xDR Challenge2025
~Smartphone with Navigation Robot~

xDR Challenge is a series of indoor localization competition by PDR benchmark standardization committee mainly focused on integrated indoor localization methods with dead-reckoning. xDR Challenge 2025 will be held as an official competition track of IPIN 2025 (Track 5). Our track is categorized as "offsite, online" track, which means that the competitors are asked to submit estimated trajectories of targets in certein time limit in real-time. Application scenario of this year's competition is navigation of people with visual impairment in a facility by cooperating a smartphone and a navigation robot (AI Suitcase). Target field of this year’s competition is a national museum and the targets of localization are pedestrian walking in the field with a navigation robot. The pedestrian (we call "target") moves with/without a navigation robot in a museum floor and views exhibits with approximately 15-minute mock tours. The targets in this competition include people with/without visual impairment. The target wears a smartphone on the chest with visual-inertial odometry system. The smartphone communicates with the navigation robot, which is equipped with UWB for ranging with the smartphone, LiDAR and IMU for self-localization. The competitors are asked to estimate trajectories of the target using IMU, VIO, UWB, and location of the navigation robot. We will evaluated the estimated trajectories of the target by utilizing multi-faceted evaluation metrics.

Definition of data and trials used in this track

Trial dataset: Dataset provided for testing and tuning the localization and tracking methods (LTM) before the actual scoring trial. The trial dataset is small subset of the dataset and provided with ground-truth data.

Scoring dataset: Dataset provided for actual benchmarking. The scoring dataset is provided without ground-truth data. The benchmarking participants are asked to submit the result of LTM by processing the scoring dataset. In this competition, the scoring dataset may includes unseen subjects and/or paths

* These are based on definitions in ISO/IEC DIS 21134 (Information technology — Computer graphics, image processing and environmental data representation — Benchmarking of integrated indoor localization and tracking methods using dead reckoning)

Testing trials: Set of data distribution and result submission through the EvaalAPI on competition organizer's servers and are reloadable. Competitors can run them as many times as they like to evaluate and fine-tune their system as well as to get used to the EvaalAPI.These trials use some data from the trial dataset.

Scoring trials: Set of data distribution and result submission through the EvaalAPI on competition organizer's servers and are only available at specific times during the competition production. These trials use data from the scoring dataset.

(URL:)https://unit.aist.go.jp/rihsa/xDR-Challenge-2025/index.html

xDR Challenge 2025 ~Smartphone with Navigation Robot~ (Hereafter referred to as xDR Challenge 2025) will be held as a sequel competition of PDR/xDR Challenge series which had been hosted by PDR Benchmark standardization committee. As in previous years' competitions, the xDR Challenge 2025 will be held as an official competition tracks of the IPIN conference. There are three categories of tracks in IPIN 2025 competition. Our track is categorized as an "Off-site, On-line" track.

The off-site competition means that
- Competition organizers have conducted measurement and prepare dataset for the competition, and
- The dataset measured will be provided to competitors, and
- The competitors are required to estimate the targets movement by applying their own algorithm and submit the estimated results.
The on-line competition means that
- the competitors are required to submit the results in real-time, but to submit in certain (longer) time limit, and
- the competitors cannot obtain whole data of each sequence for applying global optimization while constraining with provided information.

The datasets of our track consists of sensor data required for the VIO and/or PDR-based indoor localization. The data are measured in the museum with mock tours with participants. We measured pedestrian movement by using iOS device and a navigation robot. We collected sensor (gyro, accelerometer, magnets sensor) data at best-effort basis as well as results of the visual-inertial odometry (VIO) by ARKit and the UWB range and AoA signals from the navigation robot's UWB tag.
To limit complexity of the competition, the participants will receive pre-computed VIO results (location and orientation at a frame), navigation robot's location and orientation (no LiDAR data from navigation robot) and UWB ranging and AoA (thus no CIR is distributed).
Similar to the previous PDR/xDR Challenges, submitted trajectories will be evaluated by multi-faced evaluation metrics. The ground truth was obtained by a following robot that measures its location by the LiDAR & IMU, and AR marker equipped on the neck of the participant by the camera. The detailed evaluation metrics will be announced at a later date.

xDR Challenge 2025 will be conducted closely together with other tracks of the IPIN competitions under a common schedule. We also adopt common tool named EvAAL API for sharing dataset and receiving results. As aligned with other tracks, real competition will be conducted at a day in 8 - 10 September 2025. We will provide test dataset in same format with the real dataset for allowing the competitors to prepare and adjust localization algorithm/systems before the real competition.

We look forward to your participation!

To be announced.

To be announced.

Step1 Request for Trial Dataset

If potential competitors have an interest in our competition, please register in the data request form (to be opened in June). We will provide you the trial dataset after we confirm your registration.

Step2 Downloading Trial Dataset

We will provide link to the trial dataset for those who completed the request for data form. You can download the dataset with provided ID and password. Note that the organizers will manally check the request form contents so it may take a few days to send the access information. After downloading the sample data, the competitor can start testing trials.

Step3 Application of the Competition

If you decide to join the competition, please register final admission in IPIN Competition's official application form. The competitors are required to provide short and long descriptions of the system

Step4 Payment of subscription fee

In addition to the application, the competitors are required to pay the registration fee. The detail of the payment will be announced later.

Step5 Scoring trial

On a track-specific day during the second week of September, the competitors are required to join the scoring trial. We will provide unreferenced data for scoring trial through EvAAL API. Competitors will be asked to call the EvAAL API to get sensor data and submit estimated location in real-time.

Step6 Announcement of the results

The result of the xDR Challenge 2025 will be announced during the IPIN 2025 conference

To be released in June.
The detail of the dataset and an example of estimation method are described in [S. Ogiso et al, IEEE/ION PLANS 2025]. If you use this dataset, please cite this paper to reference the dataset detail.

This competition is based on results obtained from a project, Programs for Bridging the gap between R&D and the IDeal society (society 5.0) and Generating Economic and social value (BRIDGE)/Practical Global Research in the AI × Robotics Services, implemented by the Cabinet Office, Government of Japan

LTM : Localization and tracking methods
xDR : x (pedestrian / vehicle) dead reckoning
PDR : pedestrian dead reckoning
IMU : inertial masurement unit
LiDAR : light detection and ranging
VIO : visual-inertial odometry
BLE : Bluetooth low energy
UWB : ultra-wide band
AoA : Angle of Arrival

e-mail: M-xDR-Challenge2025-ml@aist.go.jp