Docking drones in mid-air is easier said that done. With the rise of modular robots, midair docking is becoming quite necessary. HiPeR Lab researchers have developed a way to dock multirotors in mid-air using relative position measurements from computer vision.
Remember the Hero Arm? It is a robotic prosthetic arm with 8kg lifting capacity that can make life easier for amputees. Daniel Melville, a one-handed gamer has gotten a Metal Gear Solid version of it. Open Bionics teamed with KONAMI for this project.
Dealing with cables comes with its own challenges. We have all had to deal with tangled cables when dealing with electronics. Robots can also have these issues. Dress packs used on robots can become problematic when cables are too tight or loose. ADDRESS is a robotic system that addresses e-stops and tangling.
Meet the DelFly Nimble: a tailless flapping wing robot that uses a single small camera and tiny processor to fly autonomously. This drone weighs around 30g. It has a VL53L0X range sensor, ESP8266 WiFi module, and a vision system consisting of a monocular camera and STM32F4 microprocessor.
In the past few years, we have covered plenty of bio-inspired robots that can take on a variety of tasks. UNSW researchers have developed a robotic device that works like a snake or an elephant’s trunk to grip and pick up objects. It can be used to handle fragile objects such as food items.
Here is a robot designed to simulate hand holding with a girlfriend or wife. Designed by Japanese engineers, Osampo Kanojo has a special rail around its elbow that mimics push and pull motion, simulating the weight of a person. It has a soft skin that maintains the temperature of a human body.
It is no secret that consumer drones leave a whole lot to be desired when it comes to their battery life. Elistair’s tethered drone, Orin 2, is made for more serious jobs. It is a a hexacopter for military, surveillance, and security applications. It can stay in the air for up to 24 hours.
We have seen plenty of robots that can jump, roll, and recover after a fall. Teaching them to jump very high comes with its own unique challenges. This video from Open Dynamic Robot Initiative shows how a quadruped learns to jump as high as possible using Bayesian optimization with unknown constraints (BOC).
Robots are getting better at negotiating rough and unknown environments all the time. This video shows a multi-layered locomotion system for legged robots that allows them to achieve safe foot placement and dynamic stability.
Drones have been used in the past to bring sensors to hazardous and hard to reach environments. This video from Aerial Robotics Lab @ Imperial College London shows an aerial sensor placement method that involves installing sensors like darts. As the researchers explain:
The proposed method is based on mechanical energy storage and an ultralight shape memory alloy trigger. The developed aerial system weighs a total of 650 grams and can execute up to 17 deployments on a single battery change. The system deploys sensors of 30 grams up to 4 meters from a target with an accuracy of ~10 cm.
[HT]
