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 Arguments
Objection to Specification is withdrawn for all the cases except for one which is not amended according to the objection. Claim objection is withdrawn in view of amendments. Claim rejection under 35 USC § 101 is withdrawn in view of amendments and further review. Applicant’s arguments filed on 03/12/2026 with respect to claim(s) 1-20 have been fully considered but they are not persuasive or moot in view of new ground of rejection provided below which was necessitated based on Applicant’s amendments to the claims. The new ground of rejection for independent claims is based on Vazquez-Santiago in view of Shi, and further in view of Jensen-Nau.
Specification
The disclosure is objected to because of the following informalities:
Page 17 Para [0050] Lines 13: 210 should be switched with 211.
Appropriate correction is required.
Claim Rejections - 35 USC § 103
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.
6. Claim(s) 1, 5-9, 11, and 15-19 are rejected under 35 U.S.C. 103 as being unpatentable over Vazquez-Santiago et al. (K. Vazquez-Santiago, C. F. Goh and K. Shimada, "Motion Planning for Kinematically Redundant Mobile Manipulators with Genetic Algorithm, Pose Interpolation, and Inverse Kinematics," 2021 IEEE 17th International Conference on Automation Science and Engineering (CASE), Lyon, France, 2021, pp. 1167-1174) (Hereinafter Vazquez-Santiago) in view of Shi et al. (Shi X, Guo Y, Chen X, Chen Z, Yang Z. Kinematics and Singularity Analysis of a 7-DOF Redundant Manipulator. Sensors (Basel). 2021 Oct 31) (Hereinafter Shi), and further in view of Jensen-Nau et al. (K. R. Jensen-Nau, T. Hermans and K. K. Leang, "Near-Optimal Area-Coverage Path Planning of Energy-Constrained Aerial Robots With Application in Autonomous Environmental Monitoring," in IEEE Transactions on Automation Science and Engineering, vol. 18, no. 3, pp. 1453-1468, July 2021).
Regarding Claim 1, Vazquez-Santiago teaches a single kinematic chain robotic control system, comprising:
A robotic manipulator (See at least Page 1167 Col 2 Para 2 “The new design, as shown in Fig.
1(c), is an 8 DoF mobile manipulator.”);
a mobile base (See at least Page 1167 Col 2 Para 2 “The mobility of the base supports
positioning to any location on the floor, as the base is non-holonomic with 2 DoF.”);
unified control circuitry, the unified control circuitry comprising:
feedback adjustment circuitry (See at least Page 1167 Col 2 Para 3 “A path following methodology in [11] applied a similar method to [5] with the addition of an error feedback control algorithm to ensure both base and manipulator follow their designated paths while ignoring the orientation of the end-effector…”);
manipulator interface circuitry communicatively coupled with the robotic manipulator (See at least Page 1167 Col 2 Para 3 “The method in [9] also maximizes the manipulability of the end-effector position to plan a trajectory with non-holonomic constraints forming part of the control algorithm, while [10] generates a trajectory for the mobile base by selecting positions that will enable better localization to reduce uncertainty…”);
base interface circuitry communicatively coupled with the mobile base (See at least Fig 7(a) shows mobile base poses which is construed as base interface circuitry communicatively capable of coupled with the mobile base, Page 1167 Col 2 Para 3 “The method in [9] also maximizes the manipulability of the end-effector position to plan a trajectory with non-holonomic constraints forming part of the control algorithm, while [10] generates a trajectory for the mobile base by selecting positions that will enable better localization to reduce uncertainty…”); and
global planning control circuitry configured to receive initial position data of a robotic manipulator having nJ joints and initial position data of the mobile base having nM joints; a joint having at least one degree of freedom; the global planning control circuitry further configured to receive a motion plan request that includes at least one waypoint for at least one articulating component coupled to one of the nJ joints of the robotic manipulator; a waypoint being a destination of the articulating component of the robotic manipulator in an external environment (See at least Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1. We adopt the Genetic Algorithm (GA) to solve for optimal time series of configurations for each sampled target. The GA can perform a stochastic search over a large search space with different units of measurement, as is the defined problem. Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome. Multiple chromosomes, describing feasible solutions are given to the GA as an initial population for optimization.”, Page 1168 “A. Path discretization and sampling - The input path is discretized into an array of targets, assuming the density of discretization is representative of a continuous path. To improve the computational efficiency, a number of waypoints are sampled from the trajectory target points, considering the changes in the path direction. The starting and end trajectory points are selected to ensure path coverage…”, Page 1170 Col 1 “D. Inverse Kinematics Solver - The software provides the capabilities for the user to define a starting pose from which to begin its search and find the closest pose where joint angle changes are minimized”); the global planning control circuitry comprising:
discretized determination circuitry to generate a first combined matrix of joint values for each of the nM mobile base joints and the nJ robotic manipulator joints, based on the motion plan request and the initial position data (See at least Page 1168 Col 2 “A. Path discretization and sampling - The input path is discretized into an array of targets, assuming the density of discretization is representative of a continuous path. To improve the computational efficiency, a number of waypoints are sampled from the trajectory target points, considering the changes in the path direction. The starting and end trajectory points are selected to ensure path coverage.”, Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1. … The variables, x θ1 t to x θ6 t , represent the manipulator joint values, while x BX t , x BY t , x Bω t represent the base (X,Y) position and orientation, respectively … We adopt the Genetic Algorithm (GA) to solve for optimal time series of configurations for each sampled target. The GA can perform a stochastic search over a large search space with different units of measurement, as is the defined problem. Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome. Multiple chromosomes, describing feasible solutions are given to the GA as an initial population for optimization.”, Page 1170 Col 1 “D. Inverse Kinematics Solver - The software provides the capabilities for the user to define a starting pose from which to begin its search and find the closest pose where joint angle changes are minimized”), …
optimization circuitry configured to generate a second combined matrix of joint values for each of the nM mobile base joints and the nJ robotic manipulator joints, based on the first combined matrix and one or more optimization parameters associated with the mobile base or the robotic manipulator (See at least Page 1168 Col 1 Para 5 “In this paper, we propose a novel approach of combining sampling of the end-effector path, GA optimization, linear interpolation, and inverse kinematics for the simultaneous base and manipulator motion planning optimization, with constraints of not only end-effector position, but end-effector orientation as well, Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1… We adopt the Genetic Algorithm (GA) to solve for optimal time series of configurations for each sampled target. The GA can perform a stochastic search over a large search space with different units of measurement, as is the defined problem. Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome. Multiple chromosomes, describing feasible solutions are given to the GA as an initial population for optimization.”, Page 1168 Col 2 “II. PROPOSED COMPUTATIONAL METHOD -…Optimize mobile manipulator configurations with the GA for the sampled waypoints.”);
matrix parsing circuitry configured to receive the second combined matrix and parse the second combined matrix to generate a third matrix of joint values for the robotic manipulator and a fourth matrix of joint values for the mobile robotic base (See at least Page 1169 Col 1 “Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome.”, Page 1170 Col 1 “D. Inverse Kinematics Solver - … The closed-form expression of the analytical solution is a more computationally efficient approach than a numerical solution, such as an iterative optimization.”, discloses iterative optimization which is construed as matrix parsing circuitry configured to receive the second combined matrix and parse the second combined matrix to generate a third matrix of joint values for the robotic manipulator and a fourth matrix of joint values for the mobile robotic base)…
controlling the robotic manipulator and the mobile base to move along the selected lowest cost
path to the first waypoint (See at least Page 1167 Col 2 Para 3 “Point-to-point task planning establishes the motions to be performed to move a robotic system from one location to another…”, Page 1169 Col 1 Para “The optimization goals are defined to minimize the base rotations, travelling distance, and manipulator joint velocities, as reflected in the cost function in Equation 2”) …
However, Vazquez-Santiago does not explicitly spell out … and to eliminate joint space solutions that would cause a singularity, wherein the singularity is a waypoint where inverse kinematics is mathematically unstable; and …
… and the unified control circuitry configured to:
determine a lowest cost path to a first waypoint, wherein the lowest cost path to the first waypoint is determined by:
defining one or more cost functions based on joint state values for each of the nj joints
and nM joints, wherein each cost function is a scalar value representing a favorability of a path based on minimizing a required travel distance along the path between successive waypoints;
assigning weights to each path based on the one or more cost functions;
optimizing one or more discrete step values in a discrete step determination matrix by
performing gradient descent cost optimization based on the one or more cost functions; and
selecting the lowest cost path;
controlling the robotic manipulator and the mobile base to move along the selected lowest cost
path to the first waypoint; and responsive to reaching the first waypoint, wherein the first waypoint is not a final waypoint:
determine the lowest cost path to a next waypoint.
Shi teaches … and to eliminate joint space solutions that would cause a singularity,
wherein the singularity is a waypoint where inverse kinematics is mathematically unstable (See at least Page 11 “4. Singularity Avoidance - … In the following section, the above-mentioned singular configuration is described based on position, and a singularity avoidance method is proposed for the avoidable singularities.”, Page 18 Para 3 “According to the above kinematic analysis, the redundant angle ϕ needs to be determined to solve the inverse kinematics…”, Page 19 “6. Conclusions … A singular configuration avoidance algorithm based on self-motion is proposed through the selection of redundant angles to avoid singular patterns. The inverse kinematics and singularity analysis of the robotic arm are verified by simulation examples, laying the foundation for the motion planning of a 7-DOF manipulator.”)…
Therefore, it would have been obvious to one of the ordinary skill in the art before the
effective filing date of the claimed invention to combine the system of Vazquez-Santiago with the teachings of Shi and include the feature of eliminating joint space solutions that would cause a singularity, thereby enhance control with increased efficiency with simplified programming (See at least “2.2. Self-Motion Analysis - According to the mechanical characteristics of the manipulator, the geometric simplification of the manipulator is carried out…”).
Jensen-Nau teaches … and the unified control circuitry configured to:
determine a lowest cost path to a first waypoint (See at least Page 1461 Col 2 Para 1 “The cost of the true optimal path for each number of waypoints was assumed to be the lowest cost found by the direct optimization…”), wherein the lowest cost path to the first waypoint is determined by:
defining one or more cost functions based on joint state values for each of the nj joints
and nM joints, wherein each cost function is a scalar value representing a favorability of a path based on minimizing a required travel distance along the path between successive waypoints (See at least Page 1461 Col 2 Para 1 “The cost of the true optimal path for each number of waypoints was assumed to be the lowest cost found by the direct optimization…”);
assigning weights to each path based on the one or more cost functions (See at least Page 1465 “The path length can only be controlled indirectly by changing the weights in the cost function, with the required weights varying based on multiple considerations such as path length, area size, and area shape.”);
optimizing one or more discrete step values in a discrete step determination matrix by
performing gradient descent cost optimization based on the one or more cost functions (See at least 1454 Col 2 Para 2 “However, their method uses a cost function based on those distances and creates a gradient descent-based velocity controller to move the path points to their optimal locations…”); and
selecting the lowest cost path (See at least Page 1461 Col 2 Para 1 “The cost of the true optimal path for each number of waypoints was assumed to be the lowest cost found by the direct optimization…”);
controlling the robotic manipulator and the mobile base to move along the selected lowest cost
path to the first waypoint (See at least Page 1456 Col 2 “III. PATH GENERATION ALGORITHM - … optimal control may be used to ensure that the path is one that the robot can follow…”, Page 1461 Col 2 Para 1 “The cost of the true optimal path for each number of waypoints was assumed to be the lowest cost found by the direct optimization…”); and responsive to reaching the first waypoint, wherein the first waypoint is not a final waypoint:
determine the lowest cost path to a next waypoint (See at least Page 1459 Col 2 Para 3 “The differential equation (19) can then be solved numerically using Euler’s method to find the next velocity ˙pj and position pj for each waypoint.”).
Therefore, it would have been obvious to one of the ordinary skill in the art before the
effective filing date of the claimed invention to combine the system of Vazquez-Santiago with the teachings of Jensen-Nau and include the feature of performing path planning optimization through gradient descent cost optimization, thereby provide scalability (See at least Page 1454 Col 2 Para 3 “Overall, simulation results show that the algorithm has distinct advantages over other path generation methods in the partial coverage scenario, with good runtimes, scalability, and path costs.”).
Regarding Claim 5, modified Vazquez-Santiago teaches all the elements of claim 1. Vazquez-Santiago further teaches the single kinematic chain robotic control system of claim 1, wherein the discretized determination circuitry generates the first combined matrix using inverse kinematics to determine 8 joint space solutions for the robotic manipulator for each mobile base solution for each waypoint in the motion plan request, wherein a mobile base solution is a possible location and orientation of the mobile base in the external environment within a sampling area, a joint space solution being a set of the nJ joint values capable of causing the articulating component of the robotic manipulator to arrive at a waypoint (See at least Page 1167 Col 1 “Abstract- …This paper presents a novel computational method for simultaneous optimization of base and manipulator robotic system with 8 DoF for welding tasks, constraining both end-effector position and orientation. The mobile manipulator consists of a 2 DoF non-holonomic base and a 6 DoF manipulator. The proposed method applies a Genetic Algorithm (GA) to solve for optimized configurations for the base and manipulator for strategically sampled end-effector waypoints. The base configurations and end-effector orientations are interpolated between the GA solutions and used as inputs for an inverse kinematics solver to find the optimal manipulator pose. The experiment results show that the proposed methods create optimized smooth and continuous motions for both the base and manipulator while constraining the end-effector position and orientation. The proposed method is a novel application of GA optimization, with improved results for path following motion planning by including sampling, interpolation, and inverse kinematics steps within the methodology.”, Page 1168 Col 2 “A. Path discretization and sampling - The input path is discretized into an array of targets, assuming the density of discretization is representative of a continuous path. To improve the computational efficiency, a number of waypoints are sampled from the trajectory target points, considering the changes in the path direction. The starting and end trajectory points are selected to ensure path coverage.”, Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1. … The variables, x θ1 t to x θ6 t , represent the manipulator joint values, while x BX t , x BY t , x Bω t represent the base (X,Y) position and orientation, respectively … We adopt the Genetic Algorithm (GA) to solve for optimal time series of configurations for each sampled target. The GA can perform a stochastic search over a large search space with different units of measurement, as is the defined problem. Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome. Multiple chromosomes, describing feasible solutions are given to the GA as an initial population for optimization.”, Page 1170 Col 1 “D. Inverse Kinematics Solver - The software provides the capabilities for the user to define a starting pose from which to begin its search and find the closest pose where joint angle changes are minimized”).
Regarding Claim 6, modified Vazquez-Santiago teaches all the elements of claim 5. Vazquez-Santiago further teaches the single kinematic chain robotic control system of claim 5, wherein the discretized determination circuitry generates the first combined matrix by further eliminating invalid joint state solutions (See at least Page 1169 Col 2 Para 2 “Chromosomes in the initial population are evaluated in terms of the fitness function. To create new generations, parents are to be selected. Parents of the new generation are selected by the tournament approach, where four candidate chromosomes selected at random are compared with regards to their fitness function value. The chromosome with the highest fitness is selected as a parent.”).
Regarding Claim 7, modified Vazquez-Santiago teaches all the elements of claim 6. Vazquez-Santiago further teaches the single kinematic chain robotic control system of claim 6, wherein invalid joint state solutions include joint state solutions that would cause a portion of the robotic manipulator to collide with the mobile base, another portion of the robotic manipulator, and/or an object in an environment surrounding the robotic manipulator (See at least Page 1168 Col 2 Para 2 “II. PROPOSED COMPUTATIONAL METHOD - The configurations of the mobile base throughout the path are constrained to avoid collision with the part. Another system constraint is to avoid collision between the manipulator and the supporting diagonal frames of the base it is mounted on.”, Page 1170 Col 2 Para 4 “B. Results - … Besides generating a smooth mobile base and manipulator trajectory, the generated motions also fulfill the collision avoidance constraints …”).
Regarding Claim 8, modified Vazquez-Santiago teaches all the elements of claim 5. Vazquez-Santiago further teaches the single kinematic chain robotic control system of claim 5, wherein the discretized determination circuitry generates the first combined matrix by further assigning weights to paths between joint state solutions for successive waypoints of the motion plan request and selecting the paths with the highest weights, where the motion plan request contains more than one waypoint, wherein a path is the required motion of the robotic manipulator and mobile base to move from waypoint to another, the weights being assigned based on cost factors (See at least Page 1169 Col 1 Para 5 “B. Genetic Algorithm formulation - … The optimization goals are defined to minimize the base rotations, travelling distance, and manipulator joint velocities, as reflected in the cost function in Equation 2.”, Page 1169 Col 2 Para 1 “… The objective function weighting favored the minimization of the joint velocities, with w3 having a value of 10, while w1 and w2 were equal to 1. Based on the defined optimization goals, a fitness function for the GA was determined to be the inverse of Equation 2.”, Page 1169 Col 2 Para 2 “Chromosomes in the initial population are evaluated in terms of the fitness function. To create new generations, parents are to be selected. Parents of the new generation are selected by the tournament approach, where four candidate chromosomes selected at random are compared with regards to their fitness function value. The chromosome with the highest fitness is selected as a parent.”).
Regarding Claim 9, modified Vazquez-Santiago teaches all the elements of claim 8. Vazquez-Santiago further teaches the single kinematic chain robotic control system of claim 8, wherein the cost factors include:
minimizing the required travel distance along the path between the joint state solutions for successive waypoints (See at least Page 1169 Col 1 Para 5 “B. Genetic Algorithm formulation - … The optimization goals are defined to minimize the base rotations, travelling distance, and manipulator joint velocities, as reflected in the cost function in Equation 2.”);
reducing the instances where the robotic manipulator must change configurations traveling between the joint states for successive waypoints; and/or
reducing the velocity and/or acceleration of the joints of the robotic manipulator and/or the joints of the mobile base when traveling between the joint states for successive waypoints (See at least Page 1169 Col 1 Para 5 “B. Genetic Algorithm formulation - … The optimization goals are defined to minimize the base rotations, travelling distance, and manipulator joint velocities, as reflected in the cost function in Equation 2.”).
Regarding Claim 11, Vazquez-Santiago teaches a non-transitory storage device that includes machine-readable instructions that, when executed by one or more processors, cause the one or more processors to perform operations, comprising:
generating a first combined matrix of joint values each of joint of a robotic manipulator having nJ joints and each joint of a mobile base having nM joints; a joint having at least one degree of freedom; the first combined matrix is based on a motion plan request, initial position data of the robotic manipulator and initial position data of the mobile robotic base; the motion plan request including at least one waypoint for at least one articulating component of the robotic manipulator; a waypoint being a destination of the articulating component of the robotic manipulator in an external environment (See at least Page 1168 Col 2 “A. Path discretization and sampling - The input path is discretized into an array of targets, assuming the density of discretization is representative of a continuous path. To improve the computational efficiency, a number of waypoints are sampled from the trajectory target points, considering the changes in the path direction. The starting and end trajectory points are selected to ensure path coverage.”, Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1. … The variables, x θ1 t to x θ6 t , represent the manipulator joint values, while x BX t , x BY t , x Bω t represent the base (X,Y) position and orientation, respectively … We adopt the Genetic Algorithm (GA) to solve for optimal time series of configurations for each sampled target. The GA can perform a stochastic search over a large search space with different units of measurement, as is the defined problem. Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome. Multiple chromosomes, describing feasible solutions are given to the GA as an initial population for optimization.”, Page 1170 Col 1 “D. Inverse Kinematics Solver - The software provides the capabilities for the user to define a starting pose from which to begin its search and find the closest pose where joint angle changes are minimized”, Page 1168 Col 1 Para 3 “…Genetic Algorithms have been applied successfully in the past to optimize manipulator motion planning as shown in [22]…”); …
generating a second combined matrix of joint values for each of the nM mobile base joints and the nJ robotic manipulator joints, based on the first combined matrix and one or more optimization parameters associated with the mobile base or the robotic manipulator (See at least Page 1168 Col 1 Para 5 “In this paper, we propose a novel approach of combining sampling of the end-effector path, GA optimization, linear interpolation, and inverse kinematics for the simultaneous base and manipulator motion planning optimization, with constraints of not only end-effector position, but end-effector orientation as well, Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1… We adopt the Genetic Algorithm (GA) to solve for optimal time series of configurations for each sampled target. The GA can perform a stochastic search over a large search space with different units of measurement, as is the defined problem. Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome. Multiple chromosomes, describing feasible solutions are given to the GA as an initial population for optimization.”, Page 1168 Col 2 “II. PROPOSED COMPUTATIONAL METHOD -…Optimize mobile manipulator configurations with the GA for the sampled waypoints.”, Page 1172 Col 2 “… A more stringent or flexible value could be used as a constraint parameter, depending on the application…”); and
parsing the second combined matrix to generate a third matrix of joint values for the robotic manipulator and a fourth matrix of joint values for the mobile base (See at least Page 1169 Col 1 “Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome.”, Page 1170 Col 1 “D. Inverse Kinematics Solver - … The closed-form expression of the analytical solution is a more computationally efficient approach than a numerical solution, such as an iterative optimization.”, discloses iterative optimization which is construed as parsing the second combined matrix to generate a third matrix of joint values for the robotic manipulator and a fourth matrix of joint values for the mobile robotic base).
However, Vazquez-Santiago does not explicitly spell out … eliminating joint space solutions that would cause a singularity, wherein the singularity is a waypoint where inverse kinematics is mathematically unstable; …
… and the unified control circuitry configured to:
defining one or more cost functions based on joint state values for each of the nj joints
and nM joints, wherein each cost function is a scalar value representing a favorability of a path based on minimizing a required travel distance along the path between successive waypoints;
assigning weights to each path based on the one or more cost functions;
optimizing one or more discrete step values in a discrete step determination matrix by
performing gradient descent cost optimization based on the one or more cost functions; and
selecting the lowest cost path;
controlling the robotic manipulator and the mobile base to move along the selected lowest cost
path to the first waypoint; and
responsive to reaching the first waypoint, wherein the first waypoint is not a final waypoint:
determine the lowest cost path to a next waypoint.
Shi teaches … eliminating joint space solutions that would cause a singularity,
wherein the singularity is a waypoint where inverse kinematics is mathematically unstable (See at least Page 11 “4. Singularity Avoidance - … In the following section, the above-mentioned singular configuration is described based on position, and a singularity avoidance method is proposed for the avoidable singularities.”, Page 18 Para 3 “According to the above kinematic analysis, the redundant angle ϕ needs to be determined to solve the inverse kinematics…”, Page 19 “6. Conclusions … A singular configuration avoidance algorithm based on self-motion is proposed through the selection of redundant angles to avoid singular patterns. The inverse kinematics and singularity analysis of the robotic arm are verified by simulation examples, laying the foundation for the motion planning of a 7-DOF manipulator.”)…
Therefore, it would have been obvious to one of the ordinary skill in the art before the
effective filing date of the claimed invention to combine the system of Vazquez-Santiago with the teachings of Shi and include the feature of eliminating joint space solutions that would cause a singularity, thereby enhance control with increased efficiency with simplified programming (See at least “2.2. Self-Motion Analysis - According to the mechanical characteristics of the manipulator, the geometric simplification of the manipulator is carried out…”).
Jensen-Nau teaches …
defining one or more cost functions based on joint state values for each of the nj joints
and nM joints, wherein each cost function is a scalar value representing a favorability of a path based on minimizing a required travel distance along the path between successive waypoints (See at least Page 1461 Col 2 Para 1 “The cost of the true optimal path for each number of waypoints was assumed to be the lowest cost found by the direct optimization…”);
assigning weights to each path based on the one or more cost functions (See at least Page 1465 “The path length can only be controlled indirectly by changing the weights in the cost function, with the required weights varying based on multiple considerations such as path length, area size, and area shape.”);
optimizing one or more discrete step values in a discrete step determination matrix by
performing gradient descent cost optimization based on the one or more cost functions (See at least 1454 Col 2 Para 2 “However, their method uses a cost function based on those distances and creates a gradient descent-based velocity controller to move the path points to their optimal locations…”); and
selecting the lowest cost path (See at least Page 1461 Col 2 Para 1 “The cost of the true
optimal path for each number of waypoints was assumed to be the lowest cost found by the direct optimization…”);
controlling the robotic manipulator and the mobile base to move along the selected lowest cost path to the first waypoint (See at least Page 1456 Col 2 “III. PATH GENERATION
ALGORITHM - … optimal control may be used to ensure that the path is one that the robot can follow…”, Page 1461 Col 2 Para 1 “The cost of the true optimal path for each number of waypoints was assumed to be the lowest cost found by the direct optimization…”); and responsive to reaching the first waypoint, wherein the first waypoint is not a final waypoint:
determine the lowest cost path to a next waypoint (See at least Page 1459 Col 2 Para 3
“The differential equation (19) can then be solved numerically using Euler’s method to find the next velocity ˙pj and position pj for each waypoint.”).
Therefore, it would have been obvious to one of the ordinary skill in the art before the
effective filing date of the claimed invention to combine the system of Vazquez-Santiago with the teachings of Jensen-Nau and include the feature of performing path planning optimization through gradient descent cost optimization, thereby provide scalability (See at least Page 1454 Col 2 Para 3 “Overall, simulation results show that the algorithm has distinct advantages over other path generation methods in the partial coverage scenario, with good runtimes, scalability, and path costs.”).
Regarding Claim 15, modified Vazquez-Santiago teaches all the elements of claim 11. Vazquez-Santiago further teaches the generation of the first combined matrix of claim 11, further comprising, using inverse kinematics to determine 8 joint space solutions for the robotic manipulator for each mobile base solution for each waypoint in the motion plan request, wherein a mobile base solution is a possible location and orientation of the mobile base in the external environment within a sampling area, a joint space solution being a set of nJ joints values capable of causing the articulating component of the robotic manipulator to arrive at a waypoint (See at least Page 1167 Col 1 “Abstract- …This paper presents a novel computational method for simultaneous optimization of base and manipulator robotic system with 8 DoF for welding tasks, constraining both end-effector position and orientation. The mobile manipulator consists of a 2 DoF non-holonomic base and a 6 DoF manipulator. The proposed method applies a Genetic Algorithm (GA) to solve for optimized configurations for the base and manipulator for strategically sampled end-effector waypoints. The base configurations and end-effector orientations are interpolated between the GA solutions and used as inputs for an inverse kinematics solver to find the optimal manipulator pose. The experiment results show that the proposed methods create optimized smooth and continuous motions for both the base and manipulator while constraining the end-effector position and orientation. The proposed method is a novel application of GA optimization, with improved results for path following motion planning by including sampling, interpolation, and inverse kinematics steps within the methodology.”, Page 1168 Col 2 “A. Path discretization and sampling - The input path is discretized into an array of targets, assuming the density of discretization is representative of a continuous path. To improve the computational efficiency, a number of waypoints are sampled from the trajectory target points, considering the changes in the path direction. The starting and end trajectory points are selected to ensure path coverage.”, Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1. … The variables, x θ1 t to x θ6 t , represent the manipulator joint values, while x BX t , x BY t , x Bω t represent the base (X,Y) position and orientation, respectively … We adopt the Genetic Algorithm (GA) to solve for optimal time series of configurations for each sampled target. The GA can perform a stochastic search over a large search space with different units of measurement, as is the defined problem. Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome. Multiple chromosomes, describing feasible solutions are given to the GA as an initial population for optimization.”, Page 1170 Col 1 “D. Inverse Kinematics Solver - The software provides the capabilities for the user to define a starting pose from which to begin its search and find the closest pose where joint angle changes are minimized”).
Regarding Claim 16, modified Vazquez-Santiago teaches all the elements of claim 15. Vazquez-Santiago further teaches the non-transitory storage device that includes machine-readable instructions that, when executed by one or more processors, cause the one or more processors to perform operations, of claim 15, further comprising, eliminating invalid joint state solutions after generating the 8 joint space solutions for the robotic manipulator for each mobile base solution for each waypoint in the motion plan request (See at least Page 1169 Col 2 Para 2 “Chromosomes in the initial population are evaluated in terms of the fitness function. To create new generations, parents are to be selected. Parents of the new generation are selected by the tournament approach, where four candidate chromosomes selected at random are compared with regards to their fitness function value. The chromosome with the highest fitness is selected as a parent.”).
Regarding Claim 17, modified Vazquez-Santiago teaches all the elements of claim 16. Vazquez-Santiago further teaches the non-transitory storage device that includes machine-readable instructions that, when executed by one or more processors, cause the one or more processors to perform operations of claim 16, wherein invalid joint state solutions include joint state solutions that would cause a portion of the robotic manipulator to collide with the mobile base, another portion of the robotic manipulator, and/or an object in an environment surrounding the robotic manipulator (See at least Page 1168 Col 2 Para 2 “II. PROPOSED COMPUTATIONAL METHOD - The configurations of the mobile base throughout the path are constrained to avoid collision with the part. Another system constraint is to avoid collision between the manipulator and the supporting diagonal frames of the base it is mounted on.”, Page 1170 Col 2 Para 4 “B. Results - … Besides generating a smooth mobile base and manipulator trajectory, the generated motions also fulfill the collision avoidance constraints …”).
Regarding Claim 18, modified Vazquez-Santiago teaches all the elements of claim 15. Vazquez-Santiago further teaches the non-transitory storage device that includes machine-readable instructions that, when executed by one or more processors, cause the one or more processors to perform operations, of claim 15, further comprising, generating the first combined matrix by further assigning weights to paths between joint state solutions for successive waypoints of the motion plan request and selecting the paths with the highest weights, where the motion plan request contains more than one waypoint, wherein a path is the required motion of the robotic manipulator and mobile base to move from one waypoint to another, the weights being assigned based on cost factors (See at least Page 1169 Col 1 Para 5 “B. Genetic Algorithm formulation - … The optimization goals are defined to minimize the base rotations, travelling distance, and manipulator joint velocities, as reflected in the cost function in Equation 2.”, Page 1169 Col 2 Para 1 “… The objective function weighting favored the minimization of the joint velocities, with w3 having a value of 10, while w1 and w2 were equal to 1. Based on the defined optimization goals, a fitness function for the GA was determined to be the inverse of Equation 2.”, Page 1169 Col 2 Para 2 “Chromosomes in the initial population are evaluated in terms of the fitness function. To create new generations, parents are to be selected. Parents of the new generation are selected by the tournament approach, where four candidate chromosomes selected at random are compared with regards to their fitness function value. The chromosome with the highest fitness is selected as a parent.”).
Regarding Claim 19, modified Vazquez-Santiago teaches all the elements of claim 18. Vazquez-Santiago further teaches the non-transitory storage device that includes machine-readable instructions that, when executed by one or more processors, cause the one or more processors to perform operations, of claim 18, wherein the cost factors include:
minimizing the required travel distance along the path between the joint states solutions for successive waypoints (See at least Page 1169 Col 1 Para 5 “B. Genetic Algorithm formulation - … The optimization goals are defined to minimize the base rotations, travelling distance, and manipulator joint velocities, as reflected in the cost function in Equation 2.”);
reducing the instances where the robotic manipulator must change configurations traveling between the joint states for successive waypoints; and/or
reducing the velocity and/or acceleration of the joints of the robotic manipulator and/or the joints of the mobile base when traveling between the joint states for successive waypoints (See at least Page 1169 Col 1 Para 5 “B. Genetic Algorithm formulation - … The optimization goals are defined to minimize the base rotations, travelling distance, and manipulator joint velocities, as reflected in the cost function in Equation 2.”).
Claim(s) 2, 3, 4, 12, 13, and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Vazquez-Santiago et al. (K. Vazquez-Santiago, C. F. Goh and K. Shimada, "Motion Planning for Kinematically Redundant Mobile Manipulators with Genetic Algorithm, Pose Interpolation, and Inverse Kinematics," 2021 IEEE 17th International Conference on Automation Science and Engineering (CASE), Lyon, France, 2021, pp. 1167-1174) (Hereinafter Vazquez-Santiago) in view of Shi et al. (Shi X, Guo Y, Chen X, Chen Z, Yang Z. Kinematics and Singularity Analysis of a 7-DOF Redundant Manipulator. Sensors (Basel). 2021 Oct 31) (Hereinafter Shi), Jensen-Nau et al. (K. R. Jensen-Nau, T. Hermans and K. K. Leang, "Near-Optimal Area-Coverage Path Planning of Energy-Constrained Aerial Robots With Application in Autonomous Environmental Monitoring," in IEEE Transactions on Automation Science and Engineering, vol. 18, no. 3, pp. 1453-1468, July 2021), and further in view of Bosscher et al. (US 20120245736 A1) (Hereinafter Bosscher).
Regarding Claim 2, modified Vazquez-Santiago teaches all the elements of claim 1. Vazquez-Santiago further teaches the single kinematic chain robotic control system of claim 1, further comprising:
…
wherein, the optimized circuitry is further configured to generate a second modified matrix of joint values for each of the nJ joints of the robotic manipulator and the nM joints of the mobile base, the second modified matrix being generated based on the first modified matrix and one or more optimization parameters associated with the mobile base or the robotic manipulator (See at least Para (See at least Page 1168 Col 1 Para 5 “In this paper, we propose a novel approach of combining sampling of the end-effector path, GA optimization, linear interpolation, and inverse kinematics for the simultaneous base and manipulator motion planning optimization, with constraints of not only end-effector position, but end-effector orientation as well, Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1… We adopt the Genetic Algorithm (GA) to solve for optimal time series of configurations for each sampled target. The GA can perform a stochastic search over a large search space with different units of measurement, as is the defined problem. Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome. Multiple chromosomes, describing feasible solutions are given to the GA as an initial population for optimization.”, Page 1168 Col 2 “II. PROPOSED COMPUTATIONAL METHOD -…Optimize mobile manipulator configurations with the GA for the sampled waypoints.”, Page 1172 Col 2 “… A more stringent or flexible value could be used as a constraint parameter, depending on the application…”);
wherein, the matrix parsing circuitry is further configured to receive the second modified matrix and parse the second modified matrix into a third modified matrix of joint values for the robotic manipulator and a fourth modified matrix of joint values for the mobile base Page 1169 Col 1 “Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome.”, Page 1170 Col 1 “D. Inverse Kinematics Solver - … The closed-form expression of the analytical solution is a more computationally efficient approach than a numerical solution, such as an iterative optimization.”, discloses iterative optimization which is construed as parsing the second combined matrix to generate a third matrix of joint values for the robotic manipulator and a fourth matrix of joint values for the mobile robotic base.
Although Vazquez-Santiago discloses feedback control algorithm (See at least Page 1167 Col 2 Para 3) and laser profile sensor (See at least Fig. 1 (c)), however, he does not explicitly spell out …
feedback adjustment circuitry configured to generate a first modified matrix of joint values for each of the nJ joints of the robotic manipulator and nM joints of the mobile base, based on the second combined matrix and a feedback matrix, the feedback matrix comprising actual joint values for the robotic manipulator joints and the mobile base joints generated by one or more sensors …
Bosscher teaches …
feedback adjustment circuitry configured to generate a first modified matrix of joint values for each of the nJ joints of the robotic manipulator and nM joints of the mobile base, based on the second combined matrix and a feedback matrix, the feedback matrix comprising actual joint values for the robotic manipulator joints and the mobile base joints generated by one or more sensors (See at least Para [0019] “… The vector q may be constructed, for example, from feedback obtained from the joints 18 themselves; that is, each joint 18 can include a sensor that indicates the current travel position of that joint 18 …”, Claim 6. “6. The method of claim 3 further comprising generating the m.times.(n-L) matrix J.sub.mod utilizing the following steps: determining an m.times.n Jacobian matrix J for the robotic device utilizing the current state of the robotic device, the current state of the robotic device comprising current displacements of each of the n joints; and utilizing the Jacobian matrix J to generate the matrix J.sub.mod, wherein J.sub.mod comprises all columns in J except for those columns corresponding to the L actively limited joints.”, Para [0033] “… Then, if a joint 112 that has an active limits flag 241 set (i.e. ActiveLimits.sub.i=1), the matrix column for that joint 112 is removed, in effect turning the Jacobian matrix J 211 from an m.times.n matrix into a modified m.times.(n-1) matrix so that the limited joint 112 is not represented in the modified Jacobian matrix J.sub.mod 251. Of course, more than one joint 112 may be limited, as indicated by the active limits flags 141. Letting L denote the number of joints whose limits are active, as indicated by the active limits flags 241, a total of L columns of the Jacobian matrix J 211 can be removed, resulting in a modified Jacobian matrix J.sub.mod 251 that is an m.times.(n-L) matrix.”), …
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to combine the system of Vazquez-Santiago with the teachings of Bosscher and include the feature of feedback adjustment circuitry being configured to generate a first modified matrix of joint values for each of the nJ joints of the robotic manipulator and nM joints of the mobile base, based on the second combined matrix and a feedback matrix, the feedback matrix comprising actual joint values for the robotic manipulator joints and the mobile base joints generated by one or more sensors, thereby provide robot joint calculation accuracy (See at least Para [0032] “As a fifth step 250 , improved joint limiting logic 250 accepts the joint active limits flags 241 , the operator-desired movement vector {dot over (x)}cmd 203 , the Jacobian matrix 211 and a scalar weighting factors 204 and generates a revised joint velocity command {dot over (q)}new 261…”).
Regarding Claim 3, modified Vazquez-Santiago teaches all the elements of claim 2. Vazquez-Santiago further teaches the feedback adjustment circuitry of claim 2, further configured to generate a first modified matrix by:
determining the joint values within the first matrix which most closely corresponds to the joint values of the actual location data for the robotic manipulator and mobile base (See at least Page 1173 Col 2 Para 3 “Finally, the GA optimization was able to generate the initial configurations used to guide the creation of a continuous motion plan for the kinematically redundant mobile manipulator.”, Page 1172 Col 2 Para 1 “Shown in Table IV, are the end-effector error in position. The error was calculated as the average error percentage between the desired and achieved end-effector position, with each (X, Y, Z) coordinate represented as the magnitude of a vector”, Page 1170 Col 1 Para 2 “… Therefore, the discretized end-effector target positions are provided as input to the IK solver, along with their respective interpolated quaternion value from the GA solution …”);
inserting the joint values of the actual location data into the beginning of the first modified matrix (See at least Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1. … The variables, x θ1 t to x θ6 t , represent the manipulator joint values, while x BX t , x BY t , x Bω t represent the base (X,Y) position and orientation, respectively. Additionally, the subscripts ti , where i = 0, 1, 2, .., n, denotes the corresponding time step of the configuration.”); and
inserting joint values for a set of modified waypoints, the modified waypoints being generated so the connections between successive modified waypoints followings a path in the external environment created by the connections between successive waypoints of the motion plan request, wherein the density of waypoints is highest between the first waypoint and second waypoint and lowest between the second to last and last waypoint, the density of waypoints referring to the time between the execution of successive waypoints (See at least Page 1169 Col 1 Para 3 “B. Genetic Algorithm formulation - … The variables, x θ1 t to x θ6 t , represent the manipulator joint values, while x BX t , x BY t , x Bω t represent the base (X,Y) position and orientation, respectively. Additionally, the subscripts ti , where i = 0, 1, 2, .., n, denotes the corresponding time step of the configuration. Therefore, each column describes the configuration at every time step. Consequently, n is dependent on the desired discretization density of the path to be followed by the end-effector. Due to the nature of welding processes, the end-effector is required to move at a constant speed. To achieve a constant speed, each column configuration in Equation 1 is to be performed at the n th time step. Each time step corresponds to a path target, and its value will depend on the desired welding speed and the distance between path target positions.”).
Regarding Claim 4, modified Vazquez-Santiago teaches all the elements of claim 2. Vazquez-Santiago further teaches the single kinematic chain robotic control system of claim 2, further comprising:
robotic manipulator interface circuitry configured to receive the third matrix and the third modified matrix from the matrix parsing circuitry (See at least Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1. … The variables, x θ1 t to x θ6 t , represent the manipulator joint values, while x BX t , x BY t , x Bω t represent the base (X,Y) position and orientation, respectively … We adopt the Genetic Algorithm (GA) to solve for optimal time series of configurations for each sampled target. The GA can perform a stochastic search over a large search space with different units of measurement, as is the defined problem. Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome. Multiple chromosomes, describing feasible solutions are given to the GA as an initial population for optimization.”, Page 1170 Col 1 “D. Inverse Kinematics Solver - … The closed-form expression of the analytical solution is a more computationally efficient approach than a numerical solution, such as an iterative optimization.”, discloses iterative optimization which is construed as receiving third matrix and the third modified matrix from the matrix parsing circuitry), and
mobile base interface circuitry configured to receive the fourth matrix and the fourth modified matrix from the matrix parsing circuitry (See at least Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1. … The variables, x θ1 t to x θ6 t , represent the manipulator joint values, while x BX t , x BY t , x Bω t represent the base (X,Y) position and orientation, respectively … We adopt the Genetic Algorithm (GA) to solve for optimal time series of configurations for each sampled target. The GA can perform a stochastic search over a large search space with different units of measurement, as is the defined problem. Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome. Multiple chromosomes, describing feasible solutions are given to the GA as an initial population for optimization.”, Page 1170 Col 1 “D. Inverse Kinematics Solver - … The closed-form expression of the analytical solution is a more computationally efficient approach than a numerical solution, such as an iterative optimization.”, discloses iterative optimization which is construed as receiving fourth matrix and the fourth modified matrix from the matrix parsing circuitry).
Regarding Claim 12, modified Vazquez-Santiago teaches all the elements of claim 11. Vazquez-Santiago further teaches the non-transitory storage device that includes machine-readable instructions that, when executed by one or more processors, cause the one or more processors to perform operations, of claim 11, further comprising:
…
generating a second modified matrix of joint values for each of the nJ joints of the robotic manipulator and the nM joints of the mobile base, the second modified matrix being generated based on the first modified matrix and one or more optimization parameters associated with the mobile base or the robotic manipulator (See at least Para (See at least Page 1168 Col 1 Para 5 “In this paper, we propose a novel approach of combining sampling of the end-effector path, GA optimization, linear interpolation, and inverse kinematics for the simultaneous base and manipulator motion planning optimization, with constraints of not only end-effector position, but end-effector orientation as well, Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1… We adopt the Genetic Algorithm (GA) to solve for optimal time series of configurations for each sampled target. The GA can perform a stochastic search over a large search space with different units of measurement, as is the defined problem. Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome. Multiple chromosomes, describing feasible solutions are given to the GA as an initial population for optimization.”, Page 1168 Col 2 “II. PROPOSED COMPUTATIONAL METHOD -…Optimize mobile manipulator configurations with the GA for the sampled waypoints.”, Page 1172 Col 2 “… A more stringent or flexible value could be used as a constraint parameter, depending on the application…”); and
parsing the second modified matrix into a third modified matrix of joint values for the robotic manipulator and a fourth modified matrix of joint values for the mobile base Page 1169 Col 1 “Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome.”, Page 1170 Col 1 “D. Inverse Kinematics Solver - … The closed-form expression of the analytical solution is a more computationally efficient approach than a numerical solution, such as an iterative optimization.”, discloses iterative optimization which is construed as parsing the second combined matrix to generate a third matrix of joint values for the robotic manipulator and a fourth matrix of joint values for the mobile robotic base.
Although Vazquez-Santiago discloses feedback control algorithm (See at least Page 1167 Col 2 Para 3) and laser profile sensor (See at least Fig. 1 (c)), however, he does not explicitly spell out …
generating a first modified matrix of joint values for each of the nJ joints of the robotic manipulator and nM joints of the mobile base, based on the second combined matrix and a feedback matrix, the feedback matrix comprising actual position data for the robotic manipulator joints and the mobile base joints generated by one or more sensors; …
Bosscher teaches …
generating a first modified matrix of joint values for each of the nJ joints of the robotic manipulator and nM joints of the mobile base, based on the second combined matrix and a feedback matrix, the feedback matrix comprising actual position data for the robotic manipulator joints and the mobile base joints generated by one or more sensors (See at least Para [0019] “… The vector q may be constructed, for example, from feedback obtained from the joints 18 themselves; that is, each joint 18 can include a sensor that indicates the current travel position of that joint 18 …”, Claim 6. “6. The method of claim 3 further comprising generating the m.times.(n-L) matrix J.sub.mod utilizing the following steps: determining an m.times.n Jacobian matrix J for the robotic device utilizing the current state of the robotic device, the current state of the robotic device comprising current displacements of each of the n joints; and utilizing the Jacobian matrix J to generate the matrix J.sub.mod, wherein J.sub.mod comprises all columns in J except for those columns corresponding to the L actively limited joints.”, Para [0033] “… Then, if a joint 112 that has an active limits flag 241 set (i.e. ActiveLimits.sub.i=1), the matrix column for that joint 112 is removed, in effect turning the Jacobian matrix J 211 from an m.times.n matrix into a modified m.times.(n-1) matrix so that the limited joint 112 is not represented in the modified Jacobian matrix J.sub.mod 251. Of course, more than one joint 112 may be limited, as indicated by the active limits flags 141. Letting L denote the number of joints whose limits are active, as indicated by the active limits flags 241, a total of L columns of the Jacobian matrix J 211 can be removed, resulting in a modified Jacobian matrix J.sub.mod 251 that is an m.times.(n-L) matrix.”); …
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to combine the system of Vazquez-Santiago with the teachings of Bosscher and include the feature of feedback adjustment circuitry being configured to generate a first modified matrix of joint values for each of the nJ joints of the robotic manipulator and nM joints of the mobile base, based on the second combined matrix and a feedback matrix, the feedback matrix comprising actual joint values for the robotic manipulator joints and the mobile base joints generated by one or more sensors, thereby provide robot joint calculation accuracy (See at least Para [0032] “As a fifth step 250 , improved joint limiting logic 250 accepts the joint active limits flags 241 , the operator-desired movement vector {dot over (x)}cmd 203 , the Jacobian matrix 211 and a scalar weighting factors 204 and generates a revised joint velocity command {dot over (q)}new 261…”).
Regarding Claim 13, modified Vazquez-Santiago teaches all the elements of claim 12. Vazquez-Santiago further teaches the generation of the first modified matrix of claim 12, further comprising,
determining the joint values within the first matrix which most closely corresponds to the joint values of the actual location data for the robotic manipulator and mobile base (See at least Page 1173 Col 2 Para 3 “Finally, the GA optimization was able to generate the initial configurations used to guide the creation of a continuous motion plan for the kinematically redundant mobile manipulator.”, Page 1172 Col 2 Para 1 “Shown in Table IV, are the end-effector error in position. The error was calculated as the average error percentage between the desired and achieved end-effector position, with each (X, Y, Z) coordinate represented as the magnitude of a vector”, Page 1170 Col 1 Para 2 “… Therefore, the discretized end-effector target positions are provided as input to the IK solver, along with their respective interpolated quaternion value from the GA solution …”);
inserting the joint values of the actual location data into the beginning of the first modified matrix (See at least Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1. … The variables, x θ1 t to x θ6 t , represent the manipulator joint values, while x BX t , x BY t , x Bω t represent the base (X,Y) position and orientation, respectively. Additionally, the subscripts ti , where i = 0, 1, 2, .., n, denotes the corresponding time step of the configuration.”); and
inserting joint values for a set of modified waypoints, the modified waypoints being generated so the connections between successive modified waypoints followings a path in the external environment created by the connections between successive waypoints of the motion plan request, wherein the density of waypoints is highest between the first waypoint and second waypoint and lowest between the second to last and last waypoint, the density of waypoints referring to the time between the execution of successive waypoints (See at least Page 1169 Col 1 Para 3 “B. Genetic Algorithm formulation - … The variables, x θ1 t to x θ6 t , represent the manipulator joint values, while x BX t , x BY t , x Bω t represent the base (X,Y) position and orientation, respectively. Additionally, the subscripts ti , where i = 0, 1, 2, .., n, denotes the corresponding time step of the configuration. Therefore, each column describes the configuration at every time step. Consequently, n is dependent on the desired discretization density of the path to be followed by the end-effector. Due to the nature of welding processes, the end-effector is required to move at a constant speed. To achieve a constant speed, each column configuration in Equation 1 is to be performed at the n th time step. Each time step corresponds to a path target, and its value will depend on the desired welding speed and the distance between path target positions.”).
Regarding Claim 14, modified Vazquez-Santiago teaches all the elements of claim 11. Vazquez-Santiago further teaches the non-transitory storage device that includes machine-readable instructions that, when executed by one or more processors, cause the one or more processors to perform operations, of claim 11, further comprising:
transmitting the third matrix and the third modified matrix to robotic manipulator interface circuitry configured to cause the robotic manipulator to execute the joint values contained within the third matrix and the third modified matrix (See at least Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1. … The variables, x θ1 t to x θ6 t , represent the manipulator joint values, while x BX t , x BY t , x Bω t represent the base (X,Y) position and orientation, respectively … We adopt the Genetic Algorithm (GA) to solve for optimal time series of configurations for each sampled target. The GA can perform a stochastic search over a large search space with different units of measurement, as is the defined problem. Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome. Multiple chromosomes, describing feasible solutions are given to the GA as an initial population for optimization.”, Page 1170 Col 1 “D. Inverse Kinematics Solver - … The closed-form expression of the analytical solution is a more computationally efficient approach than a numerical solution, such as an iterative optimization.”, discloses iterative optimization which is construed as transmitting the third matrix and the third modified matrix to robotic manipulator interface circuitry configured to cause the robotic manipulator to execute the joint values contained within the third matrix and the third modified matrix), and
transmitting the fourth matrix and the fourth modified matrix to mobile base interface circuitry configured to cause the mobile base to execute the joint values contained within the fourth matrix and the fourth modified matrix (See at least Page 1169 Col 1 “B. Genetic Algorithm formulation - Each discretized target has a corresponding configuration represented through variables in Equation 1. … The variables, x θ1 t to x θ6 t , represent the manipulator joint values, while x BX t , x BY t , x Bω t represent the base (X,Y) position and orientation, respectively … We adopt the Genetic Algorithm (GA) to solve for optimal time series of configurations for each sampled target. The GA can perform a stochastic search over a large search space with different units of measurement, as is the defined problem. Each element in Equation 1 is a GA gene, while the entire matrix is the GA chromosome. Multiple chromosomes, describing feasible solutions are given to the GA as an initial population for optimization.”, Page 1170 Col 1 “D. Inverse Kinematics Solver - … The closed-form expression of the analytical solution is a more computationally efficient approach than a numerical solution, such as an iterative optimization.”, discloses iterative optimization which is construed as transmitting the fourth matrix and the fourth modified matrix to mobile base interface circuitry configured to cause the mobile base to execute the joint values contained within the fourth matrix and the fourth modified matrix).
Claim(s) 10 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Vazquez-Santiago et al. (K. Vazquez-Santiago, C. F. Goh and K. Shimada, "Motion Planning for Kinematically Redundant Mobile Manipulators with Genetic Algorithm, Pose Interpolation, and Inverse Kinematics," 2021 IEEE 17th International Conference on Automation Science and Engineering (CASE), Lyon, France, 2021, pp. 1167-1174) (Hereinafter Vazquez-Santiago), in view of Shi et al. (Shi X, Guo Y, Chen X, Chen Z, Yang Z. Kinematics and Singularity Analysis of a 7-DOF Redundant Manipulator. Sensors (Basel). 2021 Oct 31) (Hereinafter Shi), Jensen-Nau et al. (K. R. Jensen-Nau, T. Hermans and K. K. Leang, "Near-Optimal Area-Coverage Path Planning of Energy-Constrained Aerial Robots With Application in Autonomous Environmental Monitoring," in IEEE Transactions on Automation Science and Engineering, vol. 18, no. 3, pp. 1453-1468, July 2021), and further in view of in view of Solinas et al. (US 20230177333 A1) (Hereinafter Solinas).
Regarding Claim 10, modified Vazquez-Santiago teaches all the elements of claim 1.
However, Vazquez-Santiago does not explicitly spell out the single kinematic chain robotic
control system of claim 1, wherein the optimization circuitry is further configured to generate the second combined matrix using gradient descent cost optimization.
Solinas teaches the single kinematic chain robotic control system of claim 1, wherein the
optimization circuitry is further configured to generate the second combined matrix using gradient descent cost optimization (See at least Para [0058] “… the choice of optimizer for the gradient descent and the cost function, are also for example selected manually.”, Para [0053] “… The model has a hidden layer with seven output hidden neurons, and thus corresponds to a matrix of dimensions [AltContent: rect].sup.2*7…”, Para [0075] “The one or more actuators 322 for example comprise a robotic system, such as a robotic arm trained to pull up weeds, or to pick ripe fruit from a tree…”).
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to combine the system of Vazquez-Santiago with the teachings of Solinas and include the feature of optimization circuitry being configured to generate the second combined using gradient descent cost optimization, thereby improve computational accuracy (See at least Para [0147] “An advantage of the embodiments described herein is that a control system and method based on artificial intelligence can have improved accuracy, and reduced forgetting, based on the use of iterative sampling to generate pseudo-samples. The control system is thus able to provide improved control of actuators, leading to technical advantages over prior control systems.”).
Regarding Claim 20, modified Vazquez-Santiago teaches all the elements of claim 11.
However, Vazquez-Santiago does not explicitly spell out the non-transitory storage device that
includes machine-readable instructions that, when executed by one or more processors, cause the one or more processors to perform operations, of claim 11, further comprising, generating the second combined matrix using gradient descent cost optimization..
Solinas teaches the non-transitory storage device that includes machine-readable instructions
that, when executed by one or more processors, cause the one or more processors to perform operations, of claim 11, further comprising, generating the second combined matrix using gradient descent cost optimization. (See at least Para [0058] “… the choice of optimizer for the gradient descent and the cost function, are also for example selected manually.”, Para [0053] “… The model has a hidden layer with seven output hidden neurons, and thus corresponds to a matrix of dimensions [AltContent: rect].sup.2*7…”, Para [0075] “The one or more actuators 322 for example comprise a robotic system, such as a robotic arm trained to pull up weeds, or to pick ripe fruit from a tree…”).
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to combine the system of Vazquez-Santiago with the teachings of Solinas and include the feature of generating the second combined using gradient descent cost optimization, thereby improve computational accuracy (See at least Para [0147] “An advantage of the embodiments described herein is that a control system and method based on artificial intelligence can have improved accuracy, and reduced forgetting, based on the use of iterative sampling to generate pseudo-samples. The control system is thus able to provide improved control of actuators, leading to technical advantages over prior control systems.”).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Hoppe (US 20080188986 A1) teaches a method and system to provide improved
accuracies in multi jointed robots through kinematic robot model parameters determination
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SHAHEDA HOQUE whose telephone number is (571)270-5310. The examiner can normally be reached Monday-Friday 8:00 am- 5:00 pm.
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.
/SHAHEDA HOQUE/Examiner, Art Unit 3658
/Ramon A. Mercado/Supervisory Patent Examiner, Art Unit 3658
Prosecution Insights