RobDREAM · Project

Self-Optimizing Industrial Robots That Improve Overnight Without Reprogramming

manufacturingPilotedTRL 6

You know how a good night's sleep helps you solve problems better the next day? This project applied that same idea to industrial robots. During downtime, the robots replay their work experiences and run simulations of future tasks — essentially "dreaming" — to automatically fine-tune how they move, grip objects, and navigate factory floors. The result is a mobile robot arm that gets better at its job every day without an engineer having to manually reprogram it.

By the numbers

EUR 5,401,911

EU funding for development

consortium partners

countries involved (AT, DE, ES, SE)

total project deliverables

demo deliverables including realistic manufacturing environment

The business problem

What needed solving

Manufacturers using mobile robot arms waste significant time and money on manual programming, parameter tuning, and reconfiguration every time production changes. Each new product variant, workspace layout, or grasping task means an engineer must manually adjust the robot — leading to costly downtime and slow adaptation. This is especially painful for SMEs with high product variety and small batch sizes.

The solution

What was built

The project built a realistic manufacturing environment for testing, a mobile manipulation test board for evaluating robot tasks, and a final algorithm configuration system that automatically optimizes robot perception, navigation, and grasping. These were validated through iterative real-world testing in manufacturing case studies across 17 deliverables.

Audience

Who needs this

Automotive and electronics manufacturers with mobile robot fleets needing frequent task changesWarehouse and logistics operators deploying mobile picking robotsRobot system integrators spending too much time on commissioning and parameter tuningSME manufacturers with high product variety and small lot sizesIndustrial robotics OEMs looking to add self-optimization features to their platforms

Business applications

Who can put this to work

Automotive & Discrete Manufacturing

enterprise

Target: Mid-to-large manufacturers with mobile robot fleets handling varied assembly or logistics tasks

If you are a manufacturer dealing with frequent product changeovers and spending days reprogramming mobile manipulators for each new batch — this project developed a self-optimization system that lets robots automatically tune their perception, navigation, and grasping during idle hours. The system was tested in a realistic manufacturing environment built by KUKA, targeting SME-like settings with higher product variety and smaller lot-sizes.

Logistics & Warehousing

mid-size

Target: Warehouse operators deploying mobile picking and manipulation robots

If you are a logistics company struggling with mobile robots that fumble unfamiliar packages or navigate poorly in changing warehouse layouts — this project built a mobile manipulation test board and algorithm configuration system that automatically improves object detection and grasping strategies. The 7-partner consortium including KUKA tested these across 3 demo environments to reduce setup and parameter tuning costs.

Professional Service Robotics

any

Target: Robot integrators and OEMs building mobile manipulators for industrial clients

If you are a robotics integrator losing margin on lengthy commissioning and parameter tuning for each customer site — this project delivered a final algorithm configuration system with improvements validated through manufacturing case studies. With EUR 5,401,911 in EU funding and 17 deliverables, the technology targets reduced Total Cost of Ownership for robot deployment and faster adaptation to new tasks and environments.

Frequently asked

Quick answers

What would it cost to adopt this technology?

The project was funded with EUR 5,401,911 across 7 partners over 3 years. The technology is aimed at reducing Total Cost of Ownership for mobile manipulator setup, programming, and parameter tuning. Licensing or integration costs would need to be discussed directly with the consortium, particularly KUKA Deutschland GmbH as coordinator.

Can this work at industrial scale in a real factory?

Yes — the consortium built and tested in a realistic manufacturing environment that replicates typical manufacturing setups. The project explicitly targeted increasing Technology Readiness Levels through frequent and iterative real-world testing and validation from the very beginning of the project.

Who owns the IP and can I license it?

The project was coordinated by KUKA Deutschland GmbH, a major industrial robot manufacturer. IP is distributed among 7 partners across 4 countries (AT, DE, ES, SE). Licensing arrangements would need to be negotiated with the relevant consortium members depending on which components you need.

How quickly can a robot adapt to a new task with this system?

The objective states that mobile manipulation systems will adapt more quickly to new tasks, jobs, parts, and areas of operation. The automatic optimization runs during the robot's inactive phases, meaning adaptation happens overnight without taking the robot offline during production hours.

Does this work only with KUKA robots or any mobile manipulator?

The project was developed with KUKA as lead partner and tested on their mobile manipulators. The algorithm configuration system and optimization methods are designed as a portfolio of key algorithms for perception, navigation, and manipulation. Based on available project data, integration with non-KUKA platforms would require discussion with the consortium.

What specific robot capabilities does this improve?

The system improves three core capabilities: perception (recognizing objects and environments), navigation (moving through factory floors), and manipulation and grasping (picking up and handling parts). These improvements come through automatic parameter optimization, strategy selection, and learning from simulated scenarios.

Consortium

Who built it

The consortium is led by KUKA Deutschland GmbH, one of the world's top industrial robot manufacturers — a strong signal that this technology was developed with real commercial intent, not just academic interest. The 7-partner team across 4 countries (Austria, Germany, Spain, Sweden) blends 2 industrial players with 3 universities and 2 research organizations, giving it both scientific depth and market pull. The 29% industry ratio and KUKA's coordinator role mean results were continuously tested against real manufacturing requirements. One SME partner adds agility. This is a well-balanced consortium where the end-user (KUKA) drove the development priorities.

KUKA DEUTSCHLAND GMBHCoordinator · DE
KUNGLIGA TEKNISKA HOEGSKOLANparticipant · SE
FUNDACION TECNALIA RESEARCH & INNOVATIONparticipant · ES
ALBERT-LUDWIGS-UNIVERSITAET FREIBURGparticipant · DE
DEUTSCHES ZENTRUM FUR LUFT - UND RAUMFAHRT EVparticipant · DE
RHEINISCHE FRIEDRICH-WILHELMS-UNIVERSITAT BONNparticipant · DE
Convergent Information Technologies GmbHparticipant · AT

How to reach the team

KUKA Deutschland GmbH (Germany) — contact via SciTransfer for a warm introduction to the project team

Order a report on this consortium See KUKA DEUTSCHLAND GMBH profile →

Next steps

Talk to the team behind this work.

Want to explore how self-optimizing robot technology could reduce your commissioning costs and improve robot performance? SciTransfer can connect you directly with the RobDREAM team and provide a tailored briefing for your use case.

Order a report on this topic More in Manufacturing & Industry 4.0