US 12,216,437 B2
Method, system, and computer program product for autonomous control of dynamical systems
Andrew Gilbert Miller, Pittsburgh, PA (US)
Appl. No. 17/425,509
Filed by Andrew Gilbert Miller, Pittsburgh, PA (US)
PCT Filed Jan. 27, 2020, PCT No. PCT/US2020/015264
§ 371(c)(1), (2) Date Jul. 23, 2021,
PCT Pub. No. WO2020/154743, PCT Pub. Date Jul. 30, 2020.
Claims priority of provisional application 62/796,935, filed on Jan. 25, 2019.
Claims priority of provisional application 62/796,961, filed on Jan. 25, 2019.
Prior Publication US 2022/0146992 A1, May 12, 2022
Int. Cl. G05B 13/02 (2006.01); G05B 15/02 (2006.01)
CPC G05B 13/029 (2013.01) [G05B 15/02 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
one or more hardware processors programmed or configured to:
receive data associated with a host platform;
construct a run-time dynamic envelope based on the data associated with the host platform, wherein, when constructing the run-time dynamic envelope, the one or more hardware processors are programmed and configured to:
construct an initialization dynamic envelope based on a set of actuator commands of the host platform, a response of the host platform to each actuator command of the set of actuator commands of the host platform, and a predetermined response in perception of the host platform to a dynamic action associated with an entity, wherein the initialization dynamic envelope comprises a machine learning model associated with a predicted transformation of perception, wherein the machine learning model associated with a predicted transformation of perception provides an output that comprises the predetermined response in perception of the host platform to a dynamic action associated with the entity, wherein the output is associated with a predicted transformation of perception at all positions of the host platform relative to a first position of the host platform after executing an actuator command, wherein the predicted transformation of perception is based on a notional entity as a reference point that is perceived by the host platform in a field of view of the host platform;
determine data associated with a state of the host platform based on the data associated with the host platform;
construct an idealized dynamic envelope based on the initialization dynamic envelope, the data associated with a state of the host platform, and data associated with a perception envelope; and
construct the run-time dynamic envelope based on the idealized dynamic envelope; and
construct a Worldview Relational Interaction Map (WRIM) based on the run-time dynamic envelope, wherein the WRIM comprises a coordinate system corresponding to an area of an environment of the host platform, wherein the coordinate system comprises a plurality of elements, wherein each element of the plurality of elements comprises:
one or more data attributes associated with a predicted momentum exchange resulting from the host platform co-occupying the element with the entity.