As this year’s Photoneo Technology & Applications Conference approaches, we mark a significant milestone in our Best Solution Awards. We received an impressive 11 submissions from our partners from all around the globe, highlighting the wide-reaching impact of our technology. We are genuinely thrilled to witness the scope of technology-leading applications of our MotionCam-3D and PhoXi 3D Scanner, spanning diverse sectors, ranging from logistics, automotive, steel, meat and wood processing to aviation.

 
Before delving in, let’s take a glimpse at all the incredible applicants: 

Company	Industry	HQ	Company Focus	Application	Company Size
BCN Vision	Meat Processing	Spain	Vision System Integrator	Optimization of the cut of pig’s legs during butchery process	51 – 200
ID Engineering	Automotive	Germany	Machine Builder	Synthetic Data Engine for car seats	11 – 50
EEP-Robotics	Logistics	Austria	Machine Builder	Package & Tote depalletization	11 – 50
Abagy	Steel	United States	Robot System Integrator	Automatic Welding	51 – 200
Ai Automation	Logistics	United Kingdom	Robot System Integrator	Re-palletizing of bottle packs	2 – 10
Comau	Automotive	Italy	Machine Builder	Helicopter blade surface automatic inspection	1001-5000
Extend Robotics	Start-up	United Kingdom	Robot System Integrator	Teleoperation	2 – 10
Machine Vision Consulting	Furniture & Wood	Italy	Vision System Itegrator	Furniture board inspection	2 – 10
Meliad	Horizontal solution	France	Machine Builder	Guidance for Laser cleaning	11 – 50
Senswork	Automotive	Germany	Vision System INtegrator	4x Scanner Inspection	11 – 50
Handalab	Automotive	South Korea	Start up	EV Charging	2 – 10

How we evaluated submissions 

During the assessment process, we focused on 5 key areas of the contest submissions: 

• Quality of the submission
• Innovation of project from technical and business perspectives
• Maturity of solution (i.e. in production, in the testing, prototype)
• Technical deployment possibilities
• How big is the market opportunity

Given the remarkable quality and volume of exceptional solutions, our panel of experts faced a challenge in selecting the top submissions. However, after careful deliberation, we are delighted to announce four finalists.

EEP-Robotics – Package & Tote depalletization

Logistics have been making waves in automation lately, as this fast-progressing segment is expected to reach over $121.3 USD by 2027 in the US alone according to Statista. And that’s exactly where our 1st Photoneo Awards finalist operates; meet EEP-Robotics with their advanced depalletization solution.

In pursuit of their latest challenge, EEP-Robotics needed to pair incredible accuracy with the “light-speed”. Let’s see how EEP-Robotics integrated our PhoXi 3D Scanners in their latest advancement.

Transforming Depalletization with Photoneo PhoXi 3D Scanner 

EEP-Robotics faced a demanding depalletization challenge from a leading global client: to automate the handling of diverse cardboard boxes and plastic load carriers, with high system availability, self-learning capability, and easy maintenance.

The solution required precise object detection and handling, including:

• Varied cardboard box sizes (128 x 62 x 65 mm to 395 x 265 x 270 mm)
• Seven types of plastic load carriers
• Cardboard layers, wooden pallet frames, and empty pallets
• Quick pallet height recognition and strap cutting
• Minimum 250 picks/hour with 10 gripper changes

EEP-Robotics leveraged PhoXi 3D Scanners and Photoneo automation software to address these complexities. The scanners’ ability to generate accurate 3D point clouds enabled:

• Reliable object detection and classification: AI algorithms processed the 3D data to identify and differentiate various objects on the pallet, even with challenging lighting conditions or partially obscured items.
• Precise picking and placement: Robot guidance was enhanced by accurate 3D object localization, ensuring efficient and safe handling even with varied object sizes.
• Self-learning and adaptability: The system could learn and improve its performance over time, adapting to new object types or variations in pallet configurations.

The collaboration between EEP-Robotics and Photoneo was key to success. Our 3D scanning expertise combined with EEP-Robotics’ automation prowess resulted in a robust, high-performance solution that met the client’s stringent requirements.

The final 2-in-1 depalletization solution showcases the power of 3D vision in complex logistics automation. EEP-Robotics successfully delivered a system with high availability, self-learning capabilities, and easy maintenance.

EEP’s 3D Vision-Guided Order Picking System: Detailed Application

EEP uses four PhoXi 3D Scanners to create an advanced order picking system for cardboard boxes and plastic load carriers. The system handles products delivered on Euro pallets, which are alternately fed to four source pallet positions. A robot then picks items from these pallets and places them into trays on a conveyor system.

The 3D vision sensors are crucial for detecting the position and orientation of boxes and load carriers, even identifying box joints and intermediate layers. These 4 sensors are mounted on a servo axis to adjust their height and ensure optimal scanning resolution for precise position detection.

The system also handles intermediate layers and can even reshuffle boxes or load carriers between source pallets. It’s designed to accommodate seven different types of load carriers and various cardboard box sizes. 

To achieve high performance, the system aims for a minimum of 250 picks per hour, with the robot equipped with ten different grippers for handling various items. Before placing items into the transport tray, the robot labels them. Additionally, the robot removes any stretch bands present on full pallet layers.

In essence, this application combines robotics and 3D vision to create a flexible and efficient order picking system capable of handling a wide range of products and packaging types.

ID Engineering – Synthetic Data Engine for car seats

“The customer’s requirement was that even the very first seat produced must be inspected at 100% and fully qualified and validated during the inspection process.”

Quickly, with no room for mistake, ID Engineering had to inspect and error-prove a wide variety of car seats, with 100% accuracy from the 1st seat that leaves the line. However, it can take days to train deep learning models to properly recognize and detect potential defects. 

Thankfully, our 2nd finalist managed to tackle this task with excellence. Instead of training their machines with real-world data, ID Engineering used a synthetic data engine that uses computational methods and simulations to create data, mimicking the statistical properties of real-world data. Let’s take a closer look. 

Deep learning struggles with the high item variety, short-term production changes, and special models, while traditional image processing is ineffective for low-contrast defects and cannot classify them for rework purposes. The entire process, including validation, would take days.

The ideal solution needed to be independent of 3D data availability, work with both 2D and 3D image capture options, and ensure 100% inspection and qualification of even the very first seat produced. With minimal training time and production disruption, this is where our MotionCam-3D steps in. 

3D Vision for Adaptive Car Seat Inspection

ID Engineering addressed the customer’s challenges by leveraging Photoneo’s 3D scanning capabilities in two innovative ways.

Firstly, when 3D CAD data was unavailable, Photoneo’s 3D reconstruction capability enabled the generation of synthetic data from scanned objects. This bypassed the reliance on pre-existing 3D models, making the solution adaptable to various car seat designs and production changes.

Secondly, Photoneo’s sensor served as a vision camera to inspect seat features requiring height detection along the Z-axis. Its ability to scan moving objects was crucial for seamless integration into the production line, ensuring inspection of every seat without slowing down the process.

So how did this collaboration come into fruition? In the words of ID Engineering:

“Photoneo’s precision, functionality, and robust SDK software were key factors in its selection. While the initial software development for synthetic data generation took several months, the overall solution implementation on the customer’s side was remarkably swift, taking only a few weeks. This highlights the efficiency and effectiveness of Photoneo’s technology in addressing complex inspection challenges in dynamic production environments.“

BCN Vision – Optimization of the pig legs cutting during butchery process

Are humans able to cut 12 kg pig legs into three even parts with deviation no bigger than 100 g? Paired with remarkable speed, machines had to take over, with a sub-2 seconds acquisition time. 

Our 3rd finalist, BCN Vision, has a client with truly challenging expectations, requiring unmatched precision and speed in the butchery process. The current butchery process faces a challenge in meeting the increasing demand for precision in cutting pig’s legs. Existing manual operations by human operators result in inconsistent cuts, leading to weight deviations that exceed the newly established customer standards. 

To overcome this challenge, BCN Vision had to automate, relying on our MotionCam-3D M+. The solution had to be capable of achieving a high level of precision, ensuring that the weight deviation in each cut does not exceed 100g from the standard 12kg pig’s leg. 

In addition to precision, the automated solution must also be highly efficient. The speed of acquisition and processing is critical to maintain the production throughput. The system needs to acquire the necessary data in under 2 seconds and reconstruct the point cloud representation of the pig’s leg rapidly. This will enable real-time decision-making and seamless integration into the existing butchery process.

Overall, the challenge lies in developing an automated solution that addresses the limitations of manual operations, providing the required precision, efficiency, and seamless integration to optimize the cutting process and meet the evolving customer demands. 

High-Speed, High-Accuracy, with MotionCam-3D 

The challenge of precise and efficient pig’s leg cutting was addressed through the implementation of an automated solution leveraging Photoneo MotionCam-3D M+ technology. Two of these advanced 3D cameras were integrated into a meat cutter machine, enabling dynamic cutting of ham legs into three parts based on real-time vision system analysis.

The process begins with an operator placing the ham on a conveyor belt. As it moves through the system, it passes under the scanning area equipped with the Photoneo cameras. A photocell triggers the cameras to capture detailed 3D point cloud data of the ham leg. BCN Vision’s proprietary application then collects and merges these point clouds, utilizing prior calibration for accurate alignment.

Subsequent processing, involving Python, C#, and Cognex VisionPro code, analyzes the merged point cloud to calculate the optimal cutting points based on the ham’s weight and volume. These cutting points are then relayed to the machine’s PLC, which controls the precise movement of the cutting disc. 

In result, this automated solution effectively overcomes the limitations of manual cutting, ensuring high precision and consistency in portioning ham legs. 

Extend Robotics – Teleoperation in Space

Meet the closing Photoneo Awards finalist, Extend Robotics who enchanted us with their innovative approach to vision-guided robotics. As the ambitious start-up that reaches (quite literally) beyond the frontier of Earth, Extend Robotics faced several challenges on their mission of in-orbit service and manufacturing. 

Extend Robotics stands as the promising UK-based robotic teleoperation and automation start-up, currently in an incubation period with the European Space Agency’s Business Incubation Centre (ESA BIC UK). This collaboration has led to an exploration project involving testing and evaluation at the Satellite Applications Catapult In-Orbit Service and Manufacturing (IOSM) Yard. 

What’s the culprit? Extend Robotics’ challenge lies in the need for in-orbit service and manufacturing of satellites, a process traditionally dependent on astronauts performing risky and costly extravehicular activities. 

While full automation is not feasible due to the unpredictable nature of space and the need for human insight, teleoperation offers a safer alternative. However, this approach demands a robust human-robotic interface that enables precise operations even in the harsh and variable lighting conditions of space.

Safer Satellite Repairs in Space: MotionCam-3D Across The Final Frontier

To mitigate safety risks and high costs associated with manual satellite repairs in space, a teleoperation solution was implemented, utilizing a robot controlled remotely by a human operator. The crux of this solution was the deployment of our MotionCam-3D Colour M+ camera to capture precise 3D and 2D data for creating an accurate digital twin of the space environment.

So how did Extend Robotics choose the right solution? In essence, the selection of the Photoneo camera was driven by its ability to deliver reliable data even in extreme lighting conditions, a critical requirement for space applications. The biggest advantage came with its structured light real-time sensor, coupled with embedded processing capabilities. The camera’s exceptional 3D data quality and competitive pricing further solidified the decision.

MotionCam-3D played a pivotal role by providing real-time, high-accuracy 3D and 2D data, essential for creating an immersive VR teleoperation experience. The real-time 3D scanning with structured light accuracy, along with the embedded processor handling most computations, streamlined the integration process. In comparison to alternatives, the MotionCam-3D offered superior real-time 3D data accuracy without the need for a powerful embedded computer.

Looking Ahead: Who’s The Photoneo Awards 2024 Winner? 

We are excited to see the scope of applications across numerous industries our solutions can provide. While all the submissions are worth special spotlight, we believe that the 3 finalists did outstanding work in pushing the boundaries of what’s possible in vision-guided robotics. 

Who’s the ultimate winner? We’ll see soon in the upcoming Technology and Applications Conference 2024 in Bratislava. We’re looking forward to the announcement!