Issue 41 Unmanned Systems Technology December/January 2022 PteroDynamics X-P4 l Sense & avoid l 4Front Robotics Cricket l Autonomous transport l NWFC-1500 fuel cell l DroneX report l OceanScout I Composites I DSEI 2021 report

73 The goal of the two companies’ combined r&d is to identify how best to apply UBIQ’s D•ICE ice protection solution to Mejzlik’s propellers. This will subsequently provide the latter’s customers with a means of accurately knowing when they are in icing conditions and to have the capabilities to escape them. StriekAir Engineering showcased its CarryAir UAV, a VTOL-transitioning aircraft with a blended-wing body, four electric rotor drives for hovering, VTOL mounted off the fronts of its wings and two additional lift motors fixed beneath its trailing edges. “The hover motors are deactivated once the craft is in forward flight, and the blades will fold owing to the airflow, with forward propulsion coming from a nose-mounted propeller drive,” explained Matthias Menze. “That drive has two e-motors on a common shaft for redundancy, with each working at half capacity at top speed.” The CarryAir has a 299 cm wingspan, a top speed of 110 kph, and a 25 kg MTOW with a 14.5 kg capacity for batteries and payload. It can fly for up to 3 hours on a 3 kWh battery pack, or for up to 8 hours with a gasoline range extender. The company sees it being suitable for operations including medical logistics, emergency comms relaying, and powerline and pipeline monitoring. We spoke with Vizgard to learn about its computer vision solutions for unmanned and counter-unmanned systems. “We make it as fast and easy as possible for businesses to add visual intelligence to any unmanned or counter- unmanned system,” Alex Kehoe told us. “We see computer vision as an essential piece of the puzzle for achieving safe and reliable automation.” The Vizgard stand showcased both pre-recorded and live demonstrations of its flagship software, FortifAI. Described as an ‘AI toolkit’, the software contains a growing list of bundled capabilities. These include long-range object detection and tracking, face redaction, GPS-denied precision UAV landing and even recognition of human behaviour such as walking or running. Vizgard is due to conduct live trials of its AI and sensor fusion technology on a drone-catching UAV from Gibson Robotics and the Royal Navy’s unmanned submarine after being awarded an r&d contract from the UK government’s Defence Science and Technology Laboratory. “Everything runs on board using one of our VizEdge computers, with simultaneous recording and real-time processing of navigation data, video footage and external sensor tracks such as AIS, ADS-B and radar,” Kehoe said. FlareBright was exhibiting its latest developments in machine learning and digital twinning for flight systems. “Our SnapShot micro-drone is the first incarnation of our AI,” Kelvin Hamilton said. “You just launch it into the air, its flight control system maintains a clear video capture of a single pre-assigned target, and it then performs a controlled flight back down to you at the launch point. “It operates in GNSS-denied environments, fully autonomously, and compensates for wind in real time with extremely high accuracy and updating hundreds of times per second. Our entire machine learning and digital twin DroneX | Show report Unmanned Systems Technology | December/January 2022 Mejzlik’s propeller undergoing trials in UBIQ’s anti-icing test chamber StriekAir Engineering’s blended-wing CarryAir UAV