wheel-drive cart—and mounted the original mapping robot on this platform.

Through this design, the researchers achieved the following goals:
1. **Increased Speed and Coverage**: The new mobile platform can travel at higher speeds, allowing it to cover larger geographic areas more quickly.
2. **Improved Data Collection Quality**: By adding additional sensors (such as a downward-looking RGB camera and an event camera), the quality of data collection was further enhanced.
3. **Reliable Odometry Estimation**: An external dual-wheel encoder system was designed and implemented to record odometry data, and its accuracy was verified through experiments.

Ultimately, the research team used this system to collect a 10.21-kilometer dataset on the ShanghaiTech University campus and its underground parking lot, and generated a 3D point cloud map containing 73 million points using the LIO-SAM algorithm. These data and related code have been released for other researchers to download and use. The entire project aims to provide high-quality large-scale datasets for research on mobile robot localization, mapping, and other related studies.