Tohoku University
Robotic Systems Lab
Yamazaki–Bezerra Laboratory
Intelligent robotic systems that perceive, move, manipulate, reason, and adapt in the real world.
The Tohoku Robotic Systems Laboratory, jointly led as the Yamazaki–Bezerra Laboratory, studies intelligent robotic systems that perceive, move, manipulate, communicate, reason, and adapt in real-world environments.
Our research spans four interconnected pillars — manipulation planning for deformable objects, motion learning for mobile manipulators, action support by autonomous robots, and connected robotics with edge AI. We are building robots that are useful working systems, not only demonstrations, capable of handling difficult-to-model objects, flexible materials, uncertainty, and changing environments.
We welcome students and collaborators interested in contributing to any area of our research — whether in hardware experiments, perception and AI, manipulation, mobile robotics, simulation, IoT, edge AI, software systems, or human-machine interaction.
Research
Research Pillars
Manipulation Planning and Learning of Deformable Objects
- Cloth, cables, ropes, and other flexible materials
- Vision-guided manipulation and motion planning
- Learning methods for difficult-to-model objects
Motion Learning and Planning for Mobile Manipulators
- Robots that combine locomotion and manipulation
- Motion learning for cluttered indoor and outdoor spaces
- Integrated planning for base and arm movements
Action Support by Autonomous Robots
- Robots that support people during daily tasks
- Human action and intention recognition
- Assistance for dressing, fetching, and shared work
Connected Robotics and Edge AI
- Robot-IoT collaboration across devices and sensors
- Shared world models for distributed reasoning
- Edge AI for adaptive sensing and explanation
Projects
Student Projects
Grasping and Manipulating Flexible Objects
Student projects may include experiments on robotic handling of cloth, cables, and deformable materials using vision-guided manipulation.
Scene Understanding for Everyday Tasks
Student projects may include developing perception systems that recognize objects and understand spatial relationships to support robot task planning.
Mobile Robot Navigation in Unstructured Environments
Student projects may include autonomous navigation and path planning for mobile robots operating in indoor and outdoor environments.
Videos
Watch Our Robots in Action
See our students' projects and research in action on the laboratory's YouTube channel.
Visit our YouTube channel →Join Us
We welcome students and collaborators in intelligent robotics
From hardware experiments to AI, simulation, and human-machine cooperation — there are many ways to contribute.
Join UsFacilities
Facilities
Robot Experiment Room
A dedicated space for robotic manipulation and mobile robot experiments, equipped with motion capture and sensor systems.
Student Working Area
A shared workspace with individual desks and computers, where students carry out their day-to-day research.
Hardware Building Area
A workspace with 3D printers, hand tools, and fabrication equipment for building and refining robotic hardware parts.
News
Latest News
Paper accepted at IEEE SSRR 2026
Our collaborative work on language-guided mobile manipulation has been accepted at IEEE SSRR 2026, where Assistant Prof. Ranulfo Bezerra will also serve as Publication Chair.
Paper accepted in Frontiers in Robotics and AI
Tong Wu, Ranulfo Bezerra, and Kimitoshi Yamazaki's work on real-time manipulability optimization and whole-arm obstacle avoidance has been accepted in Frontiers in Robotics and AI.
Read moreNew academic year begins
We welcome new students to the laboratory for the 2026 academic year.
Laboratory website prototype under development
We are building a new bilingual website to present the laboratory's research and activities.
Contact
Get in Touch
Questions about visiting, collaboration, or joining the lab? We are happy to hear from you.
Contact