A flight trajectory defines how exactly a quadrocopter moves in the three-dimensional space from one position to another. Automatic flight trajectory planning faces challenges such as high computational effort and a lack of precision. Hence, when low computational effort or precise control is required, programming the flight route trajectory manually might be preferable. However, this requires in-depth knowledge of how to accurately plan flight trajectories in three-dimensional space. We propose planning quadrocopter flight trajectories manually using the Programming by Demonstration (PbD) approach -- simply drawing the trajectory in the three-dimensional space by hand. This simplifies the planning process and reduces the level of in-depth knowledge required. We implemented the approach in the context of the Quadcopter Lab at Ulm University. In order to evaluate our approach, we compare the precision and accuracy of the trajectories drawn by a user using our approach as well as the required time with those manually programmed using a domain specific language. The evaluation shows that the Drawjectory workflow is, on average, 78.7 seconds faster without a significant loss of precision, shown by an average deviation 6.67 cm.

Reinforcement learning often uses neural networks to solve complex control tasks. However, neural networks are sensitive to input perturbations, which makes their deployment in safety-critical environments challenging. This work lifts recent results from formally verifying neural networks against such disturbances to reinforcement learning in continuous state and action spaces using reachability analysis. While previous work mainly focuses on adversarial attacks for robust reinforcement learning, we train neural networks utilizing entire sets of perturbed inputs and maximize the worst-case reward. The obtained agents are verifiably more robust than agents obtained by related work, making them more applicable in safety-critical environments. This is demonstrated with an extensive empirical evaluation of four different benchmarks.

The video shows quadrocopters autonomously assembling a rope bridge. This is part of a body of research in aerial construction, a field that addresses the construction of structures with the aid of flying machines. In this work, a rope bridge that can support the crossing of a person is built by quadrocopters, showing for the first time that small flying machines are capable of autonomously realizing load-bearing structures at full-scale and proceeding a step further towards real-world scenarios. Except for the required anchor points at both ends of the structure, the bridge consists exclusively of tensile elements and its connections and links are entirely realized by flying machines. Spanning 7.4 m between two scaffolding structures, the bridge consists of nine rope segments for a total rope length of about 120 m and is composed of different elements, such as knots, links, and braids.

MIT's Robust Robotics Group seems to be as thrilled with the Kinect and the hacking possibilities that emanate therefrom as we are. They've attached a Kinect to a quadrocopter, which enables completely autonomous 3-D mapping and flight--even the processing is done on board. MIT worked with the University of Washington on this project, using UW's SLAM (Simultaneous Localization And Mapping) algorithms to construct these pretty models of the environment, using the data picked up by the Kinect's sensors. The SLAM maps are actually kind of a bonus on top of the main function of the project, which is to enable fully autonomous flight in areas without GPS coverage: SLAM maps are processed off site, but they're not necessary to the operation of the quadrocopter. The project has some pretty obvious military uses, which explains why it was sponsored by the Office of Naval Research and the Army Research Office (way to mix up the naming conventions, guys).

Raffaello D'Andrea heads ETH Zurich's Flying Machine Arena Arena is at forefront of research into autonomous flying robots Quadrocopters learn amazing throwing and catching maneuvers D'Andrea says technology education needs to promote "unconstrained creation" D'Andrea says technology education needs to promote "unconstrained creation" Professor Raffaello D'Andrea isn't short of admirers for his autonomous flying robots and the amazing tricks they perform. Every week, he receives a flood of e-mails from excited people telling him how to use them, he says. "Folks have contacted me about using them to deliver burritos and pizzas, paint walls, do search and rescue, monitor the environment, flying cameras for movies ... It's just endless," D'Andrea says. "I'm not going to pass judgment on whether they are good or bad ... my role is to show people what is possible." It appears those possibilities are growing by the day at the Swiss Federal Institute of Technology in Zurich (ETH Zurich) where D'Andrea leads a team of researchers at the Flying Machine Arena (FMA).

The next step on the road to robo-domination is, apparently, juggling. From Switzerland's Institute for Dynamic Systems and Control comes this video of flying robots (quadrocopters) juggling a ball individually and in teams. The quadrocopters in the video are operated by humans (for now). They do, however, have onboard computers and are able to fly themselves if properly programmed. As humanity continues to propel itself toward the inevitable robot apocalypse, it's nice to know our future machine masters will at least be able to keep us entertained.

Can it be used as a spying tool? Will you get away with flying it undetected by whomever you're spying on? Even when it's flying 130 feet above your head, you can hear the quadrocopter buzzing back and forth. In 2014, when consumer-level quadrocopters were relatively new, I flew a DJI Phantom 2 over the doomed-for-demolition New York graffiti landmark 5 Pointz. A photographer, who had snuck in for one last glimpse of the inside before the building was flattened, noticed my camera buzzing over his head.