The Delft Mercurians team has been together since 2022 and competes in the RoboCup Soccer League. To improve their performance during the RoboCup competition, two industrial cameras from VA Imaging were installed in their test field. The student team implemented two 1.6MP high-speed cameras “MER2-160-75GC-P” with the SONY IMX273 sensor.

The RoboCup Competition

The RoboCup is the world championship for robotics where teams compete in many different challenges to push robotics and AI related innovation and research. The Delft Mercurians team has its focus on the RoboCup Small-sized Soccer League, in which they compete against other student teams from all over the world. The goal for the students is to create robots that can play soccer autonomously. Every year, the competition is being held at a different location. With a total of 24 team members, the international student team aims to compete in the RoboCup competition in Eindhoven in 2024.

The mechanical aspects of every robot must comply with the requirements set by the competition committee. Many electronic components are involved to ensure the robots drive smoothly and reliably. The strategies designed on the software side must be smart and adaptive to actually beat the enemy teams. Many different challenges are posed across different fields of engineering, making the competition multi-discipline and highly challenging.

Sponsorship

Before the Delft Mercurians team competes in the RoboCup competition in Eindhoven later this year, preparation is done with many hours of testing. In their internal testing environment, the two camera set-up’s from VA Imaging are used as a vision system. Multiple robots are positioned in the test field, which is comparable to the actual soccer field during the RoboCup Soccer League. Determining the position of every single robot on the field is key to playing flawless soccer. 

To define the exact position and movement of a robot, two of the cameras are mounted in the ceiling of the test area. The utilization of industrial cameras is crucial when addressing challenges related to positioning, localization and speed. The selected two high-speed cameras cover a large area of approximately 8.1m wide, thanks to their wide-angle M12 lenses. 

The team incorporated a vision system that includes the MER2-160-75GC-P camera, featuring the Sony IMX273 sensor. It is worth mentioning that the team selected a camera of the ‘Mercury’ 2 series for their vision system. It’s a coincidence that their team name, The Delft ‘Mercurians’ and the camera series share the Mercury theme. 

The vision hardware

The vision system in the test area includes the following machine vision components:

• MER2-160-75GC-P: This is a 1.6MP GigE camera with a C-mount, capable of capturing 75 frames per second at full resolution.
• LM12-5MP-03MM-F2.8-2-ND1: This is a 3.3MM M12-mount lens for capturing images with a non-distortion rate of < 1% and horizontal field of view of 73°.
• LADAP-C-TO-M12-V2: This is a C- to M12-mount adapter to use the M12 lens in combination with the 1.6MP C-mount camera.

An overview of all computer vision products is shown on the website. The MER2-160-75GC-P camera, equipped with the SONY IMX273 sensor, provides 75 images per second at 1.6MP, ensuring minimal motion blur. The global shutter freezes images without artifacts. That is why a global shutter camera captures high-speed moving objects without distortion.

Testing of the camera set-up

Upon receiving the cameras, the Delft Mercurians were impressed, especially compared to the previous unprofessional set-up that was used. The documentation provided by VA Imaging enabled them to start testing instantly without asking for support and implement adaptions to their software. Ever since that first moment, the team has made significant progress in testing and development. Their approach to testing, combined with the two machine vision cameras, ensured a robust testing solution and well preparation for the RoboCup competition.

Report card

Image quality: 
The Delft Mercurians team was pleased with the combination of the high frame rate and resolution of the camera. Despite the low focal length of the lens (3.3MM), no distortion was seen in the images. This has been a crucial factor calculating the robots’ positions.

Implementation process: 
The team rates the image quality 4 stars out of 5. They were particularly satisfied with the documentation provided by VA Imaging. Although not all specifications were immediately clear, eventually, all the information they needed was available. They found the SDK easy to use and straightforward, as they were able to acquire images on the first try.

Durability: 
Since receiving the cameras, the team has used them daily. They have never experienced any error messages or problems with the cameras and appreciate their reliability.

Support: 
The VA Imaging team has been a great help. From the start, by assisting the students with hardware selection, to supplying the required products, everything was as desired.

If you need assistance selecting the right hardware for your vision application, please provide a short explanation of your vision project. We are happy to help you find the most cost-effective and high-quality vision setup.

Make sure to follow Delft Mercurians via this link on LinkedIn or visit their website if you would like to keep up to date with the latest achievements of the student team. More information about the Robotics Student Association (RSA) can be found by clicking on this link.