Inside Unmanned Systems

APR-MAY 2018

Inside Unmanned Systems provides actionable business intelligence to decision-makers and influencers operating within the global UAS community. Features include analysis of key technologies, policy/regulatory developments and new product design.

Issue link: https://insideunmanned.epubxp.com/i/969777

Contents of this Issue

Navigation

Page 60 of 67

61 April/May 2018 unmanned systems inside ENGINEERING. PRACTICE. POLICY. vehicles or construction work blocked its way. "DroNet and NanoMap solve two different and somewhat complementary problems—the for- mer can be used to explore unknown dynamic indoor and outdoor environments, while the latter can quickly navigate through static and known ones," Loquercio said. In experiments, DroNet could safely navi- gate—autonomously—a number of busy city streets it had not previously seen, including straight paths and sharp curves, recognizing static and dynamic obstacles and slowing to avoid crashing into them. "At the moment, our system is still very conservative, which means that it simply stops if some obstacle blocks its way," Loquercio said. Surprisingly, DroNet could also generalize its knowledge to f ly autonomously in indoor environments it was never taught to navi- gate, such as parking lots and office corridors. "Given the fact we used a small commercial drone, the speed limit we had was 3 meters per second," Loquercio said. "However, with better hardware, we could go much faster." One weakness of DroNet "is that it does not fully exploit the agility of drones, since it limits the drone's motion to two di- mensions," Loquercio said. "The drone learned from cars and bicycles, which are not able to f ly up obstacles." In the future, Loquercio and his colleagues would like to develop drones that can autono- mously accomplish tasks given only a simply stated goal, such as reaching the nearest post office. "This requires an understanding that drones still lack," Loquercio said. " THE DRONE LEARNED FROM CARS AND BICYCLES, WHICH ARE NOT ABLE TO FLY UP OBSTACLES." Antonio Loquercio, an artifi cial intelligence researcher, University of Zurich The most powerful LabSat yet, the new LabSat 3 WIDEBAND captures and replays more GNSS signals at a much higher resolution than before. LabSat 3 WIDEBAND can record and replay the following signals: • GPS: L1 / L2 / L5 • GLONASS: L1 / L2 / L3 • BeiDou: B1 / B2 / B3 • QZSS: L1 / L2 / L5 • Galileo: E1 / E1a / E5a / E5b / E6 • SBAS: WAAS, EGNOS, GAGAN, MSAS, SDCM • IRNSS www.labsat.co.uk

Articles in this issue

Links on this page

Archives of this issue

view archives of Inside Unmanned Systems - APR-MAY 2018