Collaborative Assembly with Vision-Based Safety System

Name of demonstration

Collaborative Assembly with Vision-Based Safety System

Main objective

Demonstrating capabilities of vision-based safety system with projector and AR interface. Safety violations are captured through changes in 3D point map of working environment.

Short description

Demonstration of a vision-based safety system for human-robot collaboration in the assembly of diesel engine components. A dynamic 3D map of the working environment (robot, components + human) is continuously updated by depth sensor and utilized for safety and interaction between resources with virtual GUI. Robot’s working zone is augmented for the user to provide awareness of safety violation. Virtual GUI aims to provide instructions of the assembly sequence and map proper UI as the controller of the system. ​ 

Owner of the demonstrator

Tampere University

Responsible person

Project Researcher Dmitri Monakhov 


C29.3 - Manufacture of parts and accessories for motor vehicles


Robotics , Safety, Augmented Reality , Manufacturing, Human-Robot Collaboration.

Potential users

SME’s, Research Institute and Academy

Benefits for the users

Increasing safety awareness: Using vision-based safety system can decrease operator violation of robot safety zones by decreasing the possibility of human error As a result, this can lead to reducing production cycle time where a human doesn’t stop production line by these mistakes. Here, proper GUI notifies human with violating robot working zone and assist human to avoid this error. Based on user test and survey, both scenarios vision-based safety system compared to the baseline where the robot was not moving in the same workspace with an operator. The result of this experience can be explained by measuring of robot idle time and total execution time.

Improving user experience: the projector/AR interface can be used to provide instructions on how to execute the current task.  Research have shown that tasks performed with the help of the safety system can be completed 21-24% faster. Robot idle time is reduced by 57-64%.


This use-case utilizes projection-based and augmented reality head-mounted display technologies to overcome human error in the assembly sequence. It’s worth to mention, depth sensors such as Kinect track human body part whenever violation could happen. This is a new approach for collaborative work cell and provides a concept of safety vision-based for the operator. This approach studied in the academy research laboratory and 20 participants participated in the survey, where questionary contains 6 categories as following:  Safety, Information Processing, Ergonomics, Autonomy, Competence, and Relatedness.

Risks and limitations

Software Malfunction: As the whole system operates by a computer, there are chances that software and connections through devices get malfunctioned. These can endanger the operator’s safety while the system stopped. As a result, it requires a back-up safety system is running all the time and malfunctioning this system results in protective stop of robot system. 


Environmental disturbances: lighting conditions, dust electrical interferences can affect the depth sensor information, which can cause mistakes when estimating changes in the environment. In case of AR HMD’s field of view limitation should be encountered


Limitations of depth sensor: Scaling system to bigger environments with one depth sensor can cause point cloud to be not dense enough for accurate estimation. Synchronization between multiple depth sensors is needed.

Technology readiness level

6 - Safety approved sensors and systems are commercially available

Sectors of application

Manufacturing assembly, Quality Assurance of assembled components.

Potential sectors of application

Generation of safety zones in public environment

Hardware / Software


UR5 from Universal Robot family (tested on CB3 and UR software 3.5.4)

Standard 3LCD projector



ROS Melodic

OpenCV (2.4.x, using the one from the official Ubuntu repositories is recommended)

PCL (1.8.x, using the one from the official Ubuntu repositories is recommended)

ROS Interface to the Kinect One (Kinect v2)


Collaborative assembly with vision based safety system, a TRINITY use case

To learn more about the solution, click on the link below to access the training on the Moodle platform
Collaborative assembly with vision-based safety system

Back to LIST
Subscribe to newsletter

Please subscribe to our mailing list to be the first one informed about the new open calls, events and other hot news!

    Please check if the email you entered is correct
    All done!

    Your have successfully subscribed to Trinity newsletter.