This compilation highlights the power of the storm and what residents face in the weeks ahead. Days after powerful Hurricane Michael made landfall Wednesday just north of Mexico Beach, Florida, thousands of people have been reported missing to local authorities. Emergency crews made it their mission Friday to search for people after the Category 4 storm barreled into the Florida Panhandle before making its way to southwest Georgia and South Carolina – while also lashing North Carolina and Virginia. At least 17 people have been killed. Emergency officials said they received thousands of calls asking about missing people, but with cellphone service out across a wide area, they found it impossible to know who among those unaccounted for was safe but just unable to dial out to friends or family.

Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers. We'll also be posting a weekly calendar of upcoming robotics events for the next few months; here's what we have so far (send us your events!): Let us know if you have suggestions for next week, and enjoy today's videos. We already posted about the Atlas doing parkour video, which Marc Raibert first showed at IROS earlier this month; he also showed this video, which is just as interesting (if not quite as dramatic), since it shows SpotMini in what could be its first realistic commercial application. We have begun field testing the Spot robot for commercial usage around the world.

It turns out that small drones can leave a big impact on large aircraft. Scientists at the University of Dayton Research Institute discovered this when they flung a DJI Phantom 4 drone into the sky from a cannon. It didn't take very long before the drone crashed spectacularly with a Mooney M20 airplane. Luckily, the collision was staged as part of a study and no one was injured in the process. For the study, researchers worked to mimic a midair collision between a 2.1-lb drone and an airplane at a speed of 238 miles per hour.

Rogue drones have nearly caused air accidents, have been used as offensive weapons, to deliver drugs to prisoners, and to spy on people. So how can we fight back? This summer a packed Airbus A321 came within 100ft (30m) of disaster after encountering a drone at 15,500ft. And the number of near-misses of this sort has trebled over the last three years, with 92 incidents reported last year in the UK alone. Dozens were classified as involving a serious chance of a collision.

Typically, you've got an artist on stage singing songs and stuff, and then a bunch of spotlights beaming columns of color through some fake smoke. But something new is on the horizon, and it's equal parts creepy and futuristic. Swarms of artificially intelligent drones are starting to show up on stages around the world. Some, like the ones on Drake's latest tour, of are tiny flying lights that float above the stage. Others, like a recent Cirque du Soleil experience, featured more complex aircraft outfitted with lampshades that produced an almost ghostly effect.

In 2014, we wrote about some failsafe software from ETH Zurich that allowed a quadrotor to remain fully controllable even with one busted motor. The unbalanced torque generated by three motors means that a quadrotor can't help but spin, but with a bit of cleverness, software can compensate for the spin and keep the quadrotor stable and even allow it to obey control inputs, allowing it to land more or less safely. This is a valuable capability, but there are a few things that it doesn't address. For example, what if your quadrotor loses a rotor over an unsafe area? What if something happens to it when it's already traveling at a high speed?

Drone racing has only been a human sport for a few years, but artificial intelligence is already gunning to take over. Today, the Drone Racing League (DRL), which is one of the foremost organizations trying to turn drone racing into the next NASCAR, announced a new competition for teams to develop AI pilots for its aircraft. With backing from aerospace firm Lockheed Martin, DRL wants to recruit developers from around the world, including students and drone enthusiasts. They'll have to create an AI that's capable of flying one of DRL's standardized quadcopters through its complex race courses without preprogramming or human supervision. Teams will then compete in the DRL's upcoming 2019 season by racing against one another in the same courses as human pilots as part of the newly designated Artificial Intelligence Robotic Racing (AIRR) circuit.

The man flying the drone didn't know he was violating a temporary restriction on flights around New York City (the president was in town for the 2017 United Nations General Assembly). He didn't know he had just two minutes to land before he violated the prohibition on nighttime flights. And he didn't know his DJI Phantom 4--300 feet up, 2.5 miles away from him, and well beyond his line of sight--was flying dangerously close to an Army Black Hawk helicopter. The drone smashed to pieces. The helicopter, luckily, only suffered a dented rotor and a few scratches, according to the National Transportation Safety Board report.

Example of a recorded highway including bounding boxes and labels of detected vehicles. The color of the bounding boxes indicates the class of the detected object (car: yellow, truck: green). Every vehicle is assigned a unique id for tracking and its speed is estimated over time. Abstract-- Scenario-based testing for the safety validation of highly automated vehicles is a promising approach that is being examined in research and industry. This approach heavily relies on data from real-world scenarios to derive the necessary scenario information for testing. Measurement data should be collected at a reasonable effort, contain naturalistic behavior of road users and include all data relevant for a description of the identified scenarios in sufficient quality. However, the current measurement methods fail to meet at least one of the requirements. Thus, we propose a novel method to measure data from an aerial perspective for scenario-based validation fulfilling the mentioned requirements. Furthermore, we provide a large-scale naturalistic vehicle trajectory dataset from German highways called highD. We evaluate the data in terms of quantity, variety and contained scenarios. Our dataset consists of 16.5 hours of measurements from six locations with 110 000 vehicles, a total driven distance of 45 000 km and 5600 recorded complete lane changes. A technical proof of concept for highly automated driving (HAD) has already been shown in many demonstrations and test drives.

Last week saw the 2018 FAA Reauthorization Act become law, and the new legislation has quite a few implications for people who fly small drones or model aircraft as a hobby. Before diving into the latest changes, it's worth reviewing how the U.S. Federal Aviation Administration has regulated such things in the past. Way back in 1981, the FAA issued an "Advisory Circular" that provided guidance for people flying model aircraft. Most modelers considered those guidelines reasonable enough, but if you didn't conform to them, it was no big deal--they weren't rules, just recommendations. So, for example, if you flew a model sailplane and caught a thermal that took it more than 400 feet off the ground, the FAA really couldn't object that you were in violation of its advice to keep lower.