Systems and methods of detecting intent of spatial control
Inventors
Summer, Matthew D. • Bowman, William S. • Falendysz, Andrew D. • Makovy, Kevin M. • Hedman, Daniel R. • Truesdell, Bradley D.
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12189386-B2
Publication Date
2025-01-07
Expiration Date
Abstract
Systems and methods of manipulating/controlling robots. In many scenarios, data collected by a sensor (connected to a robot) may not have very high precision (e.g., a regular commercial/inexpensive sensor) or may be subjected to dynamic environmental changes. Thus, the data collected by the sensor may not indicate the parameter captured by the sensor with high accuracy. The present robotic control system is directed at such scenarios. In some embodiments, the disclosed embodiments can be used for computing a sliding velocity limit boundary for a spatial controller. In some embodiments, the disclosed embodiments can be used for teleoperation of a vehicle located in the field of view of a camera.
Core Innovation
The invention relates to detecting an intent of vehicle control using gesture data generated with respect to a global reference frame. The gesture data is representative of a desired motion of at least a portion of a vehicle, and it represents desired motion as desired linear velocity or desired angular velocity. The method computes an initial desired velocity vector from the gesture data and uses it to derive control information for the vehicle portion.
The invention transforms the initial desired velocity vector from the global reference frame into a control reference frame. It then identifies a nearest principal axis direction parallel to principal axes of the control reference frame by comparing the direction of the initial desired velocity vector with a set of principal axis directions. A misalignment angle is computed as a deviation from the nearest principal axis direction, and when the misalignment angle is less than or equal to an axis-snapping tolerance value, the method defines a final desired velocity vector by rotating the initial desired velocity vector to align with the nearest principal axis direction.
The invention also detects an intended regime of a velocity command by computing desired linear velocity vector and desired angular velocity vector from gesture data. It computes a linear ratio by dividing a first magnitude of the desired linear velocity vector by a linear magnitude threshold and an angular ratio by dividing a second magnitude of the desired angular velocity vector by an angular magnitude threshold. Based on rules associated with the desired linear velocity vector or the desired angular velocity vector, the method ignores the desired linear velocity vector and/or ignores the desired angular velocity vector, and sends data indicating the desired linear velocity vector and the desired angular velocity vector to at least the portion of the vehicle.
Claims Coverage
The document includes three independent claims across a method for detecting vehicle control intent, a method for detecting an intended regime of a velocity command, and one or more non-transitory computer-readable media storing corresponding instructions. Across these, the claims introduce two main inventive structures: principal-axis direction intent detection with axis snapping, and regime arbitration using normalized linear and angular ratios with rule-based ignoring before sending desired velocities.
Gesture-based intent to compute initial desired velocity in a global reference frame and snap to a nearest principal axis
Receiving gesture data generated with respect to a global reference frame representing desired motion of at least a portion of a vehicle; computing, from the gesture data, an initial desired velocity vector representing desired linear velocity or desired angular velocity; transforming the initial desired velocity vector into a control reference frame; identifying a nearest principal axis direction from principal axis directions parallel to principal axes of the control reference frame by comparing the direction of the initial desired velocity vector; computing a misalignment angle as deviation from the nearest principal axis direction; and upon determining that the misalignment angle is less than or equal to an axis-snapping tolerance value, defining a final desired velocity vector by rotating the initial desired velocity vector such that the final desired velocity vector is parallel to the nearest principal axis direction and sending information indicating the final desired velocity vector to at least the portion of the vehicle.
Rule-based regime detection using normalized linear and angular ratios and ignoring one component
Receiving gesture data generated from movement of an operator with respect to a global reference frame representing desired motion of at least a portion of a vehicle; computing from the gesture data a desired linear velocity vector and a desired angular velocity vector; computing a linear ratio by dividing a first magnitude of the desired linear velocity vector by a linear magnitude threshold and an angular ratio by dividing a second magnitude of the desired angular velocity vector by an angular magnitude threshold; upon determining that the linear ratio or the angular ratio satisfies one or more rules associated with the desired linear velocity vector, ignoring the desired linear velocity vector; upon determining that the linear ratio or the angular ratio satisfies the one or more rules associated with the desired angular velocity vector, ignoring the desired angular velocity vector; and sending data indicating the desired linear velocity vector and the desired angular velocity vector to at least the portion of the vehicle.
Non-transitory computer-readable media for regime detection by normalized ratio rules and ignoring desired velocity components
One or more non-transitory computer-readable media storing instructions that, when executed, perform operations comprising receiving gesture data generated from movement of an operator with respect to a global reference frame representing desired motion of at least a portion of a vehicle; computing from the gesture data a desired linear velocity vector and a desired angular velocity vector; computing a linear ratio by dividing a first magnitude of the desired linear velocity vector by a linear magnitude threshold and an angular ratio by dividing a second magnitude of the desired angular velocity vector by an angular magnitude threshold; upon determining that the linear ratio or the angular ratio satisfies one or more rules associated with the desired linear velocity vector, ignoring the desired linear velocity vector; upon determining that the linear ratio or the angular ratio satisfies the one or more rules associated with the desired angular velocity vector, ignoring the desired angular velocity vector; and sending data indicating the desired linear velocity vector and the desired angular velocity vector to at least the portion of the vehicle.
Across the independent claims, the document covers computing an initial desired velocity vector from gesture data in a global reference frame, transforming it into a control reference frame, and snapping its direction to a nearest principal axis when a misalignment angle is within an axis-snapping tolerance; and detecting an intended regime by computing desired linear and angular velocity vectors, normalizing them via linear and angular ratios relative to magnitude thresholds, applying rule-based checks, and ignoring one or more undesired velocity components before sending the resulting desired velocity data to the vehicle portion.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?