Pruning system and method thereof

Abstract

In some implementations, the autonomous mountable attachment (AMA) system for performing an optimal pruning action on a plant and method thereof may include receiving plant location data. In addition, the device may include analyzing the received plant location data, using at least one artificial intelligence (AI) model trained to determine an optimal pruning action on a plant. The device may include determining at least one executable instruction for performing an optimal pruning action, based on the analyzed plant location data. Moreover, the device may include sending the determined at least one executable instruction to a pruning arm controller, where the pruning arm controller causes a pruning arm to perform the optimal pruning action on the plant.

Description

PRUNING SYSTEM AND METHOD THEREOFCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application No. 63 / 600,931 filed on November 20, 2023, the contents of which are hereby incorporated by reference.All of the applications referenced above are hereby incorporated by reference.TECHNICAL FIELD

[0002] This present disclosure generally relates to a sensory system and, more specifically, to the use of imaging for the execution of a task.BACKGROUND

[0003] In modem society, cutting costs and increasing revenue are essential for profitability. This is especially true in the agricultural industry, where a single mistake can have long-lasting effects. A mistake like forgetting to water crops or applying the wrong pesticide could ruin a multitude of crops or even an entire harvest. Farmers depend on a certain crop yield in order to be sustainable. Low crop yields increase the likelihood of losses and ultimately, bankruptcy.

[0004] For millennia, the farming industry has been cultivated by the manual labor of humans. However, employing humans for farming labor has its disadvantages, which results in a lower likelihood of profitability because of higher expenses and lower revenue. Specific examples of such include wage expenses, training expenses, insurance expenses, work fatigue, language barriers, work-time restrictions (i.e., lunch, breaks, overtime, sick days, vacation, holidays, etc.), stress, anxiety, labor strikes, safety concerns, and the like. The farming industry has also utilized animals, however, they too are not without their disadvantages. For example, animals, like humans, require training, sleep, rest, instructions, food, hospitable working conditions, retirement (i.e., aging, injury, death, etc.), housing, sanitary conditions, and the like.

[0005] Unlike humans and animals, machinery does not have the aforementioned disadvantages. While some unique disadvantages are present with machinery, they are often lesser than those of animals and humans. Thus, the use of machines leads to agreater likelihood of profitability than other channels of labor. However, the utilization of machinery can often require human and or animal involvement to some extent, reducing some or all of the profitable gains realized by machine use. For example, a wheat harvester, which does not require human needs (i.e., wage, food, sleep, etc.), can still be affected by human needs since a human is required to operate the machinery. Thus, machines dependent on humans or animals still have some or all of the drawbacks associated with humans and / or animals.

[0006] Yet another challenge to agricultural success is training, the length it takes to train, the costs, the resources and the like. Farmers and the agricultural business as a whole have various training methods and requirements in order to be successful, profitable, and law-abiding. For example, crops may need to be planted, harvested, or destroyed in such a way as to remain fruitful. Yet another example, crops may need to be managed in such a way as to be lawful in accordance with laws and regulations (i.e., FDA, OSHA, USDA, etc.). Therefore, animals and laborers in some instances need to be trained in order to farm both profitably and lawfully.

[0007] Training humans and animals, however, is not without difficulties. Difficulties include, but are not limited to, length of training, cost to train, work retention (i.e., an employee can quit or retire), difficulty understanding instructions, and length of time to become an expert. In some instances, it may take a human or animal years to become trained, let alone proficient at any given task. It also may further take an abundance of resources to train. For the training of humans, that may require a language translator to translate the instructions or feedback, amongst other things. As for the training of animals, it might require obtaining animal trainers or breeders so that animals are (1) able to be trained and (2) trained for a specific task.

[0008] It would therefore be advantageous to provide a solution that would overcome the challenges noted above.SUMMARY

[0009] A summary of several example embodiments of the disclosure follows. This summary is provided for the convenience of the reader to provide a basic understanding of such embodiments and does not wholly define the breadth of the disclosure. Thissummary is not an extensive overview of all contemplated embodiments and is intended to neither identify key or critical elements of all embodiments nor to delineate the scope of any or all aspects. Its sole purpose is to present some concepts of one or more embodiments in a simplified form as a prelude to the more detailed description that is presented later. For convenience, the term “some embodiments” or “certain embodiments” may be used herein to refer to a single embodiment or multiple embodiments of the disclosure.

[0010] Certain embodiments disclosed herein include a method for performing an optimal pruning action on a plant using an autonomous mountable attachment (AMA) system. The method comprises: receiving plant location data; analyzing the received plant location data, using at least one artificial intelligence (Al) model trained to determine an optimal pruning action on a plant; determining at least one executable instruction for performing an optimal pruning action, based on the analyzed plant location data; and sending the determined at least one executable instruction to a pruning arm controller, wherein the pruning arm controller causes a pruning arm to perform the optimal pruning action on the plant.

[0011] Certain embodiments disclosed herein include an autonomous mountable attachment (AMA) system for performing an optimal pruning action on a plant. The AMA system comprises: at least one pruning arm configured to perform an optimal pruning action on a plant; at least one attachment tool affixed to the at least one pruning arm; a plurality of sensors affixed to the at least one pruning arm, configured to retrieve plant location data; a pruning arm controller configured to cause the at least one pruning arm to perform the optimal pruning action on the plant; a processing device communicatively connected to the plurality of sensors and the pruning arm controller, and configured to receive plant location data from the plurality of sensors and to determine and send at least one executable instruction to the pruning arm controller; and a vehicle mount attachment connection configured to removably affix the at least one pruning arm to at least one unmanned vehicle.

[0012] Certain embodiments disclosed herein also include a non-transitory computer readable medium having stored thereon instructions for causing a processing circuitry to execute a process. The process comprises: receiving plant location data; analyzingthe received plant location data, using at least one artificial intelligence (Al) model trained to determine an optimal pruning action on a plant; determining at least one executable instruction for performing an optimal pruning action, based on the analyzed plant location data; and sending the determined at least one executable instruction to a pruning arm controller, wherein the pruning arm controller causes a pruning arm to perform the optimal pruning action on the plant.

[0013] Certain embodiments disclosed herein also include a system for controlling an AMA system. The system comprises: a processing circuitry; and a memory, the memory containing instructions that, when executed by the processing circuitry, configure the system to: receive plant location data; analyze the received plant location data, using at least one artificial intelligence (Al) model trained to determine an optimal pruning action on a plant; determine at least one executable instruction for performing an optimal pruning action, based on the analyzed plant location data; and send the determined at least one executable instruction to a pruning arm controller, wherein the pruning arm controller causes a pruning arm to perform the optimal pruning action on the plant.BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The foregoing and other objects, features, and advantages of the disclosed embodiments will be apparent from the following detailed description taken in conjunction with the accompanying drawings.

[0015] Figure 1 is a schematic diagram of an AMA system according to an embodiment.

[0016] Figure 2 is an illustration of an AMA system coupled with a plurality of pruning arms which may be utilized in accordance with various disclosed embodiments.

[0017] Figure 3 is a flowchart illustrating a method for executing executable actions performed by an AMA system in an environment according to an embodiment.

[0018] Figure 4 is a flowchart of a method for a simultaneous localization and mapping process for executing pathing actions, according to an embodiment.

[0019] Figure 5 is a network diagram utilized to describe various disclosed embodiments.

[0020] Figure 6 is a schematic diagram of a processing device according to an embodiment.DETAILED DESCRIPTION

[0021] The embodiments disclosed herein are only examples of the many possible advantageous uses and implementations of the innovative teachings presented herein. In general, statements made in the specification of the present application do not necessarily limit any of the various claimed inventions. Moreover, some statements may apply to some inventive features but not to others. In general, unless otherwise indicated, singular elements may be plural and vice versa with no loss of generality. In the drawings, like numerals refer to like parts through several views.

[0022] The various disclosed embodiments include a method and system for processing and executing actions on plants by an autonomous mountable attachment (hereinafter “AMA”) system. According to an embodiment, the AMA system is configured to execute actions on plants, such as pruning, shearing, thinning, harvesting, weeding, and the like. In certain embodiments, a processing device is configured to receive data from the AMA system, cameras, third-party sources, and the like, to determine an optimal execution of the task. In some embodiments, machine learning models are applied to the processing device, to further improve the execution of the task.

[0023] According to an embodiment, determining an optimal execution of the task allows the AMA system to perform the tasks more accurately, more precisely, more efficiently, and less erroneously, all of which are factors that contribute to increasing agricultural success. In another embodiment, the utilization of the AMA system can reduce the use of chemicals, reduce operational costs, and reduce risks to the operator and other animals (e.g., wildlife, humans, etc.). Increasing agricultural success is desirable, as it increases crop yield (resulting in more food), increases profits (resulting in less required government subsidies), and reduces waste.

[0024] In this regard, it is realized that a human can perform tasks on plants (i.e., pruning, sheering, harvesting, etc.). However, a human cannot perform such tasks on plants in a manner, which is reliable and objective since a human will always make a subjective decision. For example, faced with the same data, a human applies subjective criteria to produce different results (e.g., a branch “looks” like it should be cut based on subjective rationale). This is inconsistent and disadvantageous. The disclosed systemsolves at least this by applying objective criteria in a manner which is reliable and repeatable for each and every determined executable action.

[0025] According to an embodiment, the AMA system is configured to be coupled with an unmanned vehicle (hereinafter “UV”). In other embodiments, the AMA system is configured to be coupled with vehicles and vehicle accessories, including but not limited to, a manned vehicle (e.g., a tractor, a truck, a car, an ATV, etc.), a trailer, a container, any machine capable of movement, a combination thereof, and the like.

[0026] Fig. 1 is an example diagram of an autonomous mountable attachment (AMA) system 100 utilized to describe the various disclosed embodiments. In an embodiment, the AMA system 100 includes at least one pruning arm 111 , a processing device 141 , and a vehicle mount attachment connection 101 . In some embodiments, the processing device 141 , includes a server, implemented as a bare metal machine, a virtual machine, a software container, a serverless function, a combination thereof, and the like. In certain embodiments, the AMA system 100 is configured to communicate with a database or may include a database (in the format of a storage). An example of a processing device 141 is discussed in more detail in Fig. 6 below.

[0027] In certain embodiments, the pruning arm 111 includes sensors 131 , such as but not limited to, a two-dimensional (2D) camera 132, a three-dimensional (3D) camera 133, a thermal camera (not shown), an infrared camera (not shown), a video camera (not shown), a light detection and ranging (LiDAR) camera (not shown), an ultrasonic measuring device (not shown), a combination thereof, and the like. The sensors 131 are deployed in proximity to plants, thereby allowing imaging (and the retrieval of other sensor data) of plants to determine which of the plants require executable action (e.g., pruning, etc.). In some embodiments, the pruning arm is a plurality of pruning arms 111 , each with its sensors 131. In certain embodiments, the sensors 131 are communicatively connected and configured to send data to the processing device 141. In certain embodiments, the sensors 131 are utilized for visual simultaneous localization and mapping.

[0028] In some embodiments, the pruning arm 111 is configured to prune (i.e., cut the branches of a plant). In some embodiments, a 3D camera in conjunction with a 2D camera is beneficial as 3D images will help determine at what angle the cut should bemade. This is not limited to pruning and can be applied to other executable actions. As a non-limiting example, harvesting may also benefit from determinable angles, as a suboptimal harvesting angle may result in bruising of the fruit or damage to the plant.

[0029] According to an embodiment, the processing device 141 is communicatively connected to a robot arm controller 151 via a network. The network may include but is not limited to Wi-Fi, Bluetooth, the internet, wired or wireless connection, local area network (LAN), metropolitan area network (MAN), wide area network (WAN), a combination thereof, and the like. In an embodiment, the robot arm controller 151 is configured to operate the pruning arm. In another embodiment, the processing device 141 is communicatively connected to at least one Global Positioning System (GPS) and / or at least one Inertial Navigation System (INS) 171. In yet other embodiment, the at least one GPS 171 and / or at least one INS 171 are communicatively connected to an operator control interface 161 configured to control the operation of the GPS and / INS 171.

[0030] In some embodiments, the pruning arm 111 includes an attachment, such as but not limited to, a cutting tool (e.g., shears 121 , laser, knife, etc.), a touching tool (e.g., claw, pruning hand, suction cup, etc.), a combination thereof, and the like. In some embodiments, the pruning arm 111 is a plurality of pruning arms 111 , each having its own attachment. In certain embodiments, the pruning arm 111 is configured to receive instructions from the processing device 141. The instructions may include but are not limited to, moving the pruning arm 111 , executing a task on a plant (i.e., pruning a specific branch by using a cutting tool attachment on the pruning arm 111), imaging other areas, and obtaining sensor data in other areas.

[0031] Fig. 2 is an example of an AMA system 200 with a plurality of pruning arms 211- 1 through 211-n, according to an embodiment.

[0032] In an embodiment, the plurality of pruning arms 211 are configured to be adjustable in height. In some embodiments, the vehicle attachment connector 201 is configured to be height-adjustable. In other embodiments, each of the plurality of pruning arms 211 may be individually adjusted and can be mounted at a given height. In some implementations, the plurality of pruning arms 211 may be at different heightsto increase the range of task execution, to better avoid other pruning arms (and other obstacles), and the like.

[0033] According to an embodiment, the pruning arms 211 are utilized to perform executions on a plant. In an embodiment, the pruning arms 211 have a tool such as a cutting tool 212-1 through 212-n. The cutting tool 212 may be utilized to e.g., prune a branch 222-1 through 222-n of a plant 221 .

[0034] According to an embodiment, an assignment bank (e.g., database 530) may be utilized to synchronize all targets into one model in which executions can be assigned to individual pruning arms.

[0035] In an embodiment, where at least one pruning arm within a plurality of pruning arms cannot function, the AMA system 200 may determine such and assign another pruning arm to perform a desired execution of a task. As a non-limiting example, if a first pruning arm is jammed and cannot move, the system 200 is configured to recognize that the first pruning arm cannot perform the task and assigns a second pruning arm (if available) to execute the assignment originally assigned to the first pruning arm. In an embodiment, the processing device 141 may determine that a pruning arm cannot function. Examples where the pruning arm may not function include but are not limited to, the pruning arm being broken, stuck, out of service, or otherwise unavailable.

[0036] Fig. 3 is an example flowchart 300 illustrating a method for controlling the operation of an autonomous mountable attachment (AMA) system according to an embodiment. In an embodiment, the method may be performed by the processing device 141 , Fig. 1.

[0037] At S310, metadata is obtained. According to an embodiment, the metadata may be obtained from, e.g., a data source. The data source may include, but is not limited to, a database (e.g., database 530), user input (e.g., a farmer using a user device, for example, user device 540), a combination thereof, and the like. In an embodiment, the retrieved metadata may include, but is not limited to, plant type (e.g., grapevine, apple tree, blueberry bush, etc.), specific plant varieties (e.g., fuji apple, granny smith apple, golden delicious apple, etc.), executable action varieties within the same specific plant (e.g., (for pruning) thinning cuts, reduction cuts, heading cuts, farmer-specific cuts, etc.), weather, any other plant-related issue, a combination thereof, and the like. In anembodiment, the metadata is continuously obtained. In an embodiment, S310 is optional.

[0038] At S320, sensor data is obtained. According to an embodiment, the sensor data may be received from, e.g., the sensors 131. In an embodiment, the sensor data may include images, videos, any sensory information, and the like, of plants, their subcomponents (e.g., fruits, branches, stems, leaves, roots, etc.), their location within an environment, the location of the AMA system within the environment, the location of items coupled with the AMA system within the environment (e.g., a vehicle, etc.), a combination thereof, and the like. In certain embodiments, the sensor data is continuously obtained.

[0039] At S330, the obtained data is analyzed to determine actions to be performed by the AMA system. According to an embodiment, the obtained data includes sensor data and metadata (if applicable). In some embodiments, actions to be performed by the AMA system may include, but are not limited to, cutting (e.g., pruning, thinning, shearing, topping, raising, etc.), harvesting, a combination thereof, and the like.

[0040] In some embodiments, the obtained data is analyzed by machine vision, simultaneous localization and mapping (SLAM), machine learning, a combination thereof, and the like.

[0041] In an embodiment, the machine learning algorithm is supervised machine learning. According to an embodiment, the supervised machine learning process used in sensor data analysis involves training a machine learning model to recognize patterns and make predictions based on labeled data. The obtained data (as shown in S310, S320) is labeled. In an embodiment, sensor data includes but is not limited to images. In an embodiment, images within the obtained data are labeled such that each image is associated with a category or class that represents what is depicted in the image. As a non-limiting example, if images of plants are being analyzed, classes may include apple tree, grape vine, fig tree, etc.

[0042] The labeled data is cleaned by various processes, such as but not limited to, resizing images to a consistent size, normalizing pixel values, augmenting the data (e.g., rotating, flipping, or cropping images to increase the diversity of the dataset, etc.), a combination thereof, and the like. The labeled datasets are divided into two subsets.The subsets include a training set and a testing / validation set. The training set is used to train the machine learning model, while the testing / validation set is used to evaluate its performance. Relevant features from the image are then extracted. The features that are extracted may include, but are not limited to, color, histograms, texture descriptors, deep learning features extracted from convolutional neural networks (CNNs), a combination thereof, and the like. As a non-limiting example, if features from plant images are extracted, the features might include the hue of fruit and leaf, and texture features of the branches. The selected model is then trained using the training dataset. During training, the model learns to recognize patterns and features that differentiate between the different classes in the dataset. Once the model successfully performs training, the model can be deployed.

[0043] According to an embodiment, sensor data (e.g., images) after the execution of a task (e.g., pruning a branch, etc.) may be received by the machine learning model. The after-images (i.e., images performed after the actions taken by AMA system) may be utilized by the model for training purposes. As a non-limiting example, the AMA system may determine to prune a branch, and an image after pruning may be taken and fed into the machine learning model to further improve the model as to whether the determination to prune (or not) was correct.

[0044] According to an embodiment, the machine learning algorithm is a reinforcement learning algorithm. In some embodiments, the reinforcement learning algorithm performs actions to determine positive and negative behavior with the goal of maximizing the total cumulative reward (positive behavior).

[0045] According to an embodiment, the machine learning algorithm is unsupervised machine learning. In some embodiments, the unsupervised machine learning algorithm groups unlabeled data. The unsupervised machine learning algorithm can determine relationships between unlabeled data to discover inherent structure. It should be appreciated that other types of machine learning algorithms can be utilized herein as well.

[0046] At S340, instructions to execute actions by the AMA system are determined. In an embodiment, instructions may include, but are not limited to an execution assignment to individual pruning arms, movement of pruning arms, movement of the AMA system,movement of a vehicle coupled with the AMA system, the type of executable action (harvest, cut, etc.), the subtype of executable action (thinning, pruning, shearing, etc.), the angle of the executable action (e.g., 45-degree angle, etc.), the depth of the executable action (e.g., 6 inches from the branch collar, etc.), the strength of the executable action (e.g., 10 newtons of force, etc.), the duration of the executable action (e.g., slow cut, fast cut, etc.), a combination thereof, and the like. In some embodiments, the most optimal set of instructions is determined. As a non-limiting example, the path of the pruning arm may be optimized using A* pathfinding.

[0047] According to an embodiment, the instructions include optimal pruning arm pathfinding (i.e., to avoid other pruning arms, branches, obstacles, and the least distance to the execution for optimization, etc.), execution determination, type of execution (e.g., take pictures, prune, harvest, etc.), angle of execution, depth of execution, strength of execution, duration of execution, a combination thereof, and the like.

[0048] At S350, instructions are sent to be executed. According to an embodiment, the instructions are sent to the robot controller 151 (Fig. 1). In some embodiments, the instructions are sent to an assignment bank (e.g., a database 530, etc.).

[0049] Fig. 4 is an example flowchart 400 illustrating a method for simultaneously localizing and mapping (SLAM) an AMA system in an environment according to an embodiment. In an embodiment, the method is performed by the processing device 141 or the localization device 550, Fig. 5. In an embodiment, the method is performed with respect to the obtained input sensor data at S410, Fig. 4.

[0050] At S410, data is obtained, including but not limited to localization data and mapping data. In an embodiment, the mapping data represents data of an environment, e.g., the field. In an embodiment, the data can be obtained via a sensor (e.g., the sensors 131 , heat sensors, sound sensors, etc.), camera, machine vision, a data source (e.g., a landmark, semi-known area, previously visited area, etc.), GPS, a combination thereof, and the like. In an embodiment, the localization data represents locationality of a given component (e.g., UV, AMA system, pruning arms, etc.) with respect to its environment in a map (i.e., where the AMA system is within a vineyard). In some embodiments, the localization data can be obtained from, but is not limited to,localization devices (i.e., devices that measure movement in an environment), GPS, localizers, landmarks, cameras (e.g., the sensors 131), odometers, a combination thereof, and the like.

[0051] At S420, a three-dimensional (3D) map is generated based on the obtained mapping data. According to an embodiment, 3D points can be determined using, for example, various images. In an embodiment, the generated 3D maps include, but are not limited to, plants (e.g., branches, leaves, stems, etc.), the land (e.g., path within a vineyard, etc.), given components (e.g., the AMA system, pruning arms, UV, etc.), a combination thereof, and the like.

[0052] At S430, the location of a given component (e.g., the AMA system, pruning arms, UV, etc.) is determined within the 3D map. In some embodiments, the location of the given component can be determined utilizing known landmarks.

[0053] At S440, an action location is determined. According to an embodiment, the action location is a location to which the AMA system should move (i.e., utilizing a vehicle, moving from one location of a vineyard to a second location), a location to which the at least one pruning arm of the AMA system should move to (i.e., from a probing location to a position wherein the arm may prune a branch), or both.

[0054] As a non-limiting example, a user using an interface (e.g., operator control interface 161 ) may prompt the AMA system to a specific path or task (e.g., prune the entire vineyard utilizing a snake path, etc.). As another non-limiting example, the AMA system, or a component therein, may determine a specific path or task based on e.g., mapping data, localization data, machine vision, etc.

[0055] At S450, instructions to perform the action are sent to the system. According to an embodiment, the instructions to perform the action may be sent to the processing device 141 , the localization device 550, a combination thereof, and the like. In some embodiments, the instructions to perform the action include a path for the AMA system (or a component coupled with) to take on land (or any other navigable surface), a path for at least one pruning arm of the AMA system to take, a type of action to be performed at a target location (e.g., prune, harvest, rotate, swivel, slow down, speed up, etc.), a combination thereof, and the like.

[0056] At S460, it is determined whether the execution of the method should continue, and if so, execution continues with S430, otherwise, execution ends.

[0057] Fig. 5 shows an example network diagram 500 utilized to describe the various disclosed embodiments. In the example network diagram 500, a processing device 141 , a database 530, a user device 540, and a localization device 550 communicate via a network 510. The network 510 may be, but is not limited to, a wireless, cellular, or wired network, a local area network (LAN), a wide area network, a metro area network (MAN), the Internet, the worldwide web (WWW), similar networks, and any combination thereof.

[0058] The processing device 141 is configured to send instructions for executing actions described herein, for example as described above with respect to Fig. 3. In another embodiment, the processing device 141 may be configured to perform simultaneous localization and mapping (SLAM) of the AMA system in an environment, for example as described above with respect to Fig. 4. In some embodiments the SLAM may include pathing of a UV the AMA system is coupled with, or the pathing of pruning arms (or other components of the AMA system) in an environment. As non-limiting examples, the AMA could be coupled with a UV in which it autonomously moves across a farm using SLAM, or SLAM is used to move a pruning arm to prune a branch while avoiding other pruning arms and branches.

[0059] Database 530 may be configured to store metadata and sensor data to be sent to the processing device 141 , for example as described above with respect to Fig. 3. In an embodiment, the database may be configured to store determined executable actions to be synchronized in which executable actions can be assigned to individual pruning arms, for example as described above with respect to Fig. 2.

[0060] The user device 540 may be, but is not limited to, a personal computer, a laptop, a tablet computer, a smartphone, a wearable computing device, a touchscreen, any device capable of receiving and displaying notifications, a combination thereof, and the like. In an embodiment the user device 540 may be utilized to input metadata into the AMA system, for example as described above with respect to Fig. 3.

[0061] The localization device 550 may be configured to perform SLAM of the AMA system or an unmanned vehicle coupled with the AMA, for example as described above with respect to Fig. 4.

[0062] A hardware layer of computing components which may be utilized to realize any or all of the processing device 141 , the database 530, the user device 540, or the localization device 550, is described further below with respect to Fig. 6.

[0063] Fig. 6 is an example schematic diagram of hardware layer 600 which may be utilized according to various disclosed embodiments. The hardware layer 600 or a similarly configured hardware layer may be utilized to realize one or more of the disclosed embodiments, for example, in use as part of the processing device 141 , the database 530, the user device 540, or the localization device 550, Fig. 5.

[0064] The hardware layer 600 includes a processing circuitry 610 coupled to a memory 620, a storage 630, and a network interface 640. In an embodiment, the components of the hardware layer 600 may be communicatively connected via a bus 650.

[0065] The processing circuitry 610 may be realized as one or more hardware logic components and circuits. For example, and without limitation, illustrative types of hardware logic components that can be used include field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), Application-specific standard products (ASSPs), system-on-a-chip systems (SOCs), graphics processing units (GPUs), tensor processing units (TPUs), general-purpose microprocessors, microcontrollers, digital signal processors (DSPs), and the like, or any other hardware logic components that can perform calculations or other manipulations of information.

[0081] The memory 620 may be volatile (e.g., random access memory, etc.), nonvolatile (e.g., read only memory, flash memory, etc.), or a combination thereof.

[0066] In one configuration, software for implementing one or more embodiments disclosed herein may be stored in the storage 630. In another configuration, the memory 620 is configured to store such software. Software shall be construed broadly to mean any type of instructions, whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise. Instructions may include code (e.g., in source code format, binary code format, executable code format, or any other suitable format of code). The instructions, when executed by the processing circuitry 610, cause the processing circuitry 610 to perform the various processes described herein.

[0067] The storage 630 may be magnetic storage, optical storage, and the like, and may be realized, for example, as flash memory or other memory technology, compact diskread only memory (CD-ROM), Digital Versatile Disks (DVDs), or any other medium which can be used to store the desired information.

[0068] The network interface 640 allows the hardware layer 600 to communicate with, for example, the processing device 141 , the database 530, the user device 540, the localization device 550, or a combination thereof.

[0069] It should be understood that the embodiments described herein are not limited to the specific architecture illustrated in Fig. 6, and other architectures may be equally used without departing from the scope of the disclosed embodiments.

[0070] It is important to note that the embodiments disclosed herein are only examples of the many advantageous uses of the innovative teachings herein. In general, statements made in the specification of the present application do not necessarily limit any of the various claimed embodiments. Moreover, some statements may apply to some inventive features but not to others. In general, unless otherwise indicated, singular elements may be in plural and vice versa with no loss of generality. In the drawings, like numerals refer to like parts through several views.

[0071] The various embodiments disclosed herein can be implemented as hardware, firmware, software, or any combination thereof. Moreover, the software is preferably implemented as an application program tangibly embodied on a program storage unit or computer readable medium consisting of parts, or of certain devices and / or a combination of devices. The application program may be uploaded to, and executed by, a machine comprising any suitable architecture. Preferably, the machine is implemented on a computer platform having hardware such as one or more central processing units (“CPUs”), a memory, and input / output interfaces. The computer platform may also include an operating system and microinstruction code. The various processes and functions described herein may be either part of the microinstruction code or part of the application program, or any combination thereof, which may be executed by a CPU, whether or not such a computer or processor is explicitly shown. In addition, various other peripheral units may be connected to the computer platform such as an additional data storage unit and a printing unit. Furthermore, a non-transitorycomputer-readable medium is any computer-readable medium except for a transitory propagating signal.

[0072] All examples and conditional language recited herein are intended for pedagogical purposes to aid the reader in understanding the principles of the disclosed embodiment and the concepts contributed by the inventor to furthering the art and are to be construed as being without limitation to such specifically recited examples and conditions. Moreover, all statements herein reciting principles, aspects, and embodiments of the disclosed embodiments, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof. Additionally, it is intended that such equivalents include both currently known equivalents as well as equivalents developed in the future, i.e., any elements developed that perform the same function, regardless of structure.

[0073] It should be understood that any reference to an element herein using a designation such as “first,” “second,” and so forth does not generally limit the quantity or order of those elements. Rather, these designations are generally used herein as a convenient method of distinguishing between two or more elements or instances of an element. Thus, a reference to first and second elements does not mean that only two elements may be employed there or that the first element must precede the second element in some manner. Also, unless stated otherwise, a set of elements comprises one or more elements.

[0074] As used herein, the phrase “at least one of” followed by a listing of items means that any of the listed items can be utilized individually, or any combination of two or more of the listed items can be utilized. For example, if a system is described as including “at least one of A, B, and C,” the system can include A alone; B alone; C alone; 2A; 2B; 2C; 3A; A and B in combination; B and C in combination; A and C in combination; A, B, and C in combination; 2A and C in combination; A, 3B, and 2C in combination; and the like.

Claims

CLAIMSWhat is claimed is:1 . A method for performing an optimal pruning action on a plant using an autonomous mountable attachment (AMA) system, comprising: receiving plant location data; analyzing the received plant location data, using at least one artificial intelligence (Al) model trained to determine an optimal pruning action on a plant; determining at least one executable instruction for performing an optimal pruning action, based on the analyzed plant location data; and sending the determined at least one executable instruction to a pruning arm controller, wherein the pruning arm controller causes a pruning arm to perform the optimal pruning action on the plant.

2. The method of claim 1 , further comprising: receiving metadata from at least one of: a data source and a user input.

3. The method of claim 2, wherein the metadata includes: a plant type, specific plant varieties, executable action varieties, weather, and other plant-related issues.

4. The method of claim 1 , wherein the optimal pruning action is any one of: a cutting action, a touching action, a thinning action, a shearing action, a topping action, a raising action, and a harvesting action.

5. The method of claim 1 , wherein analyzing the received plant location data further comprises: extracting a set of features from the received plant location data; identifying at least one pattern based on the extracted set of features; and classifying the plant location data into at least one category, wherein classifying is based on similarity of the identified at least one pattern with at least one pattern classified into at least one category during training of the at least one Al model.

6. The method of claim 1 , wherein analyzing the received plant location data further comprises: receiving state information based on the received plant location data; and determining at least one action that maximizes a cumulative reward given the received state information, based on a value function, policy, and / or model learned during training of the at least one Al model.

7. The method of claim 1 , wherein the received plant location data includes at least one of: a plant, a plant subcomponent, a plant location, a location of the AMA system, and a location of at least one component associated with the AMA system.

8. The method of claim 1 , wherein the at least one executable instruction for performing an optimal pruning action further includes any one of: assigning executable actions to a plurality of pruning arms, moving at least one pruning arm of the plurality of pruning arms, moving the AMA system, moving a vehicle coupled to the AMA system, and moving an attachment tool of the at least one pruning arm.

9. An autonomous mountable attachment (AMA) system for performing an optimal pruning action on a plant, comprising: at least one pruning arm configured to perform an optimal pruning action on a plant; at least one attachment tool affixed to the at least one pruning arm; a plurality of sensors affixed to the at least one pruning arm, configured to retrieve plant location data; a pruning arm controller configured to cause the at least one pruning arm to perform the optimal pruning action on the plant; a processing device communicatively connected to the plurality of sensors and the pruning arm controller, and configured to receive plant location data from theplurality of sensors and to determine and send at least one executable instruction to the pruning arm controller; and a vehicle mount attachment connection configured to removably affix the at least one pruning arm to at least one unmanned vehicle.

10. The AMA system of claim 9, wherein the plurality of sensors includes at least one of: a two-dimensional (2D) camera, a three-dimensional (3D) camera, a thermal camera, an infrared camera, a video camera, a light detection and ranging (LiDAR) camera, and an ultrasonic measuring device.11 . The AMA system of claim 9, wherein the at least one pruning arm is a plurality of pruning arms, each pruning arm being independently adjustable from the other pruning arms.

12. The AMA system of claim 11 , further comprising: at least one assignment bank configured to assign at least one executable instruction to at least one pruning arm in the plurality of pruning arms.

13. The AMA system of claim 11 , wherein the AMA system is configured to determine that at least one pruning arm in the plurality of pruning arms is malfunctioning and to assign at least one executable instruction that was assigned to the at least one malfunctioning pruning arm to another pruning arm.

14. The AMA system of claim 9, wherein the at least one executable instruction includes at least one of: assigning executable actions to a plurality of pruning arms, optimal pathfinding of the at least one pruning arm, moving the AMA system, moving a vehicle coupled to the AMA system, determining a type of executable action, determining a sub-type of executable action, determining a depth of the executable action, determining a strength of the executable action, and determining a duration of the executable action.

15. The AMA system of claim 9, wherein the at least one pruning arm is height adjustable.

16. The AMA system of claim 11 , wherein each pruning arm of the plurality of pruning arms is configured at a different height than each other pruning arm of the plurality of pruning arms.

17. The AMA system of claim 9, wherein the at least one attachment tool includes at least one of: a cutting tool, a touching tool, a pruning tool, a thinning tool, a shearing tool, a topping tool, a raising tool, and a harvesting tool.

18. The AMA system of claim 10, wherein the 3D camera is configured to determine an optimal angle at which the at least one pruning arm performs the optimal pruning action.

19. A non-transitory computer readable medium having stored thereon instructions for causing a processing circuitry to execute a process, the process comprising: receiving plant location data; analyzing the received plant location data, using at least one artificial intelligence (Al) model trained to determine an optimal pruning action on a plant; determining at least one executable instruction for performing an optimal pruning action, based on the analyzed plant location data; and sending the determined at least one executable instruction to a pruning arm controller, wherein the pruning arm controller causes a pruning arm to perform the optimal pruning action on the plant.

20. A system for controlling an AMA system, comprising: a processing circuitry; and a memory, the memory containing instructions that, when executed by the processing circuitry, configure the system to: receive plant location data;analyze the received plant location data, using at least one artificial intelligence (Al) model trained to determine an optimal pruning action on a plant; determine at least one executable instruction for performing an optimal pruning action, based on the analyzed plant location data; and send the determined at least one executable instruction to a pruning arm controller, wherein the pruning arm controller causes a pruning arm to perform the optimal pruning action on the plant.

Images