The First WARA Robotics Mobile Manipulation Challenge -- Lessons Learned

TL;DR

The paper reports on the inaugural WARA Robotics Mobile Manipulation Challenge (December 2024), detailing the industrial motivation, challenge setup, and four academic approaches to autonomously navigate lab spaces and load glassware into a dishwasher. Each team presents a distinct pipeline spanning perception, grasping, and task execution, with varying degrees of integration between navigation and manipulation. The study identifies robustness and scalability challenges, such as perception in clutter, pose estimation for transparent objects, and coordinating mobile bases with manipulation, and it proposes concrete improvements for a standardized second edition. Overall, the work demonstrates tangible progress toward deployable lab-automation solutions and emphasizes the value of industry-academic collaboration in translating research to practice, while acknowledging the need for reproducibility and objective evaluation in future competitions.

Abstract

The first WARA Robotics Mobile Manipulation Challenge, held in December 2024 at ABB Corporate Research in Västerås, Sweden, addressed the automation of task-intensive and repetitive manual labor in laboratory environments - specifically the transport and cleaning of glassware. Designed in collaboration with AstraZeneca, the challenge invited academic teams to develop autonomous robotic systems capable of navigating human-populated lab spaces and performing complex manipulation tasks, such as loading items into industrial dishwashers. This paper presents an overview of the challenge setup, its industrial motivation, and the four distinct approaches proposed by the participating teams. We summarize lessons learned from this edition and propose improvements in design to enable a more effective second iteration to take place in 2025. The initiative bridges an important gap in effective academia-industry collaboration within the domain of autonomous mobile manipulation systems by promoting the development and deployment of applied robotic solutions in real-world laboratory contexts.

Figures (11)

• Figure 1: Lab Automation use-case in the AstraZeneca chemical laboratories: a human operator drives a cart around the lab to collect dirty glassware, organizes it in a dishwasher tray, and finally inserts the tray in the dishwasher.
• Figure 2: All the teams received equipment to mock-up the lab automation use-case. The equipment included a cart with 4 castor wheels, a plastic bin, a plastic base hosting 2 different types of dishwasher pins (to mimic the dishwasher tray), 3 plastic bottles, and 3 plastic beakers.
• Figure 3: (a) Robot hardware setup. (b) Behavior Tree.
• Figure 4: (a) Bottle and beaker detection and localization. (b) Opening direction detection.
• Figure 5: The setup proposed by PoliMi was built around an ABB GoFa 5 collaborative robot, equipped with a Realsense D435i stereo camera and a custom parallel gripper optimized for the pin insertion task. A detachable suction cup was also employed for the bin-picking task.