3D imaging for steel profile cutting

Since 1985 HGG is a world leader in designing and building robotic cutting solutions for the heavy steel industry. Its machines are deployed all over the world and are used to rapidly cut steel profiles into free-form shapes. Its machines have been used to build iconic landmarks world-wide such as the London Eye, the Amsterdam Arena and other huge stadiums. Ever increasing demands however on the machines productivity and difficulties in finding skilled personnel, are driving a demand for more autonomy and higher accuracy.

Assignment description

One of the many challenges faced when trying to increase productivity and autonomy of steel cutting machines is accurately measuring the actual shape of the profile to be cut. This measurement allows the planned cuts to be adjusted in order to compensate for the deviations in the base material. The goal of the assignment is to design and implement an imaging solution that can measure and locate the shape of the profile in 3D very rapidly, by combining state-of-the -art object recognition algorithms with high precision vision-sensors.

  • Design a 3D imaging solution to measure a part in mere seconds with 0.1 mm accuracy
  • Develop algorithms for profile model fitting
  • Test the solution in the real world

Why is this hard!

To achieve good cutting results, it is vital that the position, orientation and actual shape of the profile to be cut are known very accurately. At the same time, these profiles can vary greatly in size and shape. Developing a solution that meets the criteria will require creative use of state-of-the-art perception algorithms together with a high-quality sensor setup.

  • High dynamic range
  • Little time
  • Bad environment

Who are we looking for

We are looking for an enthusiastic self-assertive machine vision adept that wants to take novel perception schemes onto the industrial work floor.

  • Desire to put advanced perception strategies into real world applications
  • Strong background in computer vision
  • Experience with point cloud libraries such as PCL
  • Affinity with robotics

Company supervision

HGG will provide a challenging assignment and the support needed to make it a success. The daily supervision will fall under Matthijs Jansen, graduated at Systems and Control student (Delft University of Technology) and R&D engineer at HGG.


Want to know more? Send your CV and questions to Matthijs Jansen at maj@hgg.nl.
Matthijs Jansen
R&D Engineer
T +31(0)227 50 40 30 | F +31(0)227 50 19 03 | www.hgg-group.com