One reason why implementing human-robot collaboration (HRC) is a challenge is the extensive background work involved. What tasks should collaborative robots take over from humans? What aspects should be considered in the process of identifying these tasks, and how do we construct natural collaboration? Niklas Land, Ph D in Informatics at the University of Skövde in collaboration with Volvo Group had an idea about how to investigate these aspects. Together, they set out to develop a virtual simulation tool - a framework in which the HRC station can be tested before physical implementation.
Before we replace humans with robots in tasks, there is much to consider. Human qualities cannot always be replaced. Even if we want to automate entire production lines, it doesn’t necessarily mean that we should. Niklas Land’s research question was “What types of tasks could be the objects of HRC and how can these be identified?”. His idea was to develop a decision and evaluation tool to answer this question using virtual reality. The proposed HRC task is set up virtually and analysed. Volvo Group provided an interesting case which supplied input for the development and evaluation of this virtual tool. Ph.D student Niklas Land.
Improve human work experience
“It’s a complex task to get HRC into a production line. The goal is always to support the human, to improve their functioning. I believe that when the set-up is done right, you can streamline the entire production flow,” says Niklas.
Traditionally, humans and robots are separated by steel bars and still today, this is the most common set-up. HRC has been a buzzword for at least 10 years but the technology is still very much in the starting blocks, and task identification is one of the main reasons for this. Niklas’ decision-making tool provides the industry with step-wise guidelines on how to develop an HRC application based on facts rather than assumptions.
A number of aspects in the framework
A number of aspects are considered in the tool. Firstly, the task and its conditions are identified, for example, is it a monotonous task, a task where robots can do the work faster or in environments where humans can be at potential risk. Secondly, the tool looks at how the task is being done today and how it could be done with HRC. The third aspect is about making sure the results can be measured. Otherwise it isn’t possible to evaluate the outcome.
“Imagine having this framework in place. Basically, all you need to do is create the virtual model of the production line and the specific task. Next, you go into virtual reality and test the set-up over and over again. You can easily move things around to make sure the robot and human are positioned optimally, as well as evaluate and practise the work sequence in a safe environment,” says Niklas.
Heavy lifting - the perfect task for a robot
The specific case at Volvo Group concerned the lifting of a valve cover onto a truck engine, a task with ergonomic problems, causing strain including shoulder injuries to humans. This station involves several different tasks where the robots were able to replace the heavy lifting task, the quality inspection, and potentially also the screwdriving task. From this perspective, it is a perfect task for a collaborative robot. After the model was created, Niklas and Volvo Group were able to try different angles and positions until an optimum set-up was identified.
“Niklas has really contributed a lot of new perspectives and knowledge when it comes to HRC. The creation of a virtual model of the station has been a powerful tool both when it comes to the set-up and also communication with stakeholders, which is an important aspect that should not be forgotten. Everything that Niklas has done and his wow (words of wisdom) will help us make decisions based on facts in future implementation phases,” says Tobias Synnergren of Volvo Group.
What next?
The next step is to continue to develop the demonstrator built at the ASSAR Industrial Innovation Arena and to evaluate the tool further using virtual reality (VR) and augmented reality (AR), after which the tool will be verified in an actual production environment. However, the focus now is on dealing with the COVID-19 pandemic and its repercussions. Discussions regarding the tool have been put on hold, but Niklas believes this tool will make a great difference when the time is ripe.
“This is a tool that helps save money, set-up time and unnecessary reworking,” concludes Niklas Land.
Comments