Listen

Description

A drone crash halts Swiss package delivery, AI used to detect drone faults and locate/track other drones, swarming nano-drones, mesh networked drones for precision agriculture, a drone mesh for event security, UAS for airport and public safety, and the reliability of drone sightings from aircraft.

UAV News

Swiss drone crashes near children, forcing suspension of delivery program

In May, a 22-pound Swiss Post drone crashed 50 yards away from a group of children. As a result, the drone delivery program in Switzerland has been suspended indefinitely. Swiss Post has asked Matternet to make several changes to the system: parachutes connected to the UAV by two ropes instead of one, ropes reinforced with metal braiding, and a louder emergency landing whistle.

Fault Diagnosis of Drone Propellers Using AI

Propeller defects in drones can cause them to fall from the sky and researchers have used different methods to perform fault diagnosis. But a new study by Italian researchers takes a different approach and focuses on measuring the noise emitted by the drone. Different noise patterns are used to build an artificial intelligence (AI) model that detects unbalanced blades in a UAV propeller. See Fault Diagnosis for UAV Blades Using Artificial Neural Network.

Drones Chasing Drones Using Deep Learning and AI

A Canadian study proposes target-detection and target-following using AI techniques for UAV pursuit-evasion. Deep reinforcement learning predicts the actions to apply to the follower UAV to keep track of the target UAV. A deep object detector and a search area proposal predict the position of the target UAV for tracking purposes. The proposed framework uses vision data. See Drones Chasing Drones: Reinforcement Learning and Deep Search Area Proposal.

The amazing world of nano drones

Nano drones have been studies for several years. Draper Lab’s DragonflEye drone project was announced in 2017. A real dragonfly wears a tiny backpack fitted with electronics, sensors, and a solar cell. AeroVironment has been working on a nano-hummingbird since 2011. The market for the nano drones alone is expected to reach $13.4 billion by 2023.

If one drone isn’t enough, try a drone swarm

Drones are being used for precision agriculture, but there are shortcomings with one drone and a big farm. Many battery changes may be required and the survey could take a lot of time. The Norwegian Defence Research Establishment and Rajant Corporation in the US are working on a drone swarming approach with decentralized...