Patent classifications
G05D1/0204
LEARNING DEVICE, INFORMATION PROCESSING DEVICE, AND LEARNED CONTROL MODEL
The learning system SY1 acquires control information output from a control model M by inputting to the control model M environmental information including weather information in at least one of a surrounding environment of an unmanned aerial vehicle P and an environment of a planned flight area of an unmanned aerial vehicle P, and when the unmanned aerial vehicle P takes an action based on the control information, performs reinforcement learning of the control model M using a reward r representing an evaluation of a result of the action.
METHOD AND APPARATUS FOR AIRBORNE DISSEMINATION AND IMPLANTATION OF SEEDS
This invention relates to a method and apparatus for the airborne dissemination and implantation of seeds utilizing an aerodynamic seed delivery apparatus with built-in nutrients, anti-pest, and anti-fungal properties that can be disseminated rapidly from an airborne platform. The velocity of impact and depth of penetration into specific soil types by the delivery apparatus can be controlled up to a terminal velocity kinetic energy by exploiting a specified drag coefficient, mass, and altitude of release. The seeds are delivered and imbedded into the soil at the optimal depth and orientation to maximize germination rates, since seed orientation has a pronounced effect on germination and sprout mortality rates. Flight paths for Unmanned Aerial Vehicles (UAVs) utilized for dissemination can be automated to adjust coordinates based on wind vectors, terrain elevation data, and soil permeability data to efficiently achieve a desired penetration depth across a specified geographic area.
METHOD FOR DETERMINING THE PATH OF AN UNMANNED AERIAL DEVICE AND OTHER ASSOCIATED METHODS
A modeling method using digital processing of a three-dimensional environment to establish pathways for unmanned aerial devices which are optimized according to different priorities, the method being characterized in that it comprises the following digital processing steps: (a) providing a three-dimensional model of volumes (PEXi) wherein flight is prohibited, (b) subdividing the model into individual elements (PVk), (c) determining a center (Pk) for each individual element, (d) establishing and memorizing a graph, the nodes (Pk, Ik) of which are formed by at least one portion of the centers, and the branches of which are weighted by the distances between the nodes and by at least one weighting associated with a given priority.
A method is also proposed for determining, using an unmanned aerial device, a path between two points in a three-dimensional space modeled by such a graph, and steering methods using such a determination.
CLOSED COURSE NAVIGATION THROUGH A MOVING MEDIUM
A method including propelling a vehicle disposed in a medium. The vehicle includes a body, a propulsion mechanism connected to the body, and a direction control system. The vehicle is subject to advection caused by movement of the medium. The method also includes commanding the vehicle to perform a navigation course comprising a closed course-over-ground. The method also includes periodically adjusting navigation of the vehicle along the closed course-over-ground such that a course-through-the-medium turn-rate is varied in a manner that causes a course-over-ground turn-rate of the vehicle to be held constant, thereby minimizing the impact of medium advection on vehicle speed over ground.
Fully automated launch and recovery platform for unmanned aerial vehicle
A network of automated launch and recovery platforms (LRPs) for at least one aircraft-type aerial vehicle (UAV) which automatically perform cyclic tasks of preparation, launch, and recovery without manual operation. Each LRP includes a stationary foundation in an X-Z plane, a rotatable foundation that can rotate around a Y axis of the stationary foundation, and a rotatable leverage that rotates around the Z axis at a shaft driven by a motor. A first leverage of the UAV is hooked to the rotatable leverage of the LRP such that rotation of the shaft by the motor drives the rotatable leverage and the UAV for take-off and reduces UAV to stop during recovery. The network includes a traffic control subsystem and a launch and recovery subsystem which provides initial UAV speed necessary for launch, and ensures dissipation of kinetic energy of a captured UAV during recovery.
Method and apparatus for airborne dissemination and implantation of seeds
This invention relates to a method and apparatus for the airborne dissemination and implantation of seeds utilizing an aerodynamic seed delivery apparatus with built-in nutrients, anti-pest, and anti-fungal properties that can be disseminated rapidly from an airborne platform. The velocity of impact and depth of penetration into specific soil types by the delivery apparatus can be controlled up to a terminal velocity kinetic energy by exploiting a specified drag coefficient, mass, and altitude of release. The seeds are delivered and imbedded into the soil at the optimal depth and orientation to maximize germination rates, since seed orientation has a pronounced effect on germination and sprout mortality rates. Flight paths for Unmanned Aerial Vehicles (UAVs) utilized for dissemination can be automated to adjust coordinates based on wind vectors, terrain elevation data, and soil permeability data to efficiently achieve a desired penetration depth across a specified geographic area.
WIND VELOCITY MEASUREMENT METHOD, WIND VELOCITY ESTIMATOR AND UNMANNED AERIAL VEHICLE
The present invention relates to a wind velocity measurement method, a wind velocity estimator and an unmanned aerial vehicle (UAV). The wind velocity measurement method includes: determining current wind resistance interference of a UAV by means of system identification based on flight data and attribute data of the UAV; and calculating a wind velocity of a flight environment of the UAV according to the wind resistance interference and the inherent wind resistance of the UAV. The method realizes the wind velocity measurement by identifying parameters based on the principle of system identification without a newly added wind velocity sensor and an external database. Therefore, not only hardware device costs are saved, but also an additional computing burden and a problem about real-time performance are avoided. The method is simple and requires low costs.
CONTROL APPARATUS, CONTROL METHOD, AND PROGRAM
A control apparatus includes a reception section that receives a wind speed vector measured at any time point by at least one external anemometer, a wind-power prediction section that, on the basis of the received wind speed vector, predicts a wind power to be applied to the mobile body after elapse of a predetermined time period, and a control section that controls driving of the mobile body on the basis of the predicted wind power.
Flight feedback control based on gust detection around HAPS
It is prevented that a communication relay apparatus in an upper airspace, which is suitable for constructing a three-dimensional network, falls by a strong wind. A communication relay apparatus is provided with a relay communication station that performs a radio communication with a terminal apparatus, and is capable of flying in an upper airspace by an autonomous control or an external control. This communication relay apparatus includes a flight control section that controls a flight of the communication relay apparatus based on flight control information determined so as to reduce an influence of a strong wind generated around the communication relay apparatus. The flight control information may include information for controlling at least one of a flight direction, velocity, altitude, attitude, flight route and flight pattern of the communication relay apparatus.
Airflow modeling from aerial vehicle pose
Embodiments include apparatus and methods for modeling air flow from flight responses in aerial vehicles. Sensor data is received for aerial vehicles in a geographic area. The pose (e.g., roll, pitch, and yaw) of the aerial vehicles is calculated from the sensor data. One or more wind vectors are calculated based, at least in part, on the pose. An air flow model is generated from the wind vectors.