SYNCRO
RESEARCH
PHYSICAL AI
5 mins
|
23 June 2026
Earlier this year, a new robotics research lab opened at NAMTECH in Gandhinagar. Addverb helped set it up — not just with a couple of demonstrator units, but with a working facility built around two real industrial-grade systems: an autonomous mobile robot and a collaborative manipulator arm, wired into the same infrastructure students would use to study them.
The first project to come out of that lab is the kind that justifies its existence.
Two students — Martha Surya Teja and Maganti Hemanth — took on a question that sits at the heart of modern robotics: what does it take to turn two separate robots, each excellent at one thing, into a single system that does both? They picked the Addverb Dynamo 200 — a wheeled autonomous mobile robot built for warehouse navigation — and the Addverb Syncro 5 — a six-axis collaborative arm built for precise manipulation — and set out to make them behave as one.
The result is a fully integrated mobile manipulator: a robot that can drive itself across a floor, stop precisely at a waypoint, and then pick something up and put it somewhere else. Two products. One network. One coordinated motion sequence.
Two Robots. One System.
Mobile manipulation is one of those robotics problems that looks obvious in concept and turns out to be a serious systems-integration challenge in practice. The arm and the base are designed independently — different power requirements, different control software, different communication protocols, different ideas about what "ready to move" even means. Getting them to behave like one body is the work.
Martha and Hemanth approached it the way a real engineering team would. They iterated mechanical designs in CAD to find a structure that wouldn't tip, vibrate, or pinch a cable. They built a separate power subsystem for the arm so the AMR's drive motors wouldn't introduce electrical noise into the cobot controller. They put the two robots on a single dedicated network with static IP allocation, so the Addverb Fleet Management system and the ROS 2-based cobot stack could talk without stepping on each other.
Then they wrote the teleoperation layer themselves — a custom ROS 2 package that turns joystick inputs into real-time Cartesian motion commands for the arm. The Dynamo 200 handles navigation through Fleet Management and the Addverb Visual Programming Language (VPL); the Syncro 5 handles manipulation through ROS 2 controllers. The operator drives the whole thing as a single machine.
It's not glamorous work. But it's exactly the kind of Physical AI integration required to build the systems that increasingly populate warehouses, hospitals, labs, and — eventually — humanoid platforms.
Why Mobile Manipulation Matters
Almost every robot in commercial use today does one of two things. It moves (think AMRs, AGVs, delivery robots). Or it manipulates (think cobots, industrial arms, pick-and-place stations). Combining the two on one platform unlocks an entirely different category of capability — and is the technical foundation underneath the humanoid platforms now being announced almost monthly by the world's largest companies.
A mobile manipulator can run a sequence that a fixed arm can't: drive to a shelf, pick up a part, drive to a workstation, place the part, drive back. That sequence describes most of the actual physical labour inside a warehouse, a kitchen, or a workshop. Each individual step is solved technology. Stitching them together reliably, under one operator interface, is where the real work begins — and where the next generation of robotics graduates will find themselves spending most of their careers.
There's also a quieter signal in this project. The integration here wasn't a clever new algorithm — it was mechanical design, electrical isolation, network configuration, and software interfacing. That's the unglamorous half of robotics, the half rarely shown in research papers, and the half that determines whether a system actually works on a real floor. Teaching students to take that half seriously — early, hands-on, on real industrial hardware — is what separates a robotics lab from a robotics classroom.
Inside the Lab That Made This Possible
This project was built inside the Addverb × NAMTECH Centre of Excellence (CoE) — a robotics facility designed to give students hands-on access to the same industrial-grade hardware running in production warehouses. The lab is the reason the project exists in the form it does: an AMR, a cobot, and the networking infrastructure to make them coordinate, all in one room.
For the full picture of how the CoE is set up, the philosophy behind it, and what it means for India's automation talent pipeline, read the dedicated piece: Training India's Next-Gen Automation Engineers: Inside the Addverb × NAMTECH Robotics Lab.
Where Syncro 5 and Dynamo 200 Fit
This is the kind of work the Addverb Syncro 5 cobot and the Addverb Dynamo 200 AMR were designed to support — both as deployed industrial products and as research surfaces. Across the Addverb Academic Series so far, we've seen the Syncro 5 used for vision-based motion retargeting at IISc Bangalore, bimanual handover research at IIT Gandhinagar, LLM-driven autonomy and gesture control at IIT Indore — and now, integrated with the Dynamo 200, used for mobile manipulation at NAMTECH.
Each project is different. The platforms underneath are the same.
Build on the Same Platform
If you're a faculty member, researcher, or student building on the Addverb Syncro 5 cobot or the Dynamo 200 AMR — or thinking about setting up similar lab capacity at your institution — we want to hear from you.
Write to us at [email protected], and explore open-source libraries, project templates, and academic resources at community.addverb.ai.
Technical Appendix: Architecture & Implementation Details
This section is intended for readers interested in the technical depth of the project.
Project Overview
Goal: integrate the Addverb Dynamo 200 AMR and Addverb Syncro 5 cobot into a unified mobile manipulator platform
Operating mode: autonomous navigation by the AMR, joystick-based teleoperation of the manipulator
Application domain: material handling, pick-and-place, warehouse automation, flexible manufacturing
Mechanical Design
Multiple design iterations developed in Autodesk Fusion 360 to evaluate rigidity, centre of gravity, workspace accessibility, and cable routing
Final structure: four-post 40×40 mm T-slot aluminium extrusion frame with corner brackets and fasteners
Dynamo mounting platform: laser-cut acrylic plate, designed for the integration assembly
Stable mounting for the Syncro 5 with efficient cable management
Electrical Integration
Syncro 5 powered through an independent UPS-based power subsystem, isolated from the Dynamo 200's traction battery
Architecture minimises electrical interference from drive motors to the cobot controller, onboard computer, and networking hardware
Externally mounted emergency-stop system for operational safety
Communication and Control Architecture
Dedicated router establishes a unified communication network for both systems
Dynamo 200 connected wirelessly; Syncro 5 connected over Ethernet
Static IP allocation for deterministic communication and simultaneous system access
Reliable interoperability between Addverb Fleet Management (for the AMR) and the ROS 2-based cobot control framework (for the manipulator)
Navigation Framework (Dynamo 200)
Localisation and mapping driven by the Livox MID-360 LiDAR
Environment maps, waypoints, travel corridors, and operational zones configured through Fleet Management
Mission execution workflows authored in Addverb Visual Programming Language (VPL) using Move, Wait, Set, and If blocks
Repeatable sequences including closed-loop waypoint traversal and autonomous route execution
Manipulation Framework (Syncro 5)
ROS 2 Humble software stack on the Syncro 5 platform
Custom joystick teleoperation package for real-time Cartesian control of the arm and gripper
Motion commands processed through ROS 2 nodes
Executed via the Cartesian Jogging Controller for intuitive, precise manipulation
Hardware Platform
Addverb Dynamo 200 — autonomous mobile robot, 200 kg payload class
Addverb Syncro 5 — 6-DOF industrial collaborative robot
Livox MID-360 LiDAR — for AMR localisation and mapping
Custom integration frame: T-slot aluminium extrusion + laser-cut acrylic mounting
Software Stack
AMR side: Addverb Fleet Management, Addverb Visual Programming Language (VPL)
Cobot side: ROS 2 Humble, Docker-based deployment, joint state broadcasting, gripper control, Cartesian jogging controller
Integration layer: Custom ROS 2 joystick teleoperation package
Design: Autodesk Fusion 360
Validation
Full integrated mobile manipulator platform validated within a warehouse laboratory environment at NAMTECH
Mechanical, electrical, communication, navigation, and manipulation subsystems operated reliably throughout testing
Successful demonstrations of autonomous navigation, waypoint execution, manipulator teleoperation, and coordinated pick-and-place workflows
Roadmap
Transition from teleoperated manipulation to fully autonomous task execution
Vision-based object detection and pose estimation
Event-driven synchronization between navigation and manipulation tasks
Autonomous motion planning via MoveIt 2
