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.
1.5Rotary motor and ball screw
What ifWhat if each actuator used a rotary motor and ball screw to change its length?
Claim language“a linear motor”
A rotary servo turns a ball screw, and a nut constrained against rotation moves the actuator rod along its axis. A linear encoder measures rod travel, and an inline load cell measures axial force for the compliance controller.
1.6Pose feedback with velocity trials
What ifWhat if the controller found the target posture by testing small actuator velocity commands and measuring the resulting platform motion?
Claim language“performing an inverse kinematics calculation based on the target posture”
An external pose sensor measures translation and orientation, and the controller combines their errors in a weighted error value. The controller tests short positive and negative velocity pulses on each actuator and repeats pulses that reduce the error until it falls within a specified tolerance.
1.7Factory calibrated force maps
What ifWhat if the controller retrieved actuator force settings from calibration tables indexed by platform posture and load direction?
Claim language“calculating a Jacobina matrix based on the target posture”
A calibration fixture applies known forces and moments at a grid of platform postures while six inline load cells record the corresponding changes in actuator force. During operation, the controller interpolates those recorded values between grid points and adds the resulting force offsets to the holding forces.
1.8Compliance through pose commands
What ifWhat if the compliance controller issued a revised platform posture as its output?
Claim language“generating a force command correspondingly”
A six-axis load sensor measures the external forces and moments, and a controller compares each component with its directional threshold. The controller converts each excess component into a bounded translation or rotation increment and sends the revised posture to the actuator position loops.
1.9Brake holds actuator length
What ifWhat if spring-applied rod brakes held the target posture after the motors finished positioning the platform?
Claim language“controlling a corresponding linear motor based on the position command and the force command”
Each motor moves its rod to the commanded length, and a spring-applied clamp grips the rod once it settles within position tolerance. When a platform load sensor detects a threshold crossing, the controller establishes current-limited motor control at the measured lengths and releases the clamps to permit motion.
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
Compliant insertion of ceramic cores
Six jointed linear actuators connect a fixed base to a moving platform that carries a padded gripper for ceramic casting cores. A fixture below the platform locates the receiving die, and the controller converts the programmed insertion poses into actuator lengths. During insertion, a lateral load above its directional limit shifts the commanded pose sideways in the direction of the load. An axial load above its limit pauses the feed until the load falls below the limit.
Component substitution
Ball screw drives with axial load cells
A rotary servomotor and a preloaded ball screw replace the linear motor in each of the six jointed actuators. A linear encoder measures extension, and an inline load cell measures axial force between the actuator rod and its platform joint. Each drive converts the commanded extension into screw rotation using the screw lead and uses the load cell reading in its force control loop. Above a directional load limit, the controller shifts the target pose in the direction of the excess load and recalculates the six extensions.
Parameter or scale variation
Short stroke stage for optical seating
Six jointed linear actuators with 10 mm of usable stroke connect a 140 mm base to an 80 mm moving platform that carries a lens holder. The controller centers the actuators at 5 mm extension in the neutral pose and limits seating speed to 0.5 mm/s. The controller updates the pose and force distribution at 1 kHz and applies a 1 N axial resistance setting during seating. When the axial contact load exceeds 1 N, the controller shifts the commanded pose opposite the seating direction until the load returns to the setting.
Combination with existing systems
Gantry positioning with local polishing compliance
A three-axis gantry carries the base of the six-actuator robot, and the moving platform carries a rotary polishing spindle. The gantry traverses the programmed surface path while the moving platform adjusts the spindle angle and makes short corrections along the surface normal. The controller uses the current Jacobian and the six measured actuator loads to estimate contact load along that normal. When the load exceeds the selected limit, the moving platform backs the spindle away and the gantry holds its path position until the load returns below the limit.
Automated or data-driven variant
Force sensor feedback with directional limits
A six-axis force and torque sensor mounts between the moving platform and the tool. The controller subtracts the tool's gravity contribution using its stored mass and center of mass, then transforms the measured loads into tool coordinates. It compares each signed load component with its configured limit and converts the excess into a bounded platform translation or rotation using separate gains for force and torque. The controller recalculates all six actuator lengths from the adjusted pose and ramps the pose offset toward zero as the load subsides.
Camera registration of fixture coordinates
A camera fixed to the base views three noncollinear reference marks on a removable workpiece fixture. A calibrated imaging routine estimates the fixture pose and transforms stored tool poses and directional resistance settings from fixture coordinates into robot base coordinates. The controller converts each transformed pose into six actuator lengths and computes the force distribution at that pose. When a load exceeds a limit, the controller applies the compliant pose offset along the corresponding fixture axis and updates the actuator commands.
Specific version of a part
Ribbed aluminum moving platform
A machined aluminum plate carries three pairs of joint mounting pads on its lower surface. Integral ribs connect the mounting pads to a central tool mounting boss and stiffen the plate under bending loads.
Welded steel base platform
A welded steel box frame supports a top plate with six machined joint seats. Three adjustable feet level the base and provide stable support on an uneven work surface.
Preloaded spherical platform joints
Hardened steel ball studs fasten to threaded bosses on the moving platform surface, and split sockets on the actuator ends capture the balls. Adjustable socket preload limits clearance while allowing each actuator to pivot in two directions and rotate about its axis.
Position loop with force feedforward
The driver uses a linear encoder to calculate a corrective current from the difference between commanded and measured actuator displacement. It converts the axial force command to a feedforward current using the motor force constant and adds it to the corrective current before applying the drive current limit.
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.
In each limb, the revolute or universal joint is attached to the base while the spherical joint is on the platform, and prismatic joints are connected in between.
A · US 2016/0250749 A1 · ¶ 47
1.3multiple first joint components, respectively arranged on the first surface;
The proposed 2T2R parallel mechanism consists of two RPS limbs and two UPS limbs as shown in FIG. 2(b). In each limb, the revolute or universal joint is attached to the base while the spherical joint is on the platform, and prismatic joints are connected in between.
A · US 2016/0250749 A1 · ¶ 47
1.4multiple second joint components, respectively arranged on the second surface;
1.5six linear actuators, respectively arranged between the moving platform and the base platform through the multiple first joint components and the multiple second joint component, each of the linear actuators respectively comprising a linear motor and a motor driver, wherein the moving platform has a target posture and multiple resistance forces relative to multiple directions, and each of the motor drivers is respectively configured to execute following actions:
1.6action a) performing an inverse kinematics calculation based on the target posture to obtain a linear displacement for each of the linear actuators to follow with to maintain the moving platform at the target posture and generating a position command correspondingly;
It is noted that given R and p the inverse kinematic solution can be obtained directly from (1) to give the four input limb lengths li.
A · US 2016/0250749 A1 · ¶ 53
After the preset end pose (i.e., the pose of the end effector set in advance) of each robotic arm, that is, the preset pose of an end effector (e.g., a fixture) of each robotic arm of the redundant robot is given, a set of joint angles (i.e., the angles of the joints of the redundant robot) are obtained through inverse kinematics and transmitted to the servo of the joints for controlling the joints, so that the redundant robot performs the corresponding motion.
B · US 2022/0009096 A1 · ¶ 27
1.7action b) calculating a Jacobina matrix based on the target posture;
Thus J is the 6 by 6 overall Jacobian matrix. The first four rows are the four actuation forces represented by actuation Jacobian Ja in (10) while the last two rows are constraint forces denoted by constraint Jacobian Jc.
A · US 2016/0250749 A1 · ¶ 68
In this embodiment, after the preset end pose is given, a Jacobian matrix of the configuration is solved and calculated with the concept of the corresponding null space of the Jacobian matrix (i.e., the Jacobian matrix null space) and the energy cost function, so as to obtain the configuration of the joint angle with optimal energy to take as the required inverse kinematics solution.
B · US 2022/0009096 A1 · ¶ 27
1.8action c) calculating a force distribution of each of the six linear actuators based on the Jacobian matrix and the multiple resistance forces and generating a force command correspondingly;
1.9action d) controlling a corresponding linear motor based on the position command and the force command to maintain the moving platform at the target posture; and
The calculated set of joint angles are used as the optimal inverse kinematics solution for making the redundant robot to maintain the same end pose, which is transmitted to the corresponding servo of the joint so as to perform positional control of joint, thereby realizing the required robot motion.
B · US 2022/0009096 A1 · ¶ 42
1.10action e) controlling the corresponding linear motor to perform a compliant motion when the moving platform is subjected to an external force which exceeds one of the multiple resistance forces.
An empty cell means only that no passage was reproduced there.
Statements in these documents
Finding the optimized locations of the two UPS (universal-prismatic-spherical joints) limbs is the main objective in the optimal design for giving a large singularity-free workspace and good kinematics performance represented by the condition number of a dimensionless Jacobian matrix.
A · US 2016/0250749 A1 · ¶ 5
An optimized mechanism configuration is found and the effect of joint components is also demonstrated, resulting in a practical design for AFP and other applications.
A · US 2016/0250749 A1 · ¶ 5
Because inverse kinematics of redundant robot has infinite solutions, how to efficiently solving has become a widespread concern.
B · US 2022/0009096 A1 · ¶ 3
However, the disadvantage of this calculation method is that the amount of calculation is large, the setting of the gain matrix needs to consider the computing power of the system, and the result of the iteration is uncontrollable.
B · US 2022/0009096 A1 · ¶ 5
Cite aspatent.fyi, US 2026/0355946 A1, claim 1, first published 2026-11-19T17:33:15-05:00, https://patent.fyi/US20260355946/, SHA-256 7d09185dede7c5447bad1ba347606fdab3585923f96b3d111a594a72cc31d2ba