How Can Mobile Manipulator Robots Automate Warehouse Picking? A REEMAN Engineering Project

How Can Mobile Manipulator Robots Automate Warehouse Picking? A REEMAN Engineering Project

Exploring Warehouse Shelf Picking with a 100kg AMR Platform, 7-DOF Robotic Arm, and Vertical Lifting System

Autonomous mobile robots have transformed how materials move through warehouses. But moving an item between locations and retrieving it directly from a storage shelf are two different automation challenges.

How can a robot travel to a warehouse shelf, position a robotic arm, and prepare to retrieve a specific item without relying entirely on fixed workstations?

This question became the foundation of a mobile manipulation development project involving REEMAN ROBOT and a German industrial technology customer.

The project focuses on integrating autonomous transportation with robotic manipulation to support warehouse shelf picking and industrial material handling.

REEMAN selected its Big Dog Chassis autonomous mobile robot (AMR) platform, rated for a 100kg chassis payload, and combined it with a 7-degree-of-freedom robotic arm rated for 5kg, a 610mm arm reach, and a vertical lifting mechanism with 1,200mm of travel.

The resulting Mobile Manipulator Robot architecture brings together autonomous mobility, adjustable working height, and robotic arm positioning. The project also addresses Laser navigation, 3D environmental sensing, RGB-D vision integration, gripper selection, and software development requirements.

Rather than treating transportation and picking as unrelated operations, the engineering approach explores how they can become parts of a coordinated warehouse workflow.

Project Overview: A Warehouse Automation Challenge from Germany

The project originated from a practical requirement: enabling a mobile robot to approach different storage locations and support the retrieval of materials from warehouse shelves.

The customer's objective went beyond purchasing a conventional AMR for transportation.

The intended system needed to travel between warehouse locations, operate near storage shelves, adjust the robotic arm's working height, and support the development of object retrieval tasks.

These requirements led to a broader engineering discussion about how an autonomous mobile platform, robotic arm, lifting mechanism, and perception system should work together.

Why Is Warehouse Shelf Picking More Complex Than Autonomous Material Transport?

Warehouse picking requires a robot to interact with individual objects, while conventional AMR transportation primarily involves moving loads between defined locations.

In many industrial warehouses, an AMR can transport goods after a worker or another machine has placed them on the robot.

However, retrieving an individual component from a shelf involves additional decisions.

The robot must determine where the item is stored, position itself appropriately, establish the arm's operating height, identify the target, and execute a suitable handling movement.

For this project, the intended automation sequence can be summarized as:

Task assignment → Autonomous navigation → Shelf approach → Vertical positioning → Object localization → Robotic manipulation → Material transportation.

Each step introduces different engineering requirements.

The mobile platform must reach the required location. The robotic arm must operate within the available space. The perception system must provide suitable information about the environment and target object. The end effector must match the material being handled.

This is why warehouse picking cannot be addressed by robotic arm specifications alone.

The entire system must be designed around the task.

How REEMAN Selected the Mobile Manipulator Architecture

REEMAN's approach focuses on combining three complementary capabilities: autonomous movement, flexible robotic arm positioning, and adjustable operating height.

The Big Dog Chassis provides the mobile foundation, the 7-DOF robotic arm enables localized manipulation, and the vertical lifting mechanism expands the range of potential working heights.

1. Why Use the Big Dog Chassis?

The REEMAN Big Dog Chassis provides a 100kg-rated autonomous mobile platform for integrating robotic manipulation equipment.

For this warehouse picking project, mobility is essential because the intended operations take place at different storage locations rather than at one permanently installed workstation.

The autonomous mobile base provides the foundation for transporting the integrated robotic system to designated operating areas.

Its 100kg chassis payload rating is an important consideration when selecting the robotic arm, lifting structure, controllers, and other hardware.

However, the chassis payload should not be confused with the robot's object-handling capacity.

The integrated components contribute to the load carried by the mobile platform. Their mass, mounting positions, and effects on stability must be considered during system design.

For warehouse applications, the mobile robot's footprint, available passage space, and positioning requirements are also important.

These considerations influenced the project's focus on creating a practical mobile manipulation configuration rather than simply attaching a robotic arm to a moving platform.

2. Why Was a 7-DOF Robotic Arm Selected?

A 7-DOF robotic arm provides additional joint flexibility for approaching objects from different positions and orientations.

The selected robotic arm has the following specifications:

• Degrees of freedom: 7 DOF
• Rated payload: 5kg
• Working reach: 610mm
• Repeatability: ±0.05mm

In warehouse environments, materials may be positioned near surrounding shelves or other structures.

A robotic arm therefore needs sufficient flexibility to adjust its posture while keeping the end effector within the required working area.

Seven degrees of freedom provide additional motion-planning options compared with a more constrained arm configuration.

The 610mm working reach defines the arm's rated reach, while the ±0.05mm repeatability describes its ability to return consistently to programmed positions.

Robotic arm repeatability is not the same as complete system picking accuracy.

The performance of a mobile picking system also depends on mobile base positioning, object localization, camera calibration, end-effector configuration, and mechanical integration.

During the development discussions, the customer requested comparisons of 3kg, 5kg, and 10kg arm payload options.

The purpose was to evaluate handling requirements and equipment costs before selecting a suitable configuration.

The project ultimately selected the 5kg-rated robotic arm specification.

3. Why Does Warehouse Picking Need a Vertical Lifting Mechanism?

A vertical lifting mechanism allows a robotic arm to operate from different heights, helping address the limitations of fixed-height installations.

Storage shelves frequently contain multiple levels. A robotic arm that can access one shelf position may not be able to reach another without changing its mounting height.

To address this requirement, the project selected a lifting mechanism with:

• Effective lifting travel: 1,200mm
• Maximum lifting guide load: 25kg
• Lifting guide positioning accuracy: 1mm

The lifting system adjusts the robotic arm's vertical operating position, allowing the arm's working reach to be applied at different heights.

This is particularly relevant to applications involving multiple shelf levels.

However, the final working envelope depends on more than lifting travel.

Robotic arm geometry, mounting height, shelf dimensions, object position, and surrounding clearance all influence the positions the robot can access.

The customer also considered central arm placement, overall robot height, doorway clearance, and potential elevator access.

These requirements demonstrate why vertical lifting and mobile platform design must be evaluated together.

The objective is to balance operating reach with the physical constraints of an industrial warehouse.

How Does a Mobile Manipulator Find and Handle Warehouse Items?

A mobile manipulator uses different sensing functions for navigation and robotic manipulation: environmental perception helps the robot reach the correct location, while object perception supports target localization and grasp planning.

For the mobile platform, REEMAN's project architecture uses Laser navigation and 3D environmental sensing.

These technologies support environmental awareness, localization, path planning, and obstacle detection.

Once a robot reaches the intended working location, object manipulation requires additional information.

The system must determine the target object's location and estimate how the robotic arm should approach it.

For this reason, the customer included RGB-D depth-vision integration among the project requirements.

An RGB-D camera provides image and depth information that can be processed by computer vision software to estimate an object's position in three-dimensional space.

This information can support object recognition, grasp planning, and robotic arm motion control.

The project also explored GPU-accelerated computing for computer vision and AI-based manipulation development.

These requirements reflect an important technical distinction:

Autonomous navigation moves the robot to the task. Vision-guided manipulation helps the robot interact with the target object.

Connecting the two functions requires coordinated communication between the mobile platform, perception system, robotic arm controller, and application software.

Why Are Gripper Selection and Software Integration Important?

A mobile manipulator's ability to handle different objects depends on more than the robotic arm's rated payload.

The end effector must be appropriate for the material's weight, dimensions, shape, and surface characteristics.

During the project discussions, the customer evaluated application-specific gripper solutions, including three-finger adaptive configurations.

These options are relevant when assessing how to handle objects with different geometries, such as boxes, containers, and industrial components.

Gripper selection must also account for the weight of the end effector itself, the required gripping force, and the intended handling movement.

Alongside the mechanical requirements, software integration was a major topic.

The customer expressed interest in Python SDK access, ROS integration, drag teaching, teleoperation, and the ability to develop custom AI applications.

These development requirements influence the choice of controllers, communication protocols, software interfaces, and computing hardware.

Rather than assuming that every robotics application requires the same software configuration, the project considers software architecture according to the intended handling tasks.

For industrial system integrators, this is a significant consideration when planning future functionality or adapting a mobile manipulator to different material-handling workflows.

REEMAN Mobile Manipulator Robot: Technical Specifications

The following table summarizes the selected architecture and the main specifications associated with this warehouse picking development project.

The three load specifications describe different components.

100kg refers to the mobile chassis payload, 5kg refers to the robotic arm's rated payload, and 25kg refers to the lifting guide's maximum load rating.

These values should not be added together or interpreted as a single material-handling capacity.

The selected architecture provides a foundation for developing mobile manipulation applications with adjustable working height and task-specific end effectors.

Where Can Mobile Manipulator Robots Be Used?

The engineering principles explored in this project are relevant to warehouse automation and other industrial environments where materials must be handled at multiple operating locations.

Warehouse Shelf Picking and Retrieval

Warehouse shelf picking is the primary application objective.

A mobile manipulator can be developed to approach a designated storage location, adjust the robotic arm's operating height, identify the required item through a suitable perception system, and execute an appropriate handling task.

This architecture provides a basis for connecting material retrieval with autonomous transportation.

Industrial Parts Handling

In manufacturing facilities, small components and packaged materials often need to be handled at different workstations.

A mobile robotic arm offers an approach for combining movement between these locations with localized robotic manipulation.

The appropriate arm, gripper, and control configuration depend on component weight, geometry, and handling requirements.

Multi-Workstation Material Transfer

Manufacturing layouts frequently contain equipment positioned across different operating areas.

A mobile manipulator can support the development of workflows involving material positioning, transfer, and interaction at multiple stations.

This provides a potential alternative to installing a separate fixed robotic arm at every handling location.

Robotics Research and Application Development

For robotics developers and industrial automation integrators, a mobile manipulation platform can support work involving motion planning, perception integration, robotic arm control, task coordination, and application-specific software development.

These capabilities are relevant when evaluating how robotic systems can be adapted to different industrial workflows.

Related Robotics Video — For Reference Only

Related Robotics Video: Watch the separate robotics reference on YouTube

The following video is provided as a separate robotics reference and does not show the German customer's project. The project's technical requirements and selected system configuration are described independently in this article.

Frequently Asked Questions About Mobile Manipulator Robots

Q: What is a mobile manipulator robot for warehouse picking?

A: A mobile manipulator robot combines an autonomous mobile platform with a robotic arm to support material-handling tasks at different locations. For warehouse picking, the intended workflow involves navigating to storage shelves, positioning the arm, identifying target objects, and performing item-handling operations. The complete configuration depends on shelf dimensions, material characteristics, and the required level of automation.

Q: What is the payload capacity of the REEMAN mobile manipulator?

A: The REEMAN project uses the Big Dog Chassis platform with a rated chassis payload of 100kg. Its selected 7-DOF robotic arm has a separate rated payload of 5kg, while the vertical lifting guide has a maximum load rating of 25kg. These specifications describe different mechanical components and should not be combined.

Q: Why use a 7-DOF robotic arm instead of a fixed handling mechanism?

A: A 7-DOF robotic arm offers flexible joint positioning, allowing different approach configurations within the arm's operating range. This is useful for warehouse applications involving objects at different orientations or near surrounding structures. The selected arm provides a 610mm working reach, 5kg rated payload, and ±0.05mm repeatability.

Q: Can the robot handle items on different warehouse shelf levels?

A: The selected vertical lifting mechanism provides 1,200mm of effective travel, allowing the robotic arm's operating height to be adjusted. Combined with the arm's reach, this supports the development of handling tasks at different shelf levels. Actual reachable positions depend on robot geometry, shelf configuration, mounting height, and available clearance.

Q: How does the robot navigate and identify objects?

A: The mobile platform uses Laser navigation and 3D environmental sensing to support localization and movement through the operating environment. For manipulation tasks, the project considered RGB-D depth-camera integration to provide object and spatial information for recognition and grasp planning. These functions address different stages of the intended warehouse picking workflow.

Q: Can the mobile manipulator use different grippers?

A: The project considered application-specific end-effectors, including three-finger adaptive gripper options. Gripper selection depends on object shape, size, mass, surface conditions, and the required handling action. The gripper's own weight and mechanical requirements must also be considered alongside the robotic arm's rated payload.

Q: Can REEMAN's mobile manipulator support Python, ROS, and AI development?

A: Python SDK access, ROS integration, robotic arm control interfaces, and AI-related development were included in the project's technical discussions. GPU-accelerated computing was also explored for vision and grasp planning applications. The available development environment and supported interfaces depend on the selected controllers, computing hardware, and software configuration.

Q: What are the main engineering challenges when integrating an AMR with a robotic arm?

A: Engineers must evaluate the mobile chassis payload, robotic arm reach, integrated equipment weight, center of gravity, operating clearance, positioning, and communication between controllers. For warehouse picking, shelf geometry, end-effector design, object perception, and task sequencing also affect the system architecture.

Q: How should businesses choose a mobile manipulator robot for warehouse automation?

A: Selection should begin with the intended task rather than the robot's appearance or a single specification. Important factors include material weight, shelf height, reach requirements, aisle width, available operating space, picking method, and software integration needs. REEMAN ROBOT can evaluate these requirements when developing a suitable mobile manipulation configuration.

Q: What is the main engineering value of combining the Big Dog Chassis with a robotic arm?

A: The REEMAN Big Dog Chassis provides a 100kg-rated autonomous mobile platform for integrating robotic manipulation hardware. Combined with a 7-DOF robotic arm rated for 5kg and a vertical lifting mechanism offering 1,200mm of travel, the system brings mobility, adjustable working height, and robotic manipulation into one platform. This architecture supports the development of warehouse shelf picking, industrial component handling, and multi-workstation automation solutions, extending the capabilities of conventional AMRs beyond material transportation.

Related Technical Reading

How Does an AMR Robot Navigate a Warehouse Without AGV Rails?

From Moving Materials to Handling Them: The Next Step in Warehouse Automation

The value of mobile manipulation lies in combining movement and physical interaction within a coordinated industrial workflow.

For the German customer's warehouse picking requirement, REEMAN ROBOT selected the Big Dog Chassis platform and paired it with a 7-DOF robotic arm and vertical lifting mechanism.

The architecture addresses several practical engineering challenges, including access to different shelf heights, robotic arm positioning, navigation perception, object recognition requirements, and application-specific gripper selection.

More importantly, the project highlights how warehouse automation requirements are expanding beyond transportation alone.

A conventional AMR moves materials between locations. A mobile manipulator provides a foundation for performing additional operations after reaching the required destination.

The next step in warehouse automation is not simply moving materials more efficiently, but developing robotic systems that can interact with materials where work needs to happen.

For REEMAN ROBOT, this project reflects an engineering approach centered on real application requirements and the integration of suitable robotic technologies.

REEMAN ROBOT works with industrial automation integrators, warehouse operators, and robotics developers to explore mobile manipulation configurations tailored to specific tasks.

Planning a warehouse picking robot or an AMR with a robotic arm? Contact REEMAN ROBOT to discuss your material weights, shelf dimensions, operating clearances, end-effector requirements, and software integration needs.

REEMAN ROBOT

Beyond Transportation. Towards Intelligent Mobile Manipulation.