”It is our vision for the future generation of collaborative robots that they will get closer to us and be able to manipulate objects together with us as reliable partners in a harmonic and efficient way. But first of all, we need to make sure they are SAFE to us,” says Dr. Cheng Fang, Assistant Professor from SDU Robotics at the University of Southern Denmark.
Background and Motivation
Robots are good at executing manipulations accurately and capable of handling heavy objects, while humans are superior in more complicated tasks in which significant cognitive abilities are required. A powerful human-robot collaborative system can be established by tightly and appropriately combining physical capabilities of robots with humans’ intellectual advantage to handle more complicated and challenging tasks, for instance, human-robot co-assembly.
When a worker and a cobot co-manipulate a car engine and transport it to a target place, the cobot can take most of the payload while the human worker can lead the engine to the target place and perform fine adjustment of the engine position and orientation for further accurate operation, e.g. assembly. This is a typical scenario where human-robot collaboration is required.
Similar applications can be easily found in e.g., automotive, construction, and logistics industries, where fine manipulation of bulky and heavy objects is required. Robots that are capable of co-manipulating objects with human operators are called collaborative robots (or cobots for short).
Currently, there is no safety standards for this type of human-robot co-manipulations, therefore no commercial cobots in the market is certified for the human-robot collaboration, in which continuous and physical contacts between humans and robots are required. For this reason, safety and ergonomics study in physical human-robot collaboration is of paramount importance for the development, exploitation, and deployment of such type of cobots in future factories.
High-fidelity human-robot collaboration simulation for biomechanical information of human body
The RoboCluster (now a part of Odense Robotics) project “A Benchmark for the Safety and Ergonomics Evaluation of Continuous Human-Robot Collaboration Tasks” aims to provide a solution to the safety and ergonomics evaluation of aforementioned human-robot collaboration tasks.
In this project, a musculoskeletal model of human operator and a cobot model are developed and integrated into a single simulation environment, the AnyBody Modeling System, to enable high-fidelity simulations of human-robot collaboration.
Useful biomechanical information inside human body can be obtained in the simulations, such as muscle activity levels, muscle forces, joint torques. Selected information is used to evaluate the ergonomics and safety aspects of the tasks, respectively. For example, if the activity profiles that all of the involved muscles are below 15% (i.e., permanent endurance limit), the task can then be considered as a perfect ergonomic task, which means the human operator could repeat the same task forever without feeling tired.
This is the first half of “Human Collaboration” by Dr. Cheng Fang. Read the complete article here
Bio of Cheng Fang
The project coordinator of the human collaboration project, Dr. Cheng Fang, is an assistant professor from SDU Robotics section in The Maersk Mc Kinney Moller Institute at University of Southern Denmark (SDU).
About the project
The collaboration project was part of the RoboCluster network in 2020, which is now a part of Odense Robotics. It was co-financed by the Ministry of Higher Education and Science and the Region of Southern Denmark.
This article was provided by the University of Southern Denmark and was one of the projects presented at the hybrid event Robot Innovation Summit 2020.