Reon Tabata

Experience and Awards

Tue, 26 May 2026 00:00:00 +0000

Research experience

Research Assistant — AIST Smart Mobility Research Group

Research on autonomous navigation and mapping, including multimodal registration between camera images and LiDAR point clouds.

Research internship — AIST

Developed , an augmented-reality game that demonstrates LiDAR-based 3D localization.

Designed and validated a system that uses 3D LiDAR and an RGB camera to detect a ramp entrance, guide a wheelchair-type robot to it, and perform the climb.

Selected awards and activities

Technical Award and Top 4, ABU Asia-Pacific Robot Contest 2024, as a member of the Japanese representative team.
Champion, NHK Student Robot Contest 2024, contributing to the competition’s first three-peat.
Runner-up, Tokai Regional Exchange Robocon 2023.
Design Award and National Competition qualification, KOSEN Robocon 2020 Tohoku Regional Contest.
Runner-up, Catch-Robot Battle Contest 2025.

Across eight years of robotics competitions, I worked on embedded control, wireless communication, system integration, and 3D object recognition, while also taking leadership and team-management roles.

Image-to-Point Cloud Registration Made Easy with Rectified Flow-based LiDAR Upsampling

Fri, 01 May 2026 00:00:00 +0000

This manuscript proposes converting sparse LiDAR point clouds into dense point-cloud images with a generative model, enabling image-domain feature extraction and matching methods to be applied to image-to-point-cloud registration. The manuscript was under review when the source document was prepared in May 2026.

Slope Detection and Mobile Robot Guidance Using 3D LiDAR and an RGB Camera

Sun, 01 Jun 2025 00:00:00 +0000

This work combines 3D LiDAR and an RGB camera to detect the entrance of a slope from a distance, guide a wheelchair-type mobile robot to the entrance, and execute the climb.

Autonomous Slope Detection and Guidance

Mon, 01 Jan 0001 00:00:00 +0000

This system detects a slope entrance from a distance, navigates a wheelchair-type robot to it, and performs the climb. It was developed and validated on a real robot as an undergraduate research project.

GhostScanner