Robotic systems have long looked to animals for inspiration due to the fascinating natural examples of how varying body shapes may provide specialised movements and locomotion patterns. Most robots that take their inspiration from the animal kingdom are based on creatures that walk on all fours; however, some roboticists have been investigating the viability of devices with bodies that mimic those of other animals, such as snakes.
The unique qualities of snake-inspired robots may make them superior to conventional oness for certain technological tasks. Their ability to bend and slither like a snake could be useful in a variety of medical contexts, including endoscopic treatments and other minimally invasive surgical procedures where they are inserted through the nose to reach their intended locations.
Existing methods for remote control of such devices are inefficient notwithstanding their usefulness. This is due to the fact that the electronic devices used to control the movements of snake-like robots often only allow users to select six degrees of freedom, despite the fact that they are hyper-redundant (i.e., they have a huge or infinite number of degrees of freedom).
The researchers at Leibniz Universität Hannover have created a new method for intuitively and remotely controlling the motions of hyper-redundant snake robots, allowing them to overcome this barrier. The method, described in an advance copy of a paper posted on arXiv, lets users modify the behaviour of a snake-like devices while preserving as much of its original form as feasible.
SnakeTTP system gives improved locomotion for robots
This study by Tim-Lukas Habich and colleagues proposes SnakeTTP, a unified method for intuitive telemanipulation that allows for locomotion and pivot reorientation for endoscopic tasks. “Different position and orientation specifications at highest priority and shape fitting within the null space are made possible by the new approach based on task-priority inverse kinematics. Maximizing the Frechet distance between two curves, which measures their similarity, allows shape fitting to be accomplished while end effector location and orientation are simultaneously specified.”
The SnakeTTP algorithm developed by Habich and his team was tested by having 14 individuals guide a virtual snake robot to a predetermined location. Users controlling the simulated snake robot were able to effectively accomplish the locomotion challenge and re-orient the robot’s movements within a target area with minimal form change.
As comparison to the traditional method of employing the Euclidean distance between the actual and ideal positions of the links, Habich and coworkers found that the Frechet distance-based method of form fitting reduced shape inaccuracy by up to 20.1%.
Although the new control algorithm proposed by this group of academics showed encouraging outcomes in computer simulations, no actual devices have been subjected to these tests as of yet. Future experiments in real-world settings with actual snake robots could provide even more convincing evidence of its efficacy.
Researchers may be able to use the algorithm to more accurately operate snake robots and other hyper-redundant devices (such as one inspired by octopus tentacles) in the future, improving their ability to replicate the movements of real snakes and tentacles. That could pave the way for using them in healthcare settings, especially for minimally invasive surgery on the human body.
To read our article about “Belkin steps back from supporting 190 smart home devices” click here.