About the project

  • Applications: Vision Guided Robot
  • Industrial Sectors: Automotive Industry,Food Sector,Glass Industry,Mechanical Industry,Other Sectors
  • Technologies: 2D,Matrix Camera
The robots use in industrial automation has already widespread thanks to versatility and reliability offered by these devices.

System Functions

The use of robots in the industrial automation field is already by some very widespread thanks to the versatility and reliability offered by these devices. The application of the vision systems of robotic lines allows to increase the overall system flexibility by solving the cases in which the coordination “hand-eye” is essential in the robot guide towards the gripping point or the deposit point.

Specialvideo Technology, a company specializing in designing and manufacturing of vision systems in industrial applications, develops systems for robot guidance that are tailored to the application and follows the client from the studio to the implementation. The experience gained over many years of activity allows us to study long-term reliable solutions that provide real benefits on the application and not on the individual device.


The main vision system success features for a robot guide are:

  • Lighting apparatus
  • Pieces research algorithms
  • Calibration algorithms

Lighting apparatus

The lighting equipment study plays a very important role in every vision system, in particular in robot guide systems must be made in order ​​to obtain a correct pieces illumination that is compatible with the working area. The vision system often delivers a robot working area portion which must remain free from encumbrances making hard the protections installation for ambient light. In these cases we can use the stroboscopic illuminators that make possible the vision system operation even in ambient light and without caps.

Pieces research

they are being used techniques for the contours extraction or algorithms that make use of the entire piece image , with compensation for the effects due to illumination variation; for some customizations have been used precision measurements, both linear and angular.

Calibration algorithms

To ensure high precision in robot guide measurements systems, Specialvideo Technology has developed a procedure for the vision system coordinates calibration with respect to the robot coordinate axes . This procedure allows in automatic mode to:

  • calculate the pixel to mm measures conversion factors with perspective distortion correction
  • calculate the deviation (offset and angle) between the robot axes and the vision system one
  • perform the implementation of the camera witout millimetric accuracies
  • quickly restore the functionality of the system after a maintenance intervention

The vision system can be connected to any type of robot through various interfaces (I / O, RS232, Ethernet, etc..) and it can work with both stationary and moving parts by exploiting the ability of conveyor tracking of robots.

Robot guide
for car glasses

Among the applications realized we can cite the one concerning the guide robot for gripping car glasses. In this case the purpose is to pick up the glass from a conveyor line and deposit them in the pallet collection.The glasses are arrested in almost random position within the field of view, the vision system detects the coordinates of the glass and transmits them to the robot that performs the “pick and place”.

The area captured by the camera is about 1200×900 and, thanks to special algorithms calibration and perspective correction, it was possible to achieve accuracies on the deposit (including uncertainty due to the suction cup) of few tenths millimeter. In this case the deformation introduced by the objective implies an error between the center and the sides of approximately 5mm, without adequate compensation, would not permit the deposit of the pieces within the required tolerances.

Robot guide
for alloy rims

The light alloy wheels are subjected to different processing stages after casting. Normally it runs with automatic machines enslaved by robots.
The vision system identifies the wheels placed in the machining cycle, look for the point where it will be realized the valve hole and controls the rotation until the reference position.
The system can work with different types of wheels and the set-up is through a pracita and simple self-learning procedure.