A joint paper by the CRL and the Institute of Robotics and Process Control at the Technische Universität in Braunschweig, Germany has recently been published. This marks an important step in research that had begun 3 years ago, in 2017.
This paper was published on the 29th of October 2020 by the Institute of Electrical and Electronics Engineers (IEEE) in their journal, IEEE Transactions on Medical Robotics and Bionics.
The CRL’s Jessica Burgner-Kahrs and Sven Lilge collaborated with Heiko Donat and Jochen J. Steil on research to estimate external forces applied to the tip of a concentric tube continuum robot (a robot consisting of pre-curved concentrically nested tubes) .
This was made possible through data sets provided by the CRL and the use of an artificial intelligence known as the Deep Direct Cascade Learning (DDCL), which was the German Lab’s specialty. The DDLC network used the dataset to analyze the change in shape that the robot underwent when external forces were applied to it. With this information, the network was able to estimate the quantity of the external forces.
The results were promising with the estimations being accurate on smaller scales but becoming less reliable with more complex machines.
This has implications for many situations that continuum robots can be used in, such as, in surgery and in tight-space rescue operations where external forces both on the robot and on its environment can mean the difference between success and failure. With this article and further research, we’re getting closer to continuum robots being very reliable in situations that require extreme precision and care.