Back to results
US 12,029,166 B2Granted
Harvester with automated targeting capabilities
AssigneeCeres Innovation, LLC
InventorsFrank Faulring, Carlos Ramirez, Jason Faulring, Mark Keyser, Jacob Goellner
Life of this patent
Estimating the term…
Abstract
Systems and methods here may include a vehicle with automated subcomponents for harvesting delicate items such as berries. In some examples, the vehicle includes a targeting subcomponent and a harvesting subcomponent. In some examples, the targeting subcomponent utilizes multiple cameras to create three-dimensional maps of foliage and targets. In some examples, identifying targets may be done remotely from the harvesting machine, and target coordinates communicated to the harvesting machine for robotic harvesting.
Claims19
A harvesting vehicle system comprising:
- a harvesting vehicle frame with computing device includes at least one processor and a memory including picking control systems, navigation control systems, and communication control systems, a picking subcomponent including a robotic arm in communication with the computing device;
- wherein the robotic arm including a picker head assembly to harvest targets including a vacuum assembly with a compressor, hose, and padded spoons configured to remove targets from target stems;
- wherein the vehicle includes multiple two-tiered stereoscopic cameras in a wide field of view and narrow field of view configuration, in communication with the computing device, wherein the wide field of view cameras are configured to capture and send image data of a target to the computing device;
- the computing device further configured to create three-dimensional maps of the target using the multiple wide field of view camera image data;
- wherein the computing device further configured to use the image data of the target and three-dimensional maps of the target to a capture narrow field of view image data of the target;
- the computing device further configured to direct the robotic arm and picker head to a selected target to harvest using the narrow field of view image data and three-dimensional maps of targets, the picker head assembly configured to attach the vacuum assembly and padded spoons to the selected target, and retract the selected target for harvesting.
3 more elements
A method of harvesting agriculture, comprising:
- traversing a harvesting vehicle frame across a row of agricultural plants wherein the harvesting vehicle includes a computing device with a processor and a memory,
- wherein the computing device including target acquisition control, picking control,
- wherein the harvesting vehicle including a picking subcomponent with a robotic arm with a picker head assembly, wherein the robotic arm in communication with the computing device, the picker head assembly including a vacuum assembly with a compressor, hose, and padded spoons;
- capturing and sending wide field of view image data to the target acquisition control of the computing device, using multiple wide field of view cameras on the harvesting vehicle;
- identifying targets in the agricultural plants, by the target acquisition control of the computing device, using the wide field of view image data;
- creating three-dimensional maps of targets, by the computing device using the wide field of view image data;
- determining coordinates of the targets from the three-dimensional maps;
- generating a queue of determined coordinates of targets;
- sending the queue of determined coordinates of targets to narrow field of view cameras;
- capturing and sending narrow field of view image data to the picking control of the computing device, using the narrow field of view cameras on the harvesting vehicle;
- directing, by the picker control of the computing device, the robotic arm and picker head to a first selected target coordinate in the queue of determined coordinates using the narrow field of view image data and three-dimensional maps of targets;
- and
- harvesting, by the picker control of the computing device, the first target with the picker head assembly by attaching the vacuum assembly and padded spoons to the mapped target, and retracting the first target.
10 more elements
Dependencies are inferred from the claim text.