ONAGOfly is an auto-following, palm-sized drone

---

This drone proves big things can come in small packages. 

---

In today’s market, consumers have pretty much two choices: cheaper nano drones or larger, pricier quadcopters. ONAGOfly wants to be the best of both worlds. Not only does it let users capture high-res selfies and live-stream footage to their mobile devices, the palm-sized unit only costs $200.

This consumer-friendly drone boasts a safe design, suitable for both indoor and outdoor use, and can be piloted right out of the box. It can be set to automatically follow you, or you can use its accompanying app to control the drone via Wi-Fi (up to a distance of 66 feet).

With its video game-inspired tilt control mode, ONAGOfly can be steered simply by turning its paired smartphone left and right, or up and down to fly higher and lower. Meanwhile, photos and videos can seamlessly sync to a user’s handheld gadget for instant sharing.

ONAGOfly can take off and land right from your hand, and be launched with the press of a button. The tiny UAV features built-in infrared sensors on all four sides, allowing it to avoid any potential collisions with obstacles in its way. Additionally, ONAGOfly’s GPS module enables it to automatically follow someone using the location of its connected smartphone as they run, snowboard, cycle, surf or whatever else.

According to company founder Sam Tsu, the mini ‘copter can be used by everyone of all ages and experience levels. This includes athletes, travelers, wedding planners and other drone enthusiasts.

In terms of its camera, ONAGOfly’s images and videos are being touted as comparable to that of an iPhone 6 (15MP photo and 1080P HD at 30fps video). With P2P streaming, users can watch footage in real-time from a remote device without delay. To maximize group photos, the drone can even recognize faces and detect smiles once all subjects are in the frame, and then snap the picture.

Thanks to a 1000mAh LiPo battery, users can expect around 12-15 minutes of flight time. The ONAGOfly weighs only 140 grams (0.3 pounds), and can reportedly maintain its position in wind speeds of up to 10.8 feet per second.

Interested? Head on over to ONAGOfly’s Indiegogo campaign, where the nano drone’s creators have already flown right by their goal of $150,000. Delivery is slated for February 2016.

Mayday is like an airbag system for your drones

---

Mayday is a standalone machine learning device that can detect when your quadcopter is crashing and deploy a parachute.

---

Truth be told, the failure rate associated with drones is extremely high, no matter your skill level. Like the airbag in your car that will deploy when it senses a collision, the North UAV team has taken a similar concept and applied it to the world of unmanned aerial vehicles. Not only is having to repair your ‘copter after it plunges into the water or shatters upon hitting the ground quite pricey, the people and property below it are at risk as well.

Designed with this in mind, Mayday is a standalone machine learning device that detects when your UAV is about to crash and deploys a parachute to guide it to safety. The smart on-board computer monitors a multi-rotor’s flight patterns and intervenes if something goes wrong by activating a range of servo-based countermeasures.

What’s nice is that Mayday is fully programmable. Using the two buttons on the front of the instrument, you can guide each servo head to the waypoint you want to go to in a failure event. You can even configure the Mayday to do an assortment of servo release motions to safeguard it against a crash. This allows it to be used with almost any recovery system on the market. And unlike many other RC products, the Mayday is completely input protected.

Easy to use, just as easy to install! That’s because Mayday features a two-cable interface and simple mounting, and seamlessly works with pretty much every quadcopter setup. Because it uses machine learning to determine your normal flight pattern, you can employ Mayday without having to enter in a ton of data or define certain perimeters for it to be triggered by.

Beyond that, it can also be used all by itself without any connections to a flight controller. Attach the parachute release servo to the Mayday board, throw on a small battery, and you’re good to go.

“For example, you are about to do something totally unexpected and new, like a flip, and you want to make sure Mayday doesn’t fire on accident. Simply adjust your RC servo input to the lower PWM range to tell Mayday not to fire and to try learning this new motion. Or adjust your RC servo input to the upper PWM range to override Mayday and to fire a recovery system,” says creator Kyle O’Rourke.

In terms of size, the unit measures roughly one-square-inch and weighs a half of an ounce. With an ATmega328P at its core, Mayday is equipped with handful of sensors, including an altimeter to detect relative altitude, a gyroscope for rotation speed, an accelerometer for angle and gravity, and a magnetometer for heading to relative magnetic north. Additionally, the device can be powered by a variety of sources, whether that’s a regular Li-Po to a couple of AA batteries.

Still reluctant? Mayday boasts a manual override and suppression input for those who still want some autonomous protection but need more control.

“We’re not saying that a recovery system can completely remove these risks, but we believe that having one can help reduce the total damage by a substantial amount (sort of like an airbag in a car),” O’Rourke explains.

Currently live on Kickstarter, the North UAV crew has flown right by its $12,000 goal. Delivery is expected to get underway in November 2015.

These palm-sized drones can unfold and deploy in half a second

---

Dude, is that a drone in your pocket?

---

Disaster relief efforts are among the top use cases that drone advocates have been petitioning in recent years, and rightfully so. Their unmatched ability to be released over a dangerous or inaccessible area to snap photographs and make contact with survivors far exceeds other methods being implemented today. With this in mind, researchers at EPFL and NCCR Robotics have developed an origami-inspired UAV that not only folds down into a pocketable square, but actually opens itself up and takes flight in a fraction of a second.

“You can take it out of the box, switch on the motor, and it’s ready to fly,” explained Dr. Stefano Mintchev, a professor of bio-inspired robotics at EPFL in Switzerland.

The current prototype, which was recently unveiled at the International Conference on Robotics and Automation in Seattle, features a set of arms comprised of fiberglass and inelastic polyester with propellers at their ends. When activated, the force of the rotors pulls each foldable arm out into its extended position where it’s held in place by magnets. In order for this to work, the rotors must turn in the same direction, causing the arms to rotate out the opposite way and open around two vertical folds. When the arms are fully extended, their upper section moves horizontally and locks the segment open. Otherwise, when not in use, the arms fold up in the shape of a trapezoid for easy stowing.

To maintain stability, two of the quadcopters rotors must turn clockwise, with the other two turning counter-clockwise. A sensor detects when the rotors are fully extended, then reverses the spinning direction within 50 milliseconds.

Impressively, the neatly folded drone measures 6.3″ x 6.3″ by 1.4” in size and weighs just over an ounce. When opened, it spans to roughly 2.3” x 2.3” x 1.4”.

“This quick-starting drone, while simple in appearance, is made up of a number of well-thought-out parts. The stiffness of the arms, for example, is critical to the quadrotor’s manoeuvrability. If these parts were flexible, they could bend and vibrate while in flight, causing instability and reducing the quadrotor’s response time to external commands,” the researchers explain. “Stiffness in the arms is a key factor for folding, and by spreading out horizontally the arms avoid imbalances caused by the laws of gravity. There is no need for an additional reinforcing mechanism, which would add to the weight of the device.”

At the moment, the drone must still be folded manually, but it takes less than 10 seconds for someone with practice. The team reveals that this process will be automated in future iterations along with a lighter body and stronger arms to withstand crashes. The principle of origami folding could also be applied to other types of flying devices in the form of wings, a protective cage or other innovations, the researchers claim.

Interested? Read all about the project here.

Drones are now being used to inspect bridge damage

According to a recent report from the White House Administration, one in four bridges in the United States is in dire need of significant repair or cannot handle automobile traffic. Typically, when bridges are inspected for defects, such as cracks, engineers must use hanging scaffold systems or view them from elevated platforms. It’s a slow, dangerous, expensive process and even the most experienced engineers can overlook cracks in the structure or other critical deficiencies. However, Tuft University engineers are employing wireless sensors and drones that may soon be able to examine the condition of bridges in a quicker, more efficient manner.

Led by assistant professors Babak Moaveni and Usman Khan, the Tufts University team is developing a detection system using smart sensors that are permanently attached to bridge beams and joins. Each sensor can continuously record vibrations and process the recorded signals; furthermore, any changes in the vibration response can signify damage, Moaveni explained. A wireless system would then use unmanned aerial vehicles (UAVs) to hover near the sensors and collect data while taking visual images of bridge conditions. These quadcopters would transmit data to a central collection point for analysis. According to Tufts, Khan was recently awarded $400,000 award from the National Science Foundation to explore this technology, which requires addressing significant navigational and communications challenges before it could be a reliable inspection tool.

Five years ago, Moaveni installed a series of 10 wired sensors on a 145-foot-long footbridge on the Tufts Medford/Somerville campus. These sensors measured vibrations that passed through the bridge, caused by people walking across it. In 2011, Moaveni added nearly 5,000 pounds of concrete weights on the bridge deck to simulate the effects of damage on the bridge — a load well within the bridge’s limits. Connected by cables, the sensors recorded readings on vibration levels as pedestrians walked across the span before and after installation of the concrete blocks. Tufts notes that from the changes in vibration measurements, Moaveni and his research team could successfully identify the simulated damage on the bridge, validating his vibration-based monitoring framework.

The scientists are currently working on a way of scaling the system, in hopes that it could be applied to larger, car-carrying bridges. A major goal of his research, Moaveni says, is to develop computer algorithms that can automatically detect damage in a bridge from the changes in its vibration measurements. According to Moaveni, the system should already be capable of detecting severe damage, but still needs some tweaking before it can pick up on more subtle defects. “Right now, if a bridge has severe damage, we’re pretty confident we can detect that accurately. The challenge is building the system so it picks up small, less obvious anomalies.”

This isn’t the first time a drone has been used to examine the condition of fatigued bridges. Back in 2011, a team of architects used a remote-controlled aircraft to survey the 500-year-old Stirling Bridge in Scotland and assess what repair work needed to be done. From agriculture to real estate, there are countless ways these flying apparatuses will soon, if not already, revolutionize the world around us.

 

Yes, open source hardware is taking flight!

Writing for OpenSource.com, Jason Baker of Red Hat notes that one of the best open source drone communities he’s come across is DIY drones – a site that offers forums, videos and succinct how-tos, along with an online store selling kits and components.

“DIY drones, among other things, is the host of the [Atmel-based] Ardupilot project, an Arduino-based system to help you get off the ground with a hardware, software, and firmware solution for flying nearly anything,” Baker explains.

Mustang P-47D. Image Credit: Aaron Manee, DIYDrones.com

“Versions exist for everything from fixed-wing aircraft to copters with nearly any number of propellors, and even a version for rovers for land-lovers not quite ready to take flight.”

As Baker notes, quadcopters and related vehicles are great if you want to control a flight that can be measured in meters.

“But what if you want to touch the edge of space? Not surprisingly, there’s open hardware for that too. Two of your best options for flying a little bit higher on a consumer budget are balloons and hobbyist rockets,” he says.

Image Credit: Wilfred Swinkels, DIYDrones.com

“There are plenty of instructions out there for you to try re-creating this feat on your own. Some require advanced hardware skills, but what sensors and what tracking system you include are as much a matter of your own skills and interests as anything else.”

According to Baker, open source model rocketry might be another platform of choice for DIY Makers and hobbyists, as it offers fairly easy entry and re-entry options.

“It’s an exciting time for open source flight. Even the US miliitary has recently made a decision to open source some of the work they are doing, in coordination with the Open Source Software Institute,” Baker adds. “Whether you’re an open hardware pro, or someone like me who is just getting started, there are plenty of options for diving in.”

Interested in learning more? The full text of “Open Source Hardware takes Flight” can be read here on OpenSource.com, while the DIY Drones homepage can be accessed here. Readers may also want to check out our recent article on the PAVA 9, a sleek ATmega328P-based tracker.