Designing a soft robot to move organically - to bend like a finger or twist like a wrist - has always been a process of trial and error. Now, researchers have developed a method to automatically design soft actuators based on the desired movement.
Most robots achieve grasping and tactile sensing through motorized means, which can be excessively bulky and rigid. A group of researchers has devised a way for a soft robot to feel its surroundings internally, in much the same way humans do.
The 2015 earthquake in Nepal is just one example of a recent natural disaster that required search and rescue operations. A European security research project has developed new robotic tools for such operations with great success. Some of its developments are already commercially available and ready to be deployed in order to save lives.
A new machine learning training method enables neural networks to learn directly from human-defined rules, opening new possibilities for artificial intelligence in fields from medical diagnostics to self-driving cars.
Developed by a team at the University of Toronto, mROBerTO (milli-ROBot TORonto) is designed for swarm-robotics researchers who might wish to test their collective-behavior algorithms with real physical robots.
Authenticity is an important trait, and zebrafish take it especially seriously. An interdisciplinary team of researchers discovered that zebrafish engage more with 3D-moving robotic models of themselves than with other stimuli.