Questions about how individual parts of claim 1 might be changed. A question may describe something different from what the claim language recites, or a version of it; no view is offered on which.
1.1Three axis linear gantry
What ifWhat if a gantry with three linear stages positioned the end effector?
Claim language“an arm configured to change a position of an end effector”
A fixed bridge supports a cross carriage, and the cross carriage carries a vertical slide. Servo-driven ball screws move the bridge along base rails, the carriage across the bridge, and the slide vertically to set the end effector position.
1.2Hardwired analog servo control
What ifWhat if hardwired analog circuits controlled the robot axes?
Claim language“a processor”
An analog summing amplifier subtracts each joint potentiometer voltage from its command voltage. A resistor-capacitor compensation network shapes the error signal, and a transistor bridge drives the joint motor until the feedback voltage matches the command.
1.4Direct optical position feedback
What ifWhat if the controller used live measurements of end effector position to calculate corrections?
Claim language“information representing a structural characteristic of the arm”
Two fixed cameras track a rigid marker cluster mounted near the end effector and calculate its position in the workcell frame. The controller subtracts that measured position from the target position and converts the difference into incremental joint commands.
1.5Fixed logic command correction
What ifWhat if a hardwired arithmetic circuit applied the correction directly to the joint commands?
Claim language“instructions that when executed by the processor”
Digital comparators check encoder counts against region limits and enable a dedicated correction circuit. Within those limits, a fixed-point multiplier applies a stored compliance coefficient to a joint torque estimate, and an adder feeds the corrected position command to the drive.
Section 2 · machine-generated
Variations
Further arrangements of the same subject matter, described here as of the publication date at the foot of this page.
Other field of use
Tray handling in a greenhouse
A rail-mounted arm uses a fork gripper to place seedling trays on greenhouse racks. Controller memory stores a vertical stiffness curve measured for the arm at horizontal reaches from 0.8 to 1.2 metres. A wrist load cell measures the downward force from the gripper and tray after pickup, and the processor divides that force by the stiffness at the commanded reach to estimate sag. Within that reach band, the processor adds the estimated sag to the vertical position command before computing joint targets.
Component substitution
Flexure elbow in a positioning arm
A short-reach positioning arm uses a crossed-leaf flexure pivot in place of a rolling-element elbow bearing. A rotary actuator drives the elbow, and a separate angle encoder measures elbow position. Controller memory stores the measured transverse displacement of the flexure pivot as a function of elbow angle over a calibrated interval of 5 to 12 degrees. Within that interval, the processor includes the displacement in the forward kinematic model and solves for joint targets that place the end effector at the requested position.
Parameter or scale variation
Faster updates at full reach
A 2.4-metre handling arm uses a joint servo that runs at 1 kilohertz and a structural correction task that runs at 500 hertz. For horizontal reaches from 1.8 to 2.3 metres, memory stores a three-by-three matrix that maps end-effector forces to translational deflection, with matrices measured at 50-millimetre reach increments along the transfer path. The correction task interpolates the matrix at the commanded reach and multiplies it by the gravity force calculated from the configured tool and payload mass. The servo interpolates between successive correction samples and subtracts the resulting deflection vector from the Cartesian target before computing joint commands.
Combination with existing systems
Load compensation at a sanding station
A robot arm carries a sanding head on a spring-loaded slide with an inline force sensor. Controller memory stores the arm's angular compliance about two tool axes over the joint configurations used to sand the edges of curved panels. The processor combines the measured contact force with the offset from the wrist to the contact point to calculate the bending moment, then uses the stored compliance to estimate tool tilt. Within the calibrated configurations, the processor applies an opposing orientation correction while the slide regulates contact force along the surface normal.
Automated or data-driven variant
Camera calibration of link compliance
Two fixed cameras track a rigid marker cluster attached to the end-effector mount during an automated calibration routine. The controller moves the arm through a grid of poses in the transfer region and repeats the grid with two known payload masses. Calibration software fits forearm bending and joint torsional compliance parameters to the measured changes in tool pose between the two loads at each grid point. The processor stores the fitted parameters and uses them with the configured payload mass to compensate tool position during transfers within the calibrated region.
Thermal drift correction during warmup
Temperature sensors bonded to the upper arm, forearm and wrist housing measure uneven heating during repeated machine loading. Controller memory stores thermal expansion coefficients and measured mappings from link temperature changes to endpoint displacement over the loading poses. The processor samples the sensors once per second, filters the readings and estimates the displacement in robot base coordinates at the commanded pose. Within the loading region, the processor subtracts that displacement from the Cartesian target and ramps each updated correction over the next sampling interval.
Specific version of a part
Streamed cyclic drive commands
An external controller calculates corrected joint setpoints and sends them to the robot drives over a wired link every millisecond. Each drive uses the latest setpoint and local encoder feedback to close its position loop, and the external controller retains the correction data and program.
Workspace wide gravity correction
The controller uses joint angles and payload mass to calculate the gravity torque at each joint. It multiplies those torques by stored joint compliance values and adds the resulting deflection estimates to the joint position commands on every control cycle.
Two link horizontal arm
Two rigid links rotate about parallel vertical axes to position the end effector in a horizontal plane. A ball screw at the distal end raises and lowers the end effector.
Dedicated motion control processor
A 32-bit processor with a floating-point unit reads joint encoder counts on a 1 kHz timer interrupt. It computes the joint commands and transmits them to the servo drives before the next interrupt.
Wrist compliance calibration matrix
Memory stores a six-by-six compliance matrix for the wrist assembly over a specified range of arm postures. Calibration loads and moments applied at the tool flange produce measured translations and rotations, and a fitted matrix maps those loads and moments to pose error.
Region gated deflection compensation
Stored instructions compare the shoulder and elbow encoder angles with the limits of a calibrated operating region. Inside that region, the processor uses stored stiffness coefficients and estimated joint torques to compute deflection offsets and adds them to the commanded joint positions.
Section 3 · quoted from USPTO records
Other work
Passages reproduced word for word from two earlier-published U.S. patent documents, placed beside the language of claim 1. No characterization is offered. Each passage is marked A or B for the document it comes from; ¶ numbers are the paragraph numbers in the USPTO’s published full text of that document.
1.1a robot having an arm configured to change a position of an end effector;
A robot control apparatus includes a storage unit that stores an operating program and a kinematic parameter used in a formula representing a relationship between displacement of each drive axis of a robot and a position and an orientation of a leading end of the robot and a drive unit that operates the drive axis of the robot based on the operating program and the kinematic parameter stored in the storage unit.
A · US 2017/0341234 A1 · abstract
The robot includes a robotic arm having at least one joint axis, and an end effector provided to a tip end of the robotic arm and configured to hold the substrate, and the robot is disposed adjacent to an installation position at which the substrate is placed.
B · US 2025/0026588 A1 · ¶ 8
1.2a processor; and
As illustrated in FIG. 2, the robot control device 30 according to this embodiment includes a memory 32 and a processor 34 which executes a program stored in the memory 32.
B · US 2025/0026588 A1 · ¶ 31
1.3a memory storing:
1.4first information representing a structural characteristic of the arm that causes a motion error of the robot in a first operating region; and
1.5instructions that when executed by the processor, cause the processor to perform operations comprising controlling the robot so as to reduce the motion error based on the first information when the robot operates within the first operating region.
In order for the leading end of the robot to assume a position and an orientation at the time of teaching even when the kinematic parameter has been updated after teaching of the operating program and when the drive axis is operated by executing the operating program using the present kinematic parameter, the storage unit stores the kinematic parameter before updating, and the drive unit corrects position data of at least one teaching point in the operating program based on t
A · US 2017/0341234 A1 · ¶ 6
An empty cell means only that no passage was reproduced there.
Statements in these documents
In order to improve the positioning accuracy of a robot, kinematic parameters of the robot, used in robotic control, are calibrated, as known in related art (for example, refer to Japanese Unexamined Patent Application, Publication No. 2008-012604 (“JP '604”)).
A · US 2017/0341234 A1 · ¶ 3
However, the transferring the substrate while holding may not promptly be performed because of the conduction of the determination.
B · US 2025/0026588 A1 · ¶ 6
As a result, the robot control device according to the present disclosure can promptly transfer the substrate while holding the substrate.
B · US 2025/0026588 A1 · ¶ 9
Cite aspatent.fyi, US 2026/0355951 A1, claim 1, first published 2026-11-19T17:33:15-05:00, https://patent.fyi/US20260355951/, SHA-256 2640071f306920ca421afdcd1949ea3cb7f65ce0975009402a32fb6d48dbc2e5