公开(公告)号：US20250060220A1

公开(公告)日：2025-02-20

申请号：US18450003

申请日：2023-08-15

Applicant: Oracle International Corporation

Inventor： Matthew Charles Rowe ,  Sahil Malhotra ,  Sergio Aldea Lopez ,  Oleg Gennadievich Shevelev ,  Alberto Polleri

IPC: G01C21/34

Abstract: Techniques for deriving an optimal traversal path on a racetrack are disclosed. The system partitions a track into straight and curved segments. The system identifies optimal traversals through each segment from historical traversal data. The system stitches the optimal traversals together and smooths the optimal traversals at the transition points between track segments. The system verifies that the smoothed traversals meet one or more kinematic criteria before outputting the optimal traversal path.

公开(公告)号：US20250124581A1

公开(公告)日：2025-04-17

申请号：US18380095

申请日：2023-10-13

Applicant: Oracle International Corporation

Inventor： Sergio Aldea Lopez ,  Sahil Malhotra ,  Matthew Charles Rowe ,  Oleg Gennadievich Shevelev ,  Alberto Polleri

IPC: G06T7/20 ,  G06T7/13 ,  G06T7/70 ,  G06V10/24 ,  G06V10/74

Abstract: Techniques for determining an absolute longitudinal position of a moving object on non-linear sections of a trajectory are described. In one technique, an estimated track boundary segment is generated based on a digital image associated with a moving object. For each position of multiple positions in an actual track boundary segment pertaining to a track for one or more moving objects, an alignment of the estimated track boundary segment with the actual track boundary segment is made based on that position. Also, based on the alignment, a difference measurement between the estimated track boundary segment and a portion of the actual track boundary segment is generated. After each of the positions is considered, a particular alignment, of multiple alignments, that is associated with the lowest difference measurement among the multiple positions is selected. Based on the particular alignment, a longitudinal value of the moving object is determined.

公开(公告)号：US12301388B2

公开(公告)日：2025-05-13

申请号：US18339065

申请日：2023-06-21

Applicant: Oracle International Corporation

Inventor： Matthew Charles Rowe ,  Sahil Malhotra ,  Sergio Aldea Lopez ,  Oleg Gennadievich Shevelev ,  Alberto Polleri

IPC: H04L25/03

Abstract: Techniques for smoothing a signal are disclosed. The system partitions the portion of the data sequence into a stable subsequence and an unstable subsequence of data points. The system applies a rate of change exhibited by the stable subsequence to the unstable subsequence to create a smoothed, more stable subsequence.

公开(公告)号：US20250086810A1

公开(公告)日：2025-03-13

申请号：US18367415

申请日：2023-09-12

Applicant: Oracle International Corporation

Inventor： Oleg Gennadievich Shevelev ,  Sahil Malhotra ,  Sergio Aldea Lopez ,  Matthew Charles Rowe ,  Alberto Polleri

IPC: G06T7/292 ,  G06F17/12

Abstract: Techniques for preparing data for high-precision absolute localization of a moving object along a trajectory are provided. In one technique, a sliding window of a set of adjacent points along a trajectory of a moving object is identified, along with a midpoint in the sliding window. Based on the set of adjacent points, a first polynomial equation is generated for a first dimension and a second polynomial equation is generated for a second dimension. A first derivative at a particular timestamp associated with the midpoint is a first velocity along the first dimension, while a particular first derivative at the particular timestamp is a second velocity along the second dimension. A velocity in direction of yaw is generated based on the first velocity, the second velocity, and a slip angle associated with the midpoint. A yaw angle is generated based on the velocity in direction of yaw.

公开(公告)号：US20250085434A1

公开(公告)日：2025-03-13

申请号：US18367400

申请日：2023-09-12

Applicant: Oracle International Corporation

Inventor： Oleg Gennadievich Shevelev ,  Sahil Malhotra ,  Sergio Aldea Lopez ,  Matthew Charles Rowe ,  Alberto Polleri

IPC: G01S19/13

Abstract: Techniques for preparing data for high-precision absolute localization of a moving object along a trajectory are provided. In one technique, a sequence of points is stored, where each point corresponds to a different set of Cartesian coordinates. A curve is generated that approximates a line that passes through the sequence of points. Based on the curve, a set of points is generated on the curve, where the set of points is different than the sequence of points. New Cartesian coordinates are generated for each point in the set of points. After generating the new Cartesian coordinates, Cartesian coordinates of a position of a moving object are determined.

公开(公告)号：US20240430139A1

公开(公告)日：2024-12-26

申请号：US18339065

申请日：2023-06-21

Applicant: Oracle International Corporation

Inventor： Matthew Charles Rowe ,  Sahil Malhotra ,  Sergio Aldea Lopez ,  Oleg Gennadievich Shevelev ,  Alberto Polleri

IPC: H04L25/03

Abstract: Techniques for smoothing a signal are disclosed. The system partitions the portion of the data sequence into a stable subsequence and an unstable subsequence of data points. The system applies a rate of change exhibited by the stable subsequence to the unstable subsequence to create a smoothed, more stable subsequence.

公开(公告)号：US20240420343A1

公开(公告)日：2024-12-19

申请号：US18210506

申请日：2023-06-15

Applicant: Oracle International Corporation

Inventor： Oleg Gennadievich Shevelev ,  Sahil Malhotra ,  Sergio Aldea Lopez ,  Matthew Charles Rowe ,  Alberto Polleri

IPC: G06T7/246 ,  G06T7/70

Abstract: Techniques for generating high-precision localization of a moving object on a trajectory are provided. In one technique, a particular image that is associated with a moving object is identified. A set of candidate images is selected from a plurality of images that were used to train a neural network. For each candidate image in the set of candidate images: (1) output from the neural network is generated based on inputting the particular image and said each candidate image to the neural network; (2) a predicted position of the particular image is determined based on the output and a position that is associated with said each candidate image; and (3) the predicted position is added to a set of predicted positions. The set of predicted positions is aggregated to generate an aggregated position for the particular image.