A team of researchers from the University of Tübingen and the University of Technology Graz have printed a robot arm in 3D that can mimic the movements of an elephant’s trunk.
The FDM-printed robot is equipped with a gripper on the tip and uses machine learning to walk around and adapt to new tasks, such as picking up marbles and placing them on podiums. The design was developed as an inexpensive proof of concept and can eventually be used on an industrial production line where it is capable of performing a wider range of flexible operations, such as transporting auto parts or assembling electronic devices.
Dr Sebastian Otte, co-author of the study, told New Scientist: ‘Our dream is to be able to do this in a continuous learning setting where the robot starts without any knowledge and then tries to achieve goals, and while doing so happens it generates its own examples of learning. ”
Industry 5.0: Elephant Strain
Elephant strains are one of the best works of evolution. They are equal parts flexible and strong, and offer elephants a degree of agility that you do not often see in the animal kingdom. As a result, they are a source of inspiration for many modern bionics projects in academia, with pneumatic drives often serving as artificial muscle fibers to effect flexion and elongation.
Otte and his colleagues opted for a modular design based on a set of uniform, stackable joint modules, each with three degrees of freedom (DoF). The current design contains up to ten of these modules, but the length of the robot can reportedly be doubled by using more powerful motors.
Each segment in the boot contains several motor drives that can tilt the module up to 40 ° in two axes simultaneously. In addition to bending, the robot case can also lengthen and shorten – just like the real thing. Unfortunately, calculating the reverse kinematics for robotic actuators to perform complex operations is no easy task, much less with these many DoFs. This is where artificial intelligence comes in.
Spike neural networks for navigation
The team used the so-called spiking neural network (SNN) to control the robot, which is an artificial neural network that closely mimics natural brain processes. As well as neuronal and synaptic conditions, SNNs also incorporate the concept of time into their models. By observing a set of practice movements, the SNN was able to map motor movements at corresponding robot positions, enabling the team to ‘unroll’ the target-driven navigation models with near-millimeter precision.
The researchers write: “Not only have we shown that it is possible to construct inexpensive tribal robotic weapons with basic 3D printing equipment, but we have also shown how they can be controlled using the latest repetitive neural network architectures.”
In terms of future research, the team expressed the possibility of using radar-based distance sensors to implement collision avoidance functionality, enabling the device to work with humans. Another route might be to translate the work into a snake-like robot rather than into a stationary arm, allowing it to move around for search and rescue operations.
Further details of the study can be found in the paper entitled ‘Many-Joint Robot Arm Control with Recurrent Spiking Neural Networks’. It is co-authored by Manuel Traub, Robert Legenstein and Sebastian Otte.
Low-cost robotics is an excellent example of how 3D printing can be applied to solve abstract problems. A team of researchers from Meiji University, Tokyo, recently adapted an FDM 3D printer to create an inexpensive ‘all-in-one’ manufacturing robot. The Functgraph is capable of automatically printing and attaching tool heads to alter their active functionality, allowing users to grab, rotate, and break 3D-printed objects to assemble complex mechanical systems into a single print.
Elsewhere, scientists at Tianjin University in China have printed a customizable 3D robot capable of scaling up and monitoring pipes at industrial facilities in real time. The one-piece device features a range of soft bending mechanisms and modular grippers, which allow it to climb oddly shaped infrastructure smoothly.
Subscribe to the 3D Printing Industry Newsletter for the latest news in manufacturing additives. You can also keep in touch by following us Twitter and we like Facebook.
Looking for a career in additive manufacturing? Visit 3D printing for a selection of roles in the industry.
Displayed image shows the hull robot with full bending. Photo via University of Tübingen.