Collaborative Robots (Cobot) is a hot topic in both of industry and academia. The development of collaborative robots is also identified as one of the most important and preferential research directions in Denmark.
In industry, cobots are employed as intelligent partners, which are able to work closely together with human workers to collaboratively perform complicated tasks in production lines. The safe and harmonic co-existence of cobots and human coworkers in a shared workspace without any safety guards is implemented by low-level control mechanisms, which impose kinematic and dynamic constraints (speed and force) and protective stops on cobots in risky cases. This safety strategy aims to make cobots avoid unintentional contacts with humans or reduce the risk of harming humans, when intentional contacts are required. Therefore, the nature of this strategy can only enable cobots and humans to work collaboratively in sequence for a series of process steps in a complicated task without continuous contact with each other (occasional contact may be needed, e.g., handover operation). Accordingly, continuous human-robot collaboration tasks have rarely been seen in industry even though typical scenarios such as, human-robot co-transportation and co-assembly tasks, have been discussed extensively. One of the important and intractable issues is that the safety and ergonomics cannot be verified and validated in continuous human-robot collaboration tasks.
In academia, more and more attention is projected into the continuous human-robot collaboration area, where active contacts between humans and cobots are encouraged, since humans' significant cognitive abilities in learning and adaptation to various task demands, disturbances and uncertainties, and the Cobots' superior physical capabilities can be combined to make the collaboration solutions sufficiently flexible and adaptive. For instance, when a human coworker and a cobot co-manipulate one common heavy object and transport it to a target place, the cobot can share most of the payload and the human worker can be responsible for leading the object to the target place and performing fine tuning in terms of the position and orientation of the object for further accurate operation, e.g. assembly, in a crucial stage. This is a typical scenario where continuous human-robot collaboration is necessarily required. Many researchers deal with the safety and ergonomics issues for this type of collaboration tasks case by case, however, there is no systematic framework for the verification and validation of proposed continuous human robot collaboration tasks.
In this proposal, the aim is to establish a benchmark by developing an integrated simulation environment and physical work cell to verify and validate continuous human-robot collaboration tasks in terms of safety and ergonomics. In the physical work cell, the planned coordinated motions of human and robot in a collaboration task can be measured and recorded, and fed into an integrated simulator, which combines the dynamic model of the robot and the musculoskeletal model of the human, important biomechanical properties of human body, for instance, muscle forces, joint torques and joint compression forces, will be calculated in the simulation to evaluate the safety and ergonomics of the proposed collaboration task. UR robots will be used as Cobots in this project.
