General Place Recognition

ICRA 2022 Competition

Competition Background

The ability of mobile robots to recognize previously visited or mapped areas is essential to achieving reliable autonomous systems. Place recognition for localization and SLAM is a promising method for achieving this ability. However, current approaches are negatively affected by differences in viewpoint and environmental conditions that affect visual appearance (e.g. illumination, season, time of day), and so these methods struggle to provide continuous localization in environments that change over time. These issues are compounded when trying to localize in large-scale (several-kilometer length) maps, where the effects of repeated terrain and geometries result in greater localization uncertainty. This competition aims to push visual and LiDAR state-of-the-art techniques for localization in large-scale environments, with an emphasis on environments with changing conditions.

To evaluate localization performance under long-term and large-scale tasks, we provide benchmark datasets to investigate robust re-localization under changing viewpoints and domain differences. This competition will include two challenges:

  • City-scale UGV Localization Dataset (3D-3D Localization)

Sign up website: https://www.aicrowd.com/challenges/icra2022-general-place-recognition-city-scale-ugv-localization

  • Visual Terrain Relative Navigation Dataset (2D-2D Localization)

Sign up website: https://www.aicrowd.com/challenges/icra2022-general-place-recognition-visual-terrain-relative-navigation

For the details, please refer to the dataset page.

Software Development Kit (SDK)

To accelerate the process of development, we provide a complete SDK for loading datasets and evaluating results. Python APIs are given so it would be convenient for participants to integrate and use the interfaces in their code, which include the API to:

  • Easy access to our datasets;

  • Evaluate metrics for visual/Lidar place recognition;

  • Submit results for online evaluation at crowdAI;

With this SDK, participants just need to focus on their algorithms and try to improve the place recognition accuracy.

The SDK is held in the Github repo: https://github.com/MetaSLAM/GPR_Competition

System Architecture of the MetaSLAM SDK

Schedule & Award

  • Dataset Open: 05/10/2022

  • 1-round Competition: 05/10/2022~05/24/2022

    • Presentation at ICRA 2022

  • 2-round Competition: 06/01/2022~09/01/2022

    • No1. in each track: $3000

    • No2. in each track: $2000

Excellent Competitors also have the chance to take the internship at CMU.

Sponsors

Talks

Michael Milford

Professor

@QUT

Luca Carlone

Associate Professor

@MIT

Junyan Zhu

Assistant Professor

@CMU

Ji Zhang

System Scientist

@CMU

Yimin Zhang

Chief Scientist

@Intel Labs China

Sebastian Scherer

Associate Research Professor

@CMU

Talk Schedule 05/24/2022 EST time

9:30 AM - 10:15 AM Dr. Yimin Zhang

10:15 AM - 11:00 AM Prof. Luca Carlone

11:00 AM - 12:00 PM Competition Panel Discussion

13:00 PM - 13:45 PM Prof. Jun-Yan Zhu

13:45 PM - 14:30 PM Prof. Michael Milford

14:30 PM - 15:15 PM Dr. Ji Zhang

15:15 PM - 16:00 PM Prof. Sebastian Scherer

Organizers

Sebastian Scherer Associate Research Professor@CMU

Peng Yin Project Scientist @ CMU

Zhipeng Cai Research Scientist @Embodied AI Lab, Intel

Shiqi Zhao Master@UCSD

Ivan Cisneros Master@CMU

Guangzhao Li

Master@CMU

Haowen Lai Master@Tsinghua

Ruohai Ge Mater@CMU

Contact

Email: pyin2@andrew.cmu.edu

Tel: +1 (412)-320-5786

Copyright @ 2022 The GPR-Competition Committee. All Rights Reserved.