DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
The Amendment filed on 04/22/2026 has been entered. Claims 1-20 are pending in the application.
Response to Arguments
Regarding the rejections of claims 1-20 under 35 U.S.C. 103, Applicant's arguments filed 04/22/2026 have been fully considered but they are not persuasive.
Applicant argues that “King, Shi and Gauvreau, taken singly or in combination, do not teach or suggest ‘wherein an amount of the impulse force generated is determined by a reinforcement learning model where a reward is calculated based on a rate of change of a mobility path to a correct mobility path and a penalty is calculated based on predicted damage to the plant’ as recited in amended claim 1 and similarly in amended claims 8 and 15.”
This argument is moot because the additional/modified claim language by the amendment necessitates new grounds of rejection with other prior art.
Applicant further argues “King, Shi and Gauvreau, taken singly or in combination, do not teach or suggest ‘instructing the robotic bee to generate an impulse force based on the type of the plant and the classified level of threat’” (Applicant’s Remarks, pg. 8). Specifically, Applicant argues that “There is no language in the cited passages of King and Gauvreau that teaches or suggests instructing the robotic bee to generate an impulse force” “based on the type of the plant and the classified level of threat” (Applicant’s Remarks, pg. 11), “there is no language in King regarding an associated classified level of threat” (pg. 11), and “There is no language in Gauvreau that such a modification [of a flight plan] involves generating an impulse force based on the type of plant and the classified level of threat” (pg. 11).
Examiner respectfully disagrees for at least the reasons explained in the Final Rejection of 02/05/2026. In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986).
Nevertheless, amended claim 1 now stands rejected under the combination of Baldwin (US 20240233560 A1), King, Gauvreau, Sun, and Robertson.
Baldwin discloses “classifying a level of threat” in [0026-0030] and “automatically initiat[ing] a corrective action based on the threat score” [0030]. A person of ordinary skill in the art would recognize that an aircraft with an “unsuspicious” (see Fig. 7) threat score would not require a corrective action. In combination, this corrective action is sending an emergency stop command to the robotic bee as taught by King.
King discloses “If the safety zone control 818 detects an impending hazard, the safety zone control 818 may be configured to control the UAV to take remedial action. For example, the safety zone control 818 may send an emergency stop command to the UAV to trigger the UAV to land, pause, and/or to send an updated navigation command to the UAV to alter a current path of the UAV” [0106]. An impending hazard indicates a classification of threat level (i.e. the threat level is ‘hazardous’ versus ‘non-hazardous’) of, for example, “potentially conflicting traffic (e.g., workers, other UAVs, vehicles, robots, etc.)” [0106]. A command to land and/or alter a path of the UAV to avoid such a hazard is an instruction to generate an impulse force because controlling the movement of the UAV onto an updated path away from the hazard requires force generation. See “The navigation commands may include commands to control thrust, yaw, pitch, roll, etc., of the UAV such that the UAV follows the predefined path” [0105].
Where a threat score is based at least partly on deviation from the planned/filed flight path and/or approaching known obstacles/hazards (Baldwin, [0029]), a corrective action, for example, of updating the current path of the UAV as taught by King [0106] towards the planned/filed flight path or away from obstacles/hazards, to decrease the threat score results in the UAV generating a fast-acting force to move away from the area causing the threat (an impulse force). Therefore, Baldwin/King discloses “instructing the robotic bee to generate an impulse force based on… the classified level of threat.”
Additionally, Gauvreau discloses “the propulsion system [of a UAV] includes a first motor and a second motor, and the navigation system is configured to operate the first motor to produce a first compression wave and operate the second motor to produce a second compression wave offset in phase from the first compression wave such that the first and second compression waves combine to produce the thrust [force] at a frequency that induces the plant to release pollen” [0040]. Because of Newton’s Third Law, generating a force towards the plant means that the UAV moves away from the plant. Since controlling the movement of the UAV requires force generation, Gauvreau also discloses that the force generated by the UAV depends on the type of plant; see [0051-0052] and [0066]. Therefore, Baldwin/King/Gauvreau discloses “instructing the robotic bee to generate an impulse force based on the type of the plant.”
The combination of Baldwin/King/Gauvreau/Sun/Robertson teaches the limitation “instructing the robotic bee to generate an impulse force based on the type of the plant and the classified level of threat.”
In conclusion, the combination of Baldwin/King/Gauvreau/Sun/Robertson as a whole teaches all the limitations of amended claim 1. Since the arguments against the rejections of claims 2-20 are significantly similar to those of amended claim 1, Examiner maintains the rejections of claims 2-20 on the same grounds as the rejection of amended claim 1.
Claim Objections
Claims 6-7, 13-14, and 20 are objected to because of the following informalities:
In claims 6, 13, and 20, “a reinforcement learning model to determine an amount of the impulse force” should read “the reinforcement learning model to determine the amount of the impulse force” because a reinforcement learning model and an amount of the impulse force are previously recited in claims 1, 8, and 15, respectively.
In claims 7 and 14, “wherein a reward is calculated based on a rate of change of a mobility path to a correct mobility path, wherein a penalty is calculated” should read “wherein the reward is calculated based on the rate of change of the mobility path to the correct mobility path, wherein the penalty is calculated” because a reward, rate of change of a mobility path to a correct mobility path, and a penalty are all previously recited in claim 1.
In claims 7 and 14, “damage” should read “predicted damage” or in claims 1 and 8, “predicted damage” should read “damage”. In view of [0100-0104], the “predicted damage” recited in claims 1, 8, and 15 is the same as the “damage” recited in claims 7 and 14 because the “predicted” damage is an assessment of damage already imparted to the plant.
Appropriate correction is required.
Claim Interpretation
The following definitions from the specification were used in interpreting the claims:
“An impulse force, as used herein, is a fast-acting force, which is utilized by the robotic bee to move away from the area (e.g., sticky surface of plant) causing the threat.” ([0020], [0021], [0028], [0045], [0096], [0160], [0179])
“An "unbalanced operation," as used herein, refers to movements and actions of the robotic bee causing the robotic bee to deviate from a normal mobility path.” ([0021], [0041], [0079], [0142], [0179])
“A ‘level of threat,’ as used herein, refers to an indication as to a likelihood of the robotic bee not being able to complete its requested operation due to an external influence (e.g., being stuck on a sticky surface, landing on an uneven surface structure, etc.).” ([0021], [0044], [0179])
“A mobility path, as used herein, refers to the path of movement to perform the requested operation. The rate of change of a mobility path, as used herein, refers to the rate of changing the mobility path.” ([0048], [0099], [0166])
According to paragraph [0113], the computer-readable storage mediums in a computer program product of claim 8 are non-transitory storage.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(d):
(d) REFERENCE IN DEPENDENT FORMS.—Subject to subsection (e), a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers.
The following is a quotation of pre-AIA 35 U.S.C. 112, fourth paragraph:
Subject to the following paragraph [i.e., the fifth paragraph of pre-AIA 35 U.S.C. 112], a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers.
Claims 7 and 14 are rejected under 35 U.S.C. 112(d) or pre-AIA 35 U.S.C. 112, 4th paragraph, as being of improper dependent form for failing to further limit the subject matter of the claim upon which it depends, or for failing to include all the limitations of the claim upon which it depends.
Claim 7 fails to further limit the subject matter of claim 6, which is dependent upon claim 1, because the limitations of claim 7 are included in claim 1. Similarly, claim 14 fails to further limit the subject matter of claim 13, which is dependent upon claim 8, because the limitations of claim 14 are included in claim 8.
Claims 1 and 8 recite the limitation “where a reward is calculated based on a rate of change of a mobility path to a correct mobility path and a penalty is calculated based on predicted damage to the plant.” As stated above, the “predicted damage” recited in claims 1 and 8 is the same as the “damage” recited in claims 7 and 14 because the “predicted” damage is an assessment of damage already imparted to the plant (see [0100-0104] of the specification). Therefore, the limitations of claims 7 and 14—“wherein a reward is calculated based on a rate of change of a mobility path to a correct mobility path, wherein a penalty is calculated based on damage to the plant”—do not further limit claims 1, 6, 8, and 13.
Applicant may cancel the claim(s), amend the claim(s) to place the claim(s) in proper dependent form, rewrite the claim(s) in independent form, or present a sufficient showing that the dependent claim(s) complies with the statutory requirements.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 3, 5-8, 10, 12-15, 17, and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Baldwin et al. (US 20240233560 A1, filed 01/12/2023; hereafter “Baldwin”) in view of King (US 20190303668 A1), Gauvreau, Jr. (US 20200022312 A1; hereafter “Gauvreau”), Sun et al. (JP 2021034050 A; hereafter “Sun”), and Robertson et al. (US 2019261565 A1; hereafter “Robertson”). Citations of Sun refer to the English translation provided by Examiner.
Regarding claim 1, Baldwin, in the same field of endeavor (detecting anomalous aircraft flights), discloses
detecting an unbalanced operation of the [aircraft]; classifying a level of threat of the [aircraft] not being able to complete a requested operation based on received images of the operation of the [aircraft] in response to the detected unbalanced operation of the [aircraft] (The EDT [Early Detection of Threats] system provides a machine learning-based “threat score for each active flight that address[es] two essential questions: (1) is the flight behaving normally for an aircraft of that type, and (2) is the flight where it is supposed to be relative to a flight plan and historical flight patterns” [0026]. At 102 (see Fig. 1), flight data is received from an aircraft [0028], but flight data may also be received from “visual systems” (received images) [0162]. Flight data may include “an aircraft's GPS location, altitude, ground speed, flight direction, and other data” [0039]; see also [0047], [0137], and [0140]. At 104, the system classifies, identifies, and reports “suspicious activity” of aircrafts [0029]. See “The classification schema may be any suitable method… and may compare flight tracks of real time aircraft position data against historical flight paths, planned and filed flight paths, and/or flight paths relative to known obstacles, hazards, restrictions, etc.” [0029]. For example, the planned/filed flight path is a requested operation, and the unbalanced operation is a flight track that deviates from the planned/filed flight path and/or approaches known obstacles/hazards. See also aircraft threats by filter 704 (“Highly suspicious”, "More suspicious", "Suspicious", and "Unsuspicious") in Fig. 7 and "Top Threats view 800" in Fig. 8. See also [0027], [0030], and [0033].).
Baldwin additionally discloses that “the system may automatically initiate a corrective action based on the threat score” [0030]. A person of ordinary skill in the art would recognize that an aircraft with an “unsuspicious” (see Fig. 7) threat score would not require a corrective action.
However, Baldwin does not explicitly teach “an unbalanced operation of the robotic bee being performed on a plant; determining a type of the plant using image data of the plant in response to the detected unbalanced operation of the robotic bee; …and instructing the robotic bee to generate an impulse force based on the type of the plant and the classified level of threat, wherein the impulse force is generated by the robotic bee, wherein an amount of the impulse force generated is determined by a reinforcement learning model where a reward is calculated based on a rate of change of a mobility path to a correct mobility path and a penalty is calculated based on predicted damage to the plant.”
King, in the same field of endeavor (a system for monitoring and managing plant growth with unmanned aerial vehicles), teaches
detecting an… operation of the robotic bee being performed on a plant (Robotic bee: UAV 112. Detecting an operation: “The visual observer devices 106 can monitor [detect] and control the flight operations of one or more UAVs [unmanned aerial vehicles] 112 in the grow operation 104” [0038]. After identifying an issue with a plant, the “image analysis services may then communicate the recommendations for potential remedial courses of action to the UAV to perform one or more operations. For instance, the UAV may be configured to deliver fertilizer, water, and/or so forth” [0079]. See also [0078], [0087], [0093], [0106], and [0113]. See also [0116] where a UAV may be malfunctioning.);
determining a type of the plant using image data of the plant in response to the detected… operation of the robotic bee (See Fig. 15: “At block 1502, the visual observer device determines the position of a UAV within a grow operation. … At block 1506, the visual observer device receives, from the UAV, one or more images of a plant included in the grow operation as the UAV traverses the path” [0157]. After images of the plant are deemed suitable for further processing (step 1508 in Fig. 15 / step 1110 in Fig. 11), the images are processed according to the flowchart of Fig. 11, where a target plant is identified in step 1120. “The processing of the images 502 by image analysis services may include classifying the plant 110 as a certain plant type” [0050]. See also [0027], [0029, [0034], [0049], and [0063]-[0067] for more on image data.);
classifying a level of threat of the robotic bee not being able to complete a requested operation based on received images of the operation of the robotic bee in response to the detected… operation of the robotic bee (See “the visual observer device 106 [which takes/receives images of the operation of UAV 112] may include a safety zone control 818 that is configured to scan the area around the UAV for potentially conflicting traffic (e.g., workers, other UAVs, vehicles, robots, etc.) [in other words, whether or not the UAV can continue along its assigned path/operation]… If the safety zone control 818 detects an impending hazard, the safety zone control 818 may be configured to control the UAV to take remedial action,” [0106]. In this case, the threat level of the potentially conflicting traffic is classified as either hazardous or non-hazardous. See also [0117] where a UAV may be malfunctioning and the controller 116 instructs the UAV 112 to abort its assigned mission.); and
instructing the robotic bee to generate an impulse force based on… the classified level of threat (See “If the safety zone control 818 detects an impending hazard, the safety zone control 818 may be configured to control the UAV to take remedial action. For example, the safety zone control 818 may send an emergency stop command [instruction] to the UAV to trigger the UAV to land, pause, and/or to send an updated navigation command to the UAV to alter a current path of the UAV,” [0106]. Again, in this case, the threat level is classified as either hazardous or non-hazardous. By the definition of an impulse force (a fast-acting force utilized to move away from the area causing the threat), an emergency command to land the UAV and a navigation command to alter a current path of the UAV to avoid a detected hazard are instructions to generate an impulse force. See also [0105] and [0117].),
wherein the impulse force is generated by the robotic bee (The command to generate an impulse force in [0106] is implemented by UAV control module 718 (in UAV 112; see Fig. 7), which “control[s] the operation and movement of the UAV 112” with motors 706 [0086-0087]. Controlling the movement of the UAV 112 requires the UAV to generate force. See also [0083-0087], [0105], and [0117].).
King discloses that both the central controller 116/connected visual observer device 106 and the controller of UAV 112 can visually track position of the UAV(s) 112 over time: see [0038]-[0043] for visual positioning and control of UAVs 112 by the central controller 116/connected visual observer device 106, and see “Whether in autonomous mode or remotely-piloted mode, the UAV control module 718 [stored in the memory of UAV 112] controls the operation of UAV 112. This control includes the use of outputs from the positioning module 714, sensors 724, and/or camera 726. In one example, the positioning module 714 may interface with one or more hardware sensors that determine the location/position of the UAV 112, detect other aerial drones and/or obstacles and/or physical structures around UAV 112, measure the speed and direction of the UAV 112, and provide any other inputs needed to safely control the movement of the UAV 112” [0087]. Since Baldwin teaches detecting an unbalanced operation with data from visual systems (like those of King above) and classifying a level of threat (see citations of Baldwin above), the combination of Baldwin/King teaches “determining a type of the plant using image data of the plant in response to the detected unbalanced operation of the robotic bee;” and “classifying a level of threat of the robotic bee not being able to complete a requested operation based on received images of the operation of the robotic bee in response to the detected unbalanced operation of the robotic bee.”
Furthermore, Baldwin additionally discloses that “the system may automatically initiate a corrective action based on the threat score” [0030]. In combination with King’s remedial action “to trigger the UAV to land, pause, and/or to send an updated navigation command to the UAV to alter a current path of the UAV” to avoid an impending hazard (threat) [0106], Baldwin/King teaches “instructing the robotic bee to generate an impulse force based on… the classified level of threat.” Where a threat score is based at least partly on deviation from the planned/filed flight path and/or approaching known obstacles/hazards (Baldwin, [0029]), a corrective action, for example, of updating the current path of the UAV as taught by King [0106] towards the planned/filed flight path or away from obstacles/hazards, to decrease the threat score results in the UAV generating a fast-acting force to move away from the area causing the threat (an impulse force).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified the aircraft threat detection method of Baldwin with the UAV observation and control method of King. One of ordinary skill in the art would have been motivated to make this modification “to perform the one or more operations in the grow operation” (King, Abstract).
However, Baldwin/King does not explicitly teach “instructing the robotic bee to generate an impulse force based on the type of the plant… wherein an amount of the impulse force generated is determined by a reinforcement learning model where a reward is calculated based on a rate of change of a mobility path to a correct mobility path and a penalty is calculated based on predicted damage to the plant.”
Gauvreau, in the same field of endeavor (artificial pollination by UAV), teaches
…an …operation of the robotic bee being performed on a plant (See “the propulsion system [of a UAV] includes a first motor and a second motor, and the navigation system is configured to operate the first motor to produce a first compression wave and operate the second motor to produce a second compression wave offset in phase from the first compression wave such that the first and second compression waves combine to produce the thrust at a frequency that induces the plant to release pollen” [0040].);
determining a type of the plant using image data of the plant… (See “the navigation system includes a computer vision system (not shown) configured to identify a type of plant 606 and determine the flight plan 604 based on the type” [0066].);
instructing the robotic bee to generate a… force based on the type of the plant… (Generate a force: see “the propulsion system [of a UAV] includes a first motor and a second motor, and the navigation system is configured to operate the first motor to produce a first compression wave and operate the second motor to produce a second compression wave offset in phase from the first compression wave such that the first and second compression waves combine to produce the thrust [force] at a frequency that induces the plant to release pollen” [0040]. See “generating a thrust that produces a frequency that induces the plant to release pollen includes operating first motor 404 to produce a first compression wave and operating second motor 406 to produce a second compression wave that is offset in phase from the first compression wave such that the first and second compression waves combine to produce the thrust at a frequency that induces the plant to release pollen… In some embodiments, the frequency is about 200 times per second to induce an auto-pollinating plant to release pollen. In some embodiments, the frequency is about 400 times per second to induce a cross-pollinating plant to release pollen” [0051]-[0052]. The frequency of the thrust is determined by the type of plant: auto-pollinating or cross-pollinating. When the frequency is faster, more waves (amount of force) are applied to the plant (a fast-acting force). Additionally, when combining two waves offset in phase, the maximum amplitude (amount of force) may increase; that the combined amplitude is greater than the amplitude of either waveform alone is true for the waveforms shown in Fig. 3. Therefore, Gauvreau teaches that the maximum force (thrust) is determined by the type of the plant. Because of Newton’s Third Law, generating a force towards the plant means that the UAV moves away from the plant, as required by an impulse force. See also [0066].),
wherein the… force is generated by the robotic bee (Above, the navigation system instructs the propulsion system of UAV 102 to generate the impulse force [0040]. See “Unmanned aerial vehicle 102 includes propulsion system 104 producing a thrust 106, and a navigation system (not shown) controlling the propulsion system. The navigation system flies unmanned aerial vehicle 102 above plant 108 such that thrust 106 contacts plant 108” [0036]. See also [0007] and [0037-0040].).
Baldwin/King teaches “instructing the robotic bee to generate an impulse force based on… the classified level of threat” (see citations above), and Gauvreau teaches “instructing the robotic bee to generate a… force based on the type of the plant” (see citations above). In combination, Baldwin/King/Gauvreau teaches “instructing the robotic bee to generate an impulse force based on the type of the plant and the classified level of threat.”
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified the UAV observation and control method of Baldwin/King with the pollination method of Gauvreau. One of ordinary skill in the art would have been motivated to make this modification for the benefit of “more efficient and cost-effective artificial plant pollination” (Gauvreau, [0034]).
However, Baldwin/King/Gauvreau does not explicitly teach “wherein an amount of the impulse force generated is determined by a reinforcement learning model where a reward is calculated based on a rate of change of a mobility path to a correct mobility path and a penalty is calculated based on predicted damage to the plant.”
Sun, in the same field of endeavor (path planning for autonomous robots), teaches
wherein an amount of the impulse force generated is determined by a reinforcement learning model (See “the actions in the reinforcement learning algorithm correspond to the thrusts and torques [impulse forces] of the robot, and the states in the algorithm correspond to the velocities and angular velocities of the robot” [0041]. The reward function is described in [0035], with the first term relating to obstacle (threat) avoidance: “when the AUV approaches the obstacle, it gets a negative reward warning, when the AUV moves away from the obstacle, it gets a positive reward, …when it collides with the obstacle, it gets a reward value of -1 and ends the current training cycle.” As in King, the threat level is classified as either hazardous (obstacle) or non-hazardous (not an obstacle). There is further reward for reaching the target speed: “the reward value r4＝-|△v+△Ψ| is designed using the errors between the speed, angular velocity and the target speed and target angular velocity” [0036]. Using rewards for matching target speeds and avoiding obstacles, the reinforcement learning model learns the amount of impulse force to generate.)
where a reward is calculated based on a rate of change of a mobility path to a correct mobility path (There is a reward for reaching the target speed and heading: “the reward value r4＝-|△v+△Ψ| is designed using the errors between the speed, angular velocity and the target speed and target angular velocity, where △v is the speed error and △Ψ is the heading error.” [0036]. A smaller speed error and smaller heading error result in a larger reward r4. The current speed and angular velocity represent the current mobility path, and the target speed and target angular velocity represent the correct mobility path. The total reward for the training cycle (over multiple timesteps) is therefore calculated by incorporating r4 per timestep (rate of change).).
While Sun applies the reinforcement learning model to an autonomous underwater vehicle, the reward takes into account only target points, obstacles, and the vehicle’s own movement. Detection of a UAV’s position and speed have been disclosed by Baldwin and King as stated previously above. King further discloses detection of obstacles [0106]. King also discloses that a planned path may “comprise a series of waypoints,” which are similar to Sun’s target points [0105]. Thus, Sun’s reinforcement learning model is suitable for combination with Baldwin/King/Gauvreau.
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified the UAV observation and control method of Baldwin/King/Gauvreau with the reinforcement learning model of Sun. One of ordinary skill in the art would have been motivated to make this modification “so that a complete control system based on DDPG with anti-interference capability is obtained through training” (Sun, [0036]).
However, Baldwin/King/Gauvreau/Sun does not explicitly teach “a penalty is calculated based on predicted damage to the plant.”
Robertson, in the same field of endeavor (agricultural optimization), teaches
a penalty is calculated based on predicted damage to the plant (See “To train a control policy via reinforcement learning, it is necessary to define a utility function that rewards success (in this case picking saleable fruit) and penalizes cost (e.g. time spent, energy consumed, etc.).” [0248]. In this case, the quality control subsystem grades the quality of fruit, and the control subsystem is trained “not to pick a particular target fruit, e.g. because the expected cost of picking in terms of damage to the plant or picked fruit will not outweigh the benefit provided by having one more picked fruit” [0271]. Therefore, the expected cost of predicted damage to the plant or picked fruit is a calculated penalty. See also [0161].).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified the UAV control method of Baldwin/King/Gauvreau/Sun to include a quality control method as taught by Robertson. One of ordinary skill in the art would have been motivated to make this modification for the benefit of avoiding damage to plants (Robertson, [0271]).
Regarding claim 3, Baldwin/King/Gauvreau/Sun/Robertson discloses the limitations of claim 1 as addressed above, and Gauvreau additionally discloses
wherein the impulse force is generated via compressed air (See “the propulsion system [of a UAV] includes a first motor and a second motor, and the navigation system is configured to operate the first motor to produce a first compression wave;” the compression wave is compressed air [0040].).
Regarding claim 5, Baldwin/King/Gauvreau/Sun/Robertson discloses the limitations of claim 1 as addressed above, and King additionally discloses
receiving images of an area of activity from a plurality of robotic bees (Robotic bees: UAVs 112. See “the image analysis services 128 may transmit commands to one or more UAVs 112 to fly over a particular path within the grow operation [area of activity] 104 to capture images or collect other sensor data” [0034]. Images received by a computer: “the UAV 112… may transmit images and other information to a central computing device 126 for providing image analysis services 128” [0029].),
wherein the area of activity corresponds to an area where the plurality of robotic bees are instructed to perform various operations (See “The central controller 116 may include one or more cameras to monitor the operation of the UAVs [robotic bees] 112 within each respective grow operation [area of activity] 104,” [0038]. Various operations: “the UAV 112 may be instructed to deliver fertilizer, water, and/or tools to an operation zone in a grow operation” [0044]. See also [0026] and [0079].),
wherein the images of the area of activity comprise the image data of the plant (Images of the area of activity: see “the image analysis services 128 may transmit commands to one or more UAVs 112 to fly over a particular path within the grow operation [area of activity] 104 to capture images or collect other sensor data” [0034]. Those images comprise the image data of the plant: “The UAV is configured to capture one or more images 502 of plants 110 or groups of plants as the UAV traverses the path” [0049]. See also [0027].).
Regarding claim 6, Baldwin/King/Gauvreau/Sun/Robertson disclose the limitations of claim 1 as addressed above, and Baldwin additionally discloses that “the system may automatically initiate a corrective action based on the threat score” [0030].
King additionally teaches determining generation of an impulse force by the robotic bee based on the classified level of threat (See “If the safety zone control 818 detects an impending hazard, the safety zone control 818 may be configured to control the UAV to take remedial action. For example, the safety zone control 818 may send an emergency stop command [instruction] to the UAV to trigger the UAV to land, pause, and/or to send an updated navigation command to the UAV to alter a current path of the UAV,” [0106]. Again, in this case, the threat level is classified as either hazardous or non-hazardous. By the definition of an impulse force (a fast-acting force utilized to move away from the area causing the threat), an emergency command to land the UAV and a navigation command to alter a current path of the UAV to avoid a detected hazard are instructions to generate an impulse force. See also [0105] and [0117].).
Gauvreau additionally teaches
…determine an amount of the… force to be generated by the robotic bee based on the type of the plant… (See “generating a thrust that produces a frequency that induces the plant to release pollen includes operating first motor 404 to produce a first compression wave and operating second motor 406 to produce a second compression wave that is offset in phase from the first compression wave such that the first and second compression waves combine to produce the thrust at a frequency that induces the plant to release pollen… the frequency is about 200 times per second to induce an auto-pollinating plant to release pollen. In some embodiments, the frequency is about 400 times per second to induce a cross-pollinating plant to release pollen” [0051]-[0052]. The frequency of the thrust is determined by the type of plant: auto-pollinating or cross-pollinating. When the frequency is faster, more waves (amount of force) are applied to the plant (a fast-acting force). Additionally, when combining two waves offset in phase, the maximum amplitude (amount of force) may increase; that the combined amplitude is greater than the amplitude of either waveform alone is true for the waveforms shown in Fig. 3. Therefore, Gauvreau teaches that the maximum force (thrust) is determined by the type of the plant. Because of Newton’s Third Law, generating a force towards the plant means that the UAV moves away from the plant, as required by an impulse force. See also [0040] and [0066].).
Sun additionally teaches
training a reinforcement learning model to determine an amount of the impulse force to be generated… based on… the classified level of threat (See “the actions in the reinforcement learning algorithm correspond to the thrusts and torques [impulse forces] of the robot, and the states in the algorithm correspond to the velocities and angular velocities of the robot” [0041]. The reward function is described in [0035], with the first term relating to obstacle (threat) avoidance: “when the AUV approaches the obstacle, it gets a negative reward warning, when the AUV moves away from the obstacle, it gets a positive reward, …when it collides with the obstacle, it gets a reward value of -1 and ends the current training cycle.” As in claim 1, the threat level is classified as either hazardous (obstacle) or non-hazardous (not an obstacle). There is further reward for reaching the target speed: “the reward value r4＝-|△v+△Ψ| is designed using the errors between the speed, angular velocity and the target speed and target angular velocity” [0036]. Using rewards for matching target speeds and avoiding obstacles, the reinforcement learning model learns the amount of impulse force to generate.).
Therefore, the combination of Baldwin/King/Gauvreau/Sun/Robertson as a whole teaches the claim.
Regarding claim 7, Baldwin/King/Gauvreau/Sun/Robertson disclose the limitations of claim 6 as addressed above, and Sun further teaches
wherein a reward is calculated based on a rate of change of a mobility path to a correct mobility path (There is a reward for reaching the target speed and heading: “the reward value r4＝-|△v+△Ψ| is designed using the errors between the speed, angular velocity and the target speed and target angular velocity, where △v is the speed error and △Ψ is the heading error.” [0036]. A smaller speed error and smaller heading error result in a larger reward r4. The current speed and angular velocity represent the current mobility path, and the target speed and target angular velocity represent the correct mobility path. The total reward for the training cycle (over multiple timesteps) is therefore calculated by incorporating r4 per timestep (rate of change).).
Robertson further teaches
wherein a penalty is calculated based on damage to the plant (See “To train a control policy via reinforcement learning, it is necessary to define a utility function that rewards success (in this case picking saleable fruit) and penalizes cost (e.g. time spent, energy consumed, etc.).” [0248]. In this case, the quality control subsystem grades the quality of fruit, and the control subsystem is trained “not to pick a particular target fruit, e.g. because the expected cost of picking in terms of damage to the plant or picked fruit will not outweigh the benefit provided by having one more picked fruit” [0271]. Therefore, the expected cost of damage to the plant or picked fruit is a calculated penalty. See also [0161].).
Therefore, the combination of Baldwin/King/Gauvreau/Sun/Robertson as a whole teaches the claim.
Regarding claims 8 and 15, these claim limitations are significantly similar to those of claim 1; and, thus, are rejected on the same grounds. Note that King discloses that computing devices 600 have a memory 610 and processors 604 in Fig. 6, a UAV 112 has memory 712 and processors 710 in Fig. 7, a visual observer device has memory 810 and processors 806 in Fig. 8, and a central controller 116 has memory 910 and processors 906 in Fig. 9.
Regarding claims 10 and 17, these claim limitations are significantly similar to those of claim 3; and, thus, are rejected on the same grounds.
Regarding claims 12 and 19, these claim limitations are significantly similar to those of claim 5; and, thus, are rejected on the same grounds.
Regarding claims 13 and 20, these claim limitations are significantly similar to those of claim 6; and, thus, are rejected on the same grounds.
Regarding claim 14, these claim limitations are significantly similar to those of claim 7; and, thus, are rejected on the same grounds.
Claims 2-3, 9-10, and 16-17 are rejected under 35 U.S.C. 103 as being unpatentable over Baldwin in view of King, Gauvreau, Sun, and Robertson, and further in view of Miller et al. (US 20200278696 A1; hereafter “Miller”).
Regarding claim 2, Baldwin/King/Gauvreau/Sun/Robertson disclose the limitations of claim 1 as addressed above. However, Baldwin/King/Gauvreau/Sun/Robertson does not explicitly teach “wherein the impulse force is generated via a spring.”
Miller, in the same field of endeavor (UAV systems), teaches
wherein the impulse force is generated via a spring (See “As shown in FIG. 6(c), the spring mechanism is then deployed and activated, in order to launch or jump the device 1 toward the next location. This provides a large impulse [force] for the stalk 4 of the device 1 to lift off, and the device is launched or takes off generally in the direction of the next location” [0218]. See also Fig. 6A-6C, [0181]-[0182], and [0190].).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified the UAV observation and control method of Baldwin/King/Gauvreau/Sun/Robertson with the launch method of Miller. One of ordinary skill in the art would have been motivated to make this modification for the benefit of launching a UAV in a manner “less susceptible to wind, dust or crop (for example) hitting the propellers. This may involve launching the device to a higher altitude than would otherwise be required” (Miller, [0219]).
Regarding claims 9 and 16, these claim limitations are significantly similar to those of claim 2; and, thus, are rejected on the same grounds.
Regarding claim 3, Baldwin/King/Gauvreau/Sun/Robertson disclose the limitations of claim 1 as addressed above. However, Baldwin/King/Gauvreau/Sun/Robertson does not explicitly teach “wherein the impulse force is generated via compressed air.”
Miller, in the same field of endeavor (drone systems), teaches
wherein the impulse force is generated via compressed air (See “In an alternative, the device may be provided with compressed gas or CO2 canisters to assist in the take-off or hop launch. This may be instead of or in addition to the spring loaded jumping mechanism… and with the propellers then used to control the movement of the device 1 once airborne” [0182].).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified the UAV observation and control method of Baldwin/King/Gauvreau/Sun/Robertson with the launch method of Miller. One of ordinary skill in the art would have been motivated to make this modification for the benefit of launching a UAV in a manner “less susceptible to wind, dust or crop (for example) hitting the propellers. This may involve launching the device to a higher altitude than would otherwise be required” (Miller, [0219]).
Regarding claims 10 and 17, these claim limitations are significantly similar to those of claim 3; and, thus, are rejected on the same grounds.
Claims 4, 11, and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Baldwin in view of King, Gauvreau, Sun, and Robertson, and further in view of Rinaldi et al. (US 20170267347 A1; hereafter “Rinaldi”).
Regarding claim 4, Baldwin/King/Gauvreau/Sun/Robertson disclose the limitations of claim 1 as addressed above. However, Baldwin/King/Gauvreau/Sun/Robertson does not explicitly teach “wherein the unbalanced operation of the robotic bee is due to an uneven surface of the plant or a sticky surface of the plant.”
Rinaldi, in the same field of endeavor (UAV systems), teaches
wherein the unbalanced operation of the robotic bee is due to an uneven surface of the plant or a sticky surface of the plant (See “The drone 120 also includes a severing module 145 to sever the suspension means [holding a payload] 135, e.g., to keep the drone 120 from crashing [unbalanced operation] …if the cable is tangled in an obstacle like a tree. …In some embodiments, the package delivery module 130 determines whether to sever the suspension means 135 based on an additional load on the suspension means 135” [0050]. A cable tangled in a tree would occur because of the uneven surface of the tree.).
King discloses a UAV may carry a payload for horticultural operations: “For instance, the UAV may be configured to deliver fertilizer, water, and/or so forth. In this regard, the UAV may be configured to carry cargo or may comprise a compartment or a receptacle” [0079].
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified the UAV observation and control method of Baldwin/King/Gauvreau/Sun/Robertson with the force overload detection means of Rinaldi. One of ordinary skill in the art would have been motivated to make this modification for the benefit of “avoiding the drone 120 from being dragged down” or causing damage because of a plant obstacle (Rinaldi, [0050]).
Regarding claims 11 and 18, these claim limitations are significantly similar to those of claim 4; and, thus, are rejected on the same grounds.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Moya Ly whose telephone number is (571)272-5832. The examiner can normally be reached Monday-Friday 10:00 am-6:00 pm ET.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Ramon Mercado can be reached at (571) 270-5744. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/MOYA LY/Examiner, Art Unit 3658
/Ramon A. Mercado/Supervisory Patent Examiner, Art Unit 3658
Prosecution Insights