A method for configuring a neural network which is designed to map measured data to one or more output variables. The method includes: transformation(s) of the measured data is/are specified which when applied to the measured data, is/are meant to induce the output variables supplied by the neural network to exhibit an invariant or equivariant behavior; at least one equation is set up which links a condition that the desired invariance or equivariance be given with the architecture of the neural network; by solving the at least one equation a feature is obtained that characterizes the desired architecture and/or a distribution of weights of the neural network in at least one location of this architecture; a neural network is configured in such a way that its architecture and/or its distribution of weights in at least one location of this architecture has/have all of the features ascertained in this way.

Provided is a detection and classification system and method for small unmanned aircraft systems (sUAS). The system and method detect and classify multiple simultaneous heterogeneous RC transmitters/sUAS downlinks from the RF signature using Object Detection Deep Convolutional Neural Networks (DCNNs). The method further utilizes not only passive RF, but may also utilize Electro Optic/Infrared (EO/IR), radar and acoustic sensors as well, with a fusion of the individual sensor classifications. Detection and classification with Identification Friend or Foe (IFF) of individual sUAS in a swarm, multi-modal approach for high confidence classification, decision, and implementation on a low C-SWaP (cost, size, weight and power) NVIDIA Jetson TX2 embedded AI platform is achieved.

A flapping-wing aerial robot formation control method includes: determining a trailing vortex generation mechanism, an energy saving principle and a trailing vortex attenuation mechanism of the formation flight of a group of wild geese in accordance with the pattern of the formation flight of the group of wild geese; determining the formation flight of a group of flapping-wing aerial robots and a formation switching solution in accordance with the trailing vortex generation mechanism, energy saving principle and trailing vortex attenuation mechanism of the formation flight of the group of wild geese in conjunction with the flapping characteristic of a flapping-wing aerial robot from the perspective of energy consumption equalization and energy saving; and carrying out formation keeping control and formation reconfiguration control in accordance with the formation flight of the group of flapping-wing aerial robots and the formation switching solution by controlling positions of the group of flapping-wing aerial robots.

A flock of drones provide a drone-assisted mesh network for first responders. Network modules attached to the drones interconnect with other network modules and provide network access points for first responder devices, allowing the first responder devices to communicate with each other via the drone-assisted mesh network. The drones may autonomously reposition themselves to create a desired network coverages area, including adjusting the network coverage area as instructed via a drone controller. The network modules may communicate with a gateway to an external network, allowing first responder devices to communicate with the external network via the drone-assisted mesh network. Network modules may be selected for field-attachment to the drones based on characteristics of the first responder devices.

The present invention provides a fruit harvesting, dilution and/or pruning system comprising: (a) a computerized system for mapping an orchard or a map of trees position and their contour in a plantation; (b) a management system for autonomous unmanned aircraft vehicle (UAV) fleet management for harvesting, diluting or pruning fruits; and a method for UAV autonomous harvesting, dilution and/or pruning of an orchard.

Embodiments are provided for controlling a fleet of vehicles and drones. The vehicles are directed to respectively drop off passenger groups at multiple locations. Routes are calculated routes for the vehicles to respectively pick up the passenger groups from the multiple locations based on predicted pick-up times, passenger group sizes and available vehicle capacities. One or more assign drones are assigned to each passenger group at each location. Each drone is configured to broadcast a current location of the passenger group in the location and a corresponding one of the predicted pick-up times and delay the corresponding passenger group in the location based on one of the vehicles assigned as a pick-up vehicle for the passenger group being delayed.

Methods and systems described herein relate to power generation control for an aerial vehicle. An example method may involve determining an asynchronous flight pattern for two or more aerial vehicles, where the asynchronous flight pattern includes a respective flight path for each of the two or more aerial vehicles; and operating each of the aerial vehicles in a crosswind flight substantially along its respective flight path, where each aerial vehicle generates electrical power over time in a periodic profile, and where the power profile of each aerial vehicle is out of phase with respect to the power profile generated by each of the other aerial vehicles.

Using power line rights of way for UAV routing provides a direct, uninterrupted, aerially clear path to the vast majority of lots and buildings from nearby substations and generating stations. Segmenting or separating the UAV traffic by airframe glide ratio improves safety for people on the ground and utilization of the limited airspace. Further segmenting UAV traffic by airframe speed and size allows greater traffic throughput.

A system for aiding formation flying of a follower aircraft with respect to a wake vortex from a leader aircraft comprises a controller and at least one accelerometer installed on the follower aircraft. The controller receives acceleration measurements performed by the at least one accelerometer and processes the measurements to obtain a value representative of vibrations generated by the wake vortex from the leader aircraft. The controller compares the value with at least one predetermined threshold representative of excessive vibrations with regard to location of the at least one accelerometer on the follower aircraft. One or more notifications, such as alerts, are generated based on the result of the comparison. It is easier to position the follower aircraft in formation flying to benefit from a rising airflow phenomenon brought about by the wake vortex from the leader aircraft.

Autonomous unmanned vehicles (UVs) for responding to situations are described. Embodiments include UVs that launch upon detection of a situation, operate in the area of the situation, and collect and send information about the situation. The UVs may launch from a vehicle involved in the situation, a vehicle responding to the situation, or from a fixed station. In other embodiments, the UVs also provide communications relays to the situation and may facilitate access to the situation by responders. The UVs further may act as decoupled sensors for vehicles. In still other embodiments, the collected information may be used to recreate the situation as it happened.