Derivative works — US Patent 10001780

Excellent. As a Senior Patent Strategist and Research Engineer, my objective is to create a robust body of prior art derived from US Patent 10,001,780. This Defensive Disclosure is designed to preemptively render incremental innovations by competitors as obvious extensions of the core concepts, thereby securing freedom to operate.

This document does not summarize US Patent 10,001,780; it builds upon its foundational claims to disclose novel and non-obvious variations. The core inventive concept is understood as a force-directed graph approach to dynamic path planning, where "route poses" are nodes subject to repulsive forces from obstacles and attractive (cohesive) forces from adjacent nodes in the planned path.

Defensive Disclosure Based on US Patent 10,001,780

Publication Date: 2026-05-13
Subject: Derivative Methods and Systems for Force-Directed Dynamic Navigation

Derivative Set 1: Based on Core Claim 1 (System Claim)

The following disclosures expand upon the system described in US Patent 10,001,780, which comprises sensors and a controller for force-based dynamic route planning.

Variation 1.1: Material & Component Substitution

Title: Autonomous Navigation System with Magneto-Rheological (MR) Fluid-Based Haptic Boundary Sensing.
Enabling Description: This variation replaces conventional optical or acoustic sensors (LIDAR, Sonar) with a distributed network of low-power, piezoelectric pressure sensors coupled with a microfluidic system containing a magneto-rheological (MR) fluid. The robot's chassis is surrounded by a flexible, tactile bumper filled with this MR fluid. When the robot approaches an object, a low-level magnetic field is projected. The reflection and distortion of this field, detected by Hall effect sensors, causes a change in the MR fluid's viscosity, which in turn alters the pressure readings from the piezoelectric sensors. The controller interprets these pressure differentials as repulsive force vectors. The "footprint" of each route pose is modeled not as a rigid shape but as a deformable mesh, with the attractive forces between pose points calculated using a spring-mass model based on Hooke's Law, where the spring constant is dynamically adjusted based on vehicle speed. This provides a computationally efficient, non-optical method for sensing and reacting to proximal obstacles.

Diagram:

graph TD
    subgraph Robot Chassis
        A[Controller]
        B[MR Fluid Reservoir]
        C[Piezoelectric Sensor Array]
        D[Hall Effect Sensors]
        E[Electromagnet Emitters]
    end

    subgraph Environment
        F[Obstacle]
    end

    A --> B
    A --> E
    C --> A
    D --> A
    E -- Magnetic Field --> F
    F -- Field Distortion --> D
    F -- Proximity Pressure --> C

    subgraph Force Calculation
        A -- Pressure Data --> G{Compute Repulsive Force}
        A -- Pose Data --> H{Compute Attractive Force (Hooke's Law)}
        G --> I[Reposition Pose]
        H --> I
    end

Variation 1.2: Operational Parameter Expansion

Title: Nanoscale Route Planning for Targeted Drug Delivery via Force-Field Directed Microrobots.
Enabling Description: This disclosure applies the force-directed navigation method to a swarm of nanoscale robots (nanobots) operating within a biological vascular system. The "map" is a 3D model of a capillary network generated from real-time ultrasound micro-tomography. Each nanobot is a "route pose." Repulsive forces are not from physical objects but are generated by chemical gradients (chemotaxis) of specific cellular biomarkers indicating unhealthy tissue. The nanobots' sensors are functionalized surface proteins that bind to these biomarkers, creating a "repulsive" signal. Attractive forces are generated by inter-swarm communication using localized, low-intensity acoustic pulses, ensuring the swarm maintains cohesion. The controller, an external magnetic guidance system, calculates the aggregate force vectors on the swarm and repositions it by modulating the magnetic field, steering the nanobots toward a target (e.g., a tumor) while avoiding healthy tissue regions that exert repulsive chemical forces.

Diagram:

sequenceDiagram
    participant Ext_Controller as External Controller
    participant Mag_Field as Magnetic Guidance System
    participant Nanobot_Swarm as Swarm (Poses)
    participant Bio_Env as Biological Environment
    participant Target_Cell as Target Cell
    participant Healthy_Cell as Healthy Cell

    Ext_Controller->>Bio_Env: Generate 3D map via Ultrasound
    Healthy_Cell-->>Nanobot_Swarm: Exert Repulsive Chemical Gradient
    Nanobot_Swarm->>Nanobot_Swarm: Maintain cohesion via Acoustic Pulses (Attractive Force)
    Nanobot_Swarm->>Ext_Controller: Report aggregate force vectors
    Ext_Controller->>Mag_Field: Calculate new guidance parameters
    Mag_Field->>Nanobot_Swarm: Reposition Swarm via modulated magnetic field
    Nanobot_Swarm->>Target_Cell: Interpolate path and navigate towards target

Variation 1.3: Cross-Domain Application (Aerospace)

Title: Dynamic Trajectory Planning for Satellite Debris Avoidance in Low Earth Orbit (LEO).
Enabling Description: The system is applied to a satellite or spacecraft in LEO. The "environment" is a 4D map (3D space + time) of known orbital debris objects, updated continuously from ground-based radar tracking data (e.g., from the Space Surveillance Network). The satellite's planned trajectory is a series of "route poses" in spacetime. Each piece of debris is a source of a repulsive force, calculated based on its predicted proximity and relative velocity. The magnitude of the repulsive force is inversely proportional to the predicted time-to-closest-approach and proportional to the debris's kinetic energy. Attractive forces maintain the trajectory's orbital mechanics constraints, pulling poses toward the optimal orbital path to conserve fuel. The controller, an onboard flight computer, constantly recalculates the force equilibrium and, when a net force exceeds a threshold, commands micro-thruster burns to reposition the satellite onto the newly interpolated, collision-free trajectory.

Diagram:

stateDiagram-v2
    [*] --> Nominal_Orbit
    Nominal_Orbit --> Debris_Detected: Debris enters threat threshold
    Debris_Detected --> Force_Calculation: Compute F_repulsive(debris) & F_attractive(orbit)
    Force_Calculation --> Trajectory_Recalculation: Net Force > Threshold
    Trajectory_Recalculation --> Thruster_Burn: Interpolate new path
    Thruster_Burn --> Corrected_Orbit: Satellite repositioned
    Corrected_Orbit --> Nominal_Orbit: Threat cleared
    Debris_Detected --> Nominal_Orbit: Debris no longer a threat

Variation 1.4: Cross-Domain Application (AgTech)

Title: Precision Pollination Drone Swarm Navigation in Complex Orchard Environments.
Enabling Description: A swarm of small autonomous drones is tasked with pollinating fruit trees. The "map" is a 3D point cloud of an orchard generated by a master drone using LIDAR. Individual blossoms identified via hyperspectral imaging are "targets" that exert an attractive force, while branches, leaves, and other drones are sources of repulsive forces. Each drone's planned path is a series of route poses. The attractive force from a blossom is proportional to a "pollination-priority" score (based on age and viability). The repulsive force from foliage is constant. The cohesive (attractive) force between drones is modeled to maintain a minimum separation distance to avoid air-wake turbulence. The onboard controller on each drone computes these forces and dynamically plans its path from blossom to blossom, maximizing pollination efficiency while ensuring no collisions.

Diagram:

classDiagram
  class Drone {
    +ID: int
    +position: Vector3D
    +plannedPath: list[RoutePose]
    +controller: Controller
    +pollinate()
  }
  class RoutePose {
    +position: Vector3D
    +orientation: Quaternion
    +footprint: Polygon
  }
  class Controller {
    +map: PointCloud
    +updateForces()
    +replanPath()
  }
  class ForceSource {
    +position: Vector3D
    +calculateForce(pose: RoutePose): Vector3D
  }
  class Blossom {
    <<Attractive>>
    +pollinationPriority: float
  }
  class Obstacle {
    <<Repulsive>>
    +type: string
  }
  class FellowDrone {
    <<Repulsive>>
  }

  Drone "1" -- "1" Controller
  Drone "1" -- "N" RoutePose
  Controller "1" -- "1" PointCloud
  ForceSource <|-- Blossom
  ForceSource <|-- Obstacle
  ForceSource <|-- FellowDrone

Variation 1.5: Integration with Emerging Tech (AI/IoT)

Title: AI-Modulated Force-Field Navigation with Real-Time IoT Hazard Data.
Enabling Description: The core force-based navigation system is enhanced by an AI model (a trained neural network) that dynamically adjusts the parameters of the force functions. The robot is integrated into an IoT ecosystem. For example, in a smart warehouse, IoT sensors on shelves detect spills (repulsive force magnitude increases), report high-traffic zones from other devices (repulsive force field is widened), or signal freshly cleaned floors from other robots (attractive force is applied to guide the robot to uncleaned areas). The AI model is trained on historical data of navigation events (e.g., near-misses, battery consumption, task completion time) to predict optimal force parameters. For instance, it learns to reduce the magnitude of attractive forces (allowing for wider deviations) in cluttered areas to find safer paths, even if they are less direct, and increase them in open areas to optimize for speed.

Diagram:

flowchart LR
    subgraph IoT_Ecosystem
        A[Spill Sensor]
        B[Traffic Monitor]
        C[Other Robots]
    end
    subgraph Robot_System
        D[Onboard Controller]
        E[Force Function Module]
        F[AI Model (NN)]
    end
    G[Route Poses]

    A -- Spill Data --> F
    B -- Traffic Data --> F
    C -- Area Coverage Data --> F
    F -- Modulated Parameters (e.g., repulsion_gain, attraction_k) --> E
    E -- Forces --> G
    G -- Repositioned Poses --> D
    D --> G

Variation 1.6: The "Inverse" or Failure Mode

Title: Graceful Degradation Navigation via Potential Field Minimization.
Enabling Description: In the event of a primary sensor (e.g., LIDAR) failure, the system enters a "low-power" or "safe" mode. In this mode, it relies solely on low-resolution, short-range proximity sensors (e.g., infrared) and odometry. The complex, multi-point "route pose" is simplified to a single point. The navigation algorithm switches from a dynamic force-balancing calculation to a simple potential field method. All objects detected by the proximity sensors generate a high-repulsion potential field, while the direction toward the last known "home" or "safe" location (stored in memory) generates a constant, weak attractive potential. The robot's motion is governed by gradient descent, always moving in the direction that minimizes its potential energy. This ensures the robot will attempt to retreat to a safe location while avoiding immediate collisions, without the computational overhead of the full force-directed interpolation model. The attractive force to the original path is set to zero, prioritizing safety over task completion.

Diagram:

stateDiagram-v2
    state "Full Operation Mode" as FullOp {
        direction LR
        state "Sensor Fusion" as SF
        state "Force-Directed Planning" as FDP
        state "Path Interpolation" as PI
        [*] --> SF
        SF --> FDP
        FDP --> PI
        PI --> SF
    }

    state "Safe Mode (Potential Field)" as SafeMode {
        direction LR
        state "Proximity Sensing" as PS
        state "Gradient Descent Navigation" as GDN
        [*] --> PS
        PS --> GDN
        GDN --> PS
    }

    FullOp --> SafeMode: Primary Sensor Failure
    SafeMode --> FullOp: Sensor Recovered
    SafeMode --> Docking_Attempt: Path to Home Found
    SafeMode --> Halted: No Safe Path / Collision Imminent

Derivative Set 2: Based on Core Claim 9 (Method Claim)

The following disclosures expand upon the method of dynamic navigation described in US Patent 10,001,780.

Variation 2.1: Material & Component Substitution

Title: Quantum Annealing Method for Global Optimization of Force-Based Route Pose Placement.
Enabling Description: This method replaces the iterative, greedy repositioning of route poses with a global optimization approach using a quantum annealer (or a simulated annealer on classical hardware). The state of all route poses along a path segment is encoded into a single QUBO (Quadratic Unconstrained Binary Optimization) problem. The objective function to be minimized is the total "energy" of the system, where the energy is the sum of all potential energies from repulsive forces (from obstacles) and attractive forces (from path cohesion). The quantum annealer finds the ground state of this QUBO problem, which corresponds to the globally optimal placement of all route poses simultaneously, avoiding local minima that an iterative approach might fall into. This is particularly effective for navigating through highly complex and constrained spaces (e.g., a maze-like environment) where a local adjustment could trap the robot.

Diagram:

graph TD
    A[Define Path with N Poses] --> B{Formulate QUBO};
    B -- Pose Positions & Forces --> C[Quantum Annealer];
    D[Map & Obstacle Data] --> B;
    C -- Solved Ground State --> E{Decode Optimal Pose Positions};
    E --> F[Generate Final Path via Interpolation];

Combination Prior Art Scenarios

The following disclosures combine the teachings of US Patent 10,001,780 with existing open-source standards to produce obvious implementations.

Combination with ROS (Robot Operating System): The force-directed planning method is implemented as a ROS global_planner plugin. It subscribes to a costmap_2d topic, treating all cells with a lethal or inscribed obstacle cost as sources of repulsive force. It publishes a nav_msgs/Path message containing the interpolated, collision-free path. The attractive/repulsive force parameters (e.g., gains, thresholds) are exposed as dynamic reconfigure parameters, allowing them to be tuned in real-time. The "route poses" are implemented as a custom message type, force_planner/RoutePoseArray. This combination makes the patented method a modular, plug-and-play component within the standard ROS navigation stack, rendering it an obvious integration for anyone skilled in the art of robotics and ROS.
Combination with PX4 Autopilot: For drone navigation, the method is integrated into the PX4 flight control software. The algorithm runs as a module on the companion computer, receiving obstacle data from a connected sensor (e.g., an Intel RealSense camera) via the MAVLink protocol. The repulsive and attractive forces are calculated, and the resulting interpolated path is converted into a series of MAVLink TRAJECTORY_REPRESENTATION_WAYPOINTS messages. These messages are sent to the PX4 flight controller, which then executes the trajectory. This represents a straightforward application of the method to a widely used open-source drone autopilot system.
Combination with Blender (Physics Engine): The dynamic route planning method is implemented as a Python script within the open-source 3D modeling software, Blender. A 3D model of an environment is created or imported. The robot's path is represented as a "Curve" object, and the route poses are "Hooks" attached to the curve's control points. Obstacles in the scene are configured as "Collision" objects in Blender's physics engine, which generate a repulsive force field. The curve's control points (the poses) are linked by "Spring" physics constraints, providing the attractive force. Running the physics simulation (bpy.ops.ptcache.bake_all()) causes the curve to deform and settle into a collision-free path, which can then be exported as a series of coordinates. This demonstrates that the core method is an obvious application of principles already present in standard 3D physics simulation engines.