Hagyományos kamerák okosítása mobilis robotokhoz (KUKA Robotics)

Dr. Kiss Bálint
Kölső konzuelns:
Komlósi István (KUKA Robotics)
External supervisor e-mail:
Önálló laboratórium 1 - Irányítórendszerek főspecializáció, MSc Vill.
Önálló laboratórium 2 - Irányítórendszerek főspecializáció, MSc Vill.
Önálló laboratórium 1 - Vizuális informatika főspecializáció, MSc Info.
Önálló laboratórium 2 - Vizuális informatika főspecializáció, MSc Info.
Önálló laboratórium - Szoftverfejlesztés és rendszertervezés specializáció, BSc Info.
Önálló laboratórium - Irányítórendszerek ágazat, BSc Vill.
Hallgatói létszám:
Szakdolgozat / Diplomaterv
TDK dolgozat

 Non-smart Cameras acting as Smart Cameras in Industrial Mobile Robotic Applications

In industrial robotic applications in some areas of production where dust and moisture is present relying only on laser based positioning might not ensure millimeter accuracy that is required for the automation process, e.g. when driving into cells where load must be picked up or put down. Hence, robots are often equipped with cameras that read markers contributing to fine positioning and position verification. Industrial cameras providing out of the box marker detection are usually more expensive than normal cameras by a large factor.

The goal of this project is to provide a framework to integrate non-smart cameras (cameras without configurable marker detection) to allow marker detection for fine positioning and position verification. Via this software solution customers could integrate other cameras to their system and provide accurate positioning or position verification based on marker features. 

Use Cases

Position Verification
  • Robot drives to position based on laser navigation
  • Camera detects and reads tag to verify that the position is correct
Fine Positioning
  • Based on extrinsic and intrinsic camera calibration the position and the offset of a tag to the robot can be calculated
  • The offset is used to do a fine movement to minimize the error of the position, e.g. at positions where load should be picked up
Work Packages for students
Step 1:
  • Create a microservice that can interface USB or SICK cameras and detect tags in the image, e.g. with OpenCV
  • Provide the data through a REST API that can be integrated into the robot system’s software architecture
Step 2:
Create mechanisms to calibrate cameras intrinsically and extrinsically to allow accurate position detection
With this project a student can
  • gain knowledge and deep insight to industrial mobile robot systems
  • gain knowledge in camera systems used for robotic applications
  • gain knowledge of mobile robot localization and fine positioning
  • learn how to integrate an external system to an already existing robotic solutions
  • deepen programming knowledge and get acquainted with software technologies used for industrial mobile robot systems
  • acquire knowledge in image processing