A dual-arm mobile robot for repetitive warehouse and factory work. Arms from JPY 260,000, complete robot from JPY 2,000,000.
Standard robot: Manmaru YOR
Manmaru YOR is a dual-arm omnidirectional mobile robot based on the open-source YOR design. We redesigned the arms, arm layout, sheet-metal frame, and parts sourcing in-house. Each unit is assembled in Japan, calibrated, tested, and delivered with a ROS 2 environment set up. Motions are trained with our in-house Sim2Real pipeline. Depending on the task, we also use third-party robots such as Unitree and AgiBot.
Differences from the original YOR
| Item | Change |
|---|---|
| Arms | Two reBot Arm B601-DM arms |
| Arm layout | Mounted on 45-degree brackets so both arms work together in a shared central workspace |
| Sheet-metal frame | Redesigned in-house |
| Parts sourcing | Reworked so parts can be sourced and assembled in Japan |
| Productization | Delivered assembled, calibrated, tested, and with ROS 2 set up |
Specifications (Manmaru YOR)
- Arms
- reBot Arm B601-DM ×2
- Degrees of freedom
- 6+1 per arm (6 axes + 1 gripper axis)
- Reach
- 767 mm per arm (607 mm without gripper)
- Payload
- 1.5 kg per arm
- Repeatability
- ±0.2 mm per arm
- Base
- Omnidirectional, 4 independently steered and driven wheels (REV 3" MAXSwerve ×4)
- Height
- 841.1 mm (as drawn)
- Arm width
- 602.5 mm (stowed)
Values are nominal design specifications and may change depending on deployment conditions.
Public resources
Sim2Real
Validate the task in simulation before you deploy
ManmaruAI develops its own Sim2Real pipeline. We rebuild the target task in simulation, train the robot motion there, and transfer the result to the real robot. On-site teaching is reduced to checking and tuning on the robot.
Our standard robot, Manmaru YOR, is a dual-arm mobile robot based on the open-source YOR design and productized in Japan. Arms start at JPY 260,000 and the complete robot at JPY 2,000,000.
- 01Rebuild the task in simulation
We model the workpiece, shelves or totes, and the robot to recreate the task in Sim.
- 02Train and validate in Sim
The motion is trained and its success rate and trial count are recorded in Sim.
- 03Transfer to the robot (Sim2Real)
The trained motion moves to the real robot, where we record its success rate and transfer days.
- 04Check and tune on site
We check the motion under site conditions and tune it before operation.
Vision technology: quality assurance for robot work
A camera captures the workpiece the robot handles, and AI judges whether it has an anomaly. Along with the result, the reason and the anomaly location are shown on screen.
Inspection is built into the robot task itself, for example checking molded parts after removal or checking items before and after induction and packing. Parts judged anomalous can be removed by the robot using the location data.
We start by validating imaging conditions and confirm the judgment criteria with good and defective samples before integrating inspection into the robot task.
Anomaly judgment
Judges anomalies from an image file or a still image from a camera.
Localization
Shows where the anomaly is on the image, together with the reason.
Integration with the task
Feeds the result into robot motion to remove or recheck anomalous parts.
How we provide it
Self-build from public resources
Build it yourself from the published BOM and robot model.
Buy a complete robot
From JPY 2,000,000Delivered assembled, calibrated, and tested. Arms alone start at JPY 260,000.
Development support
Quoted individuallyConfiguration changes for your R&D goals and motion development with Sim2Real.
Request a workpiece fit check
Tell us about the workpiece, the task, and your site. A fit check from JPY 300,000 confirms whether Sim2Real can deliver it.