The Robotics Roundup is a weekly newspost going over some of the most exciting developments in robotics over the past week.
In today’s edition we have:
- Silicon retinas to help robots navigate the world
- Could a Robot Help Save Deteriorating Infrastructure?
- A Mini Cheetah robot is now learning to play soccer
- Jellyfish-like tentacles lets robot grasp soft objects without crushing them
- The fastest ever laundry-folding robot is here. And it’s likely still slower than you
Silicon retinas to help robots navigate the world
Visual input systems in most modern robotics systems are based on the standard aperture based camera capturing still images for processing. Event based camera technology instead operates more similarly to our retinas, capturing discreet updates to sub portions of the image. This lends itself perfectly to use in vision based robotics, and may prove to be a significant improvement over systems currently in use.
Could a Robot Help Save Deteriorating Infrastructure?
Robotics technology lends itself perfectly to infrastructure inspection, allowing the collection of reliable data from dangerous or difficulty to access locations. This article provides an overview of what Gecko Robotics is up to in the remote inspection space.
A Mini Cheetah robot is now learning to play soccer
Researchers from the Hybrid Robotics Lab at the University of California taught their Mini Cheetah robot how to be a goalie. The video included in the article showcases the robot’s impressive skills, as well as a partner Cheetah taking some shots on goal.
Jellyfish-like tentacles lets robot grasp soft objects without crushing them
Reliable grasping and manipulation of oddly shaped and fragile objects has long been an issue for robotic manipulators. Harvard University researchers have developed a novel gripper using pressurized tubing that provides a secure grip on irregular objects, without risking damage to the grasped object.
The fastest ever laundry-folding robot is here. And it’s likely still slower than you
Laundry folding is a deceptively difficult process to automate. While this new development in robotic folding technology is still several orders of magnitude slower than manual folding, each step brings us significantly closer to a practical household folding robot.