A team of researchers from the University of Tübingen and the University of Technology in Graz have 3D printed a robot arm capable of mimicking the movements of an elephant’s trunk.
Equipped with a snap-on device, the FDM printed robot uses machine learning to travel and adapt to new tasks, such as lifting marble and placing them on podiums. Developed as proof of the low-cost concept, the design could eventually be used on an industrial production line where it would be able to perform a wider variety of flexible operations, such as transporting car parts or assembling devices. electronic.
Dr. Sebastian Otte, a co-author of the study, told New Scientist: “Our dream is that we can do this in a lifelong learning system where the robot starts without any knowledge and then tries to achieve goals and, while it does, it generates its own examples of learning. ”
Industry 5.0: elephant trunks
Elephant trunks are one of the best works of evolution. They are equally flexible and strong parts and give elephants a level of dexterity that you don’t often see in the animal kingdom. As a result, they are a source of inspiration for many modern bionic projects in academia, with pneumatic actuators that often act as artificial muscle fibers to achieve bending and extension.
Otte and his colleagues opted for a modular design, which is based on a set of uniform, stackable common modules with three degrees of freedom (DoF) each. The current design includes up to ten of these modules, but the length of the robot can be doubled with the use of more powerful motors.
Each segment of the boot houses several gear-driven motors that can tilt the module up to 40 ° in two axes simultaneously. In addition to bending, the robotic trunk is also able to lengthen and shorten – just like the real thing. Unfortunately, calculating the reverse kinematics for robotic actuators to perform complex operations is not an easy task, much less with so many DoFs. This is where artificial intelligence comes in.
Spiking neural networks for navigation
The team used what is called a spiking neural network (SNN) to control the robot, which is an artificial neural network that closely mimics the natural processes of the brain. In addition to incorporating neural and synaptic states, SNNs also include the concept of time in their models. Observing a set of training movements, SNN was able to map the movements of the engine to the appropriate positions of the robot, allowing the team to “unroll” target-oriented navigation models with an accuracy of almost a millimeter.
The researchers write: “Not only have we shown that it is possible to build low-cost robotic trunk arms with basic 3D printing equipment, but we have also demonstrated how they can be controlled using the latest recurring neural network architectures. . ”
Regarding future research, the team expressed the possibility of incorporating radar-based distance sensors to implement collision avoidance functionality, allowing the device to work in tandem with humans. Another way could be to translate the work into a snake-type robot, rather than a stationary arm, through which it could “slide” around it for search and rescue operations.
More details of the study can be found in the paper entitled “Control of multi-joint robot arms with recurrent spike neural networks.” He is co-author of Manuel Traub, Robert Legenstein and Sebastian Otte.
Cheap robotics is a prime example of how 3D printing can be applied to solve abstract problems. A team of researchers from Meiji University, Tokyo, recently customized a 3D FDM printer to create a low-cost “all-in-one” manufacturing robot. Functgraph is able to automatically print and attach custom tool heads to change their active functionality, allowing users to catch, rotate and break 3D printed objects to assemble complex mechanical systems into a single print job.
Elsewhere, scientists at Tianjin University in China have previously printed a customizable 3D robot capable of scaling and monitoring pipelines from industrial facilities in real time. The one-piece device features a series of soft bending mechanisms and modular pliers, allowing it to flexibly climb the strangely shaped infrastructure.
Subscribe to 3D printing industry newsletter for the latest additive manufacturing news. You can also stay connected by watching us on Twitter and we like it Facebook.
Are you looking for a career in additive manufacturing? Visit 3D printing jobs for a selection of roles in the industry.
The image shown shows the trunk robot at full flexion. Photo by the University of Tübingen.