MULTIMODAL TRAJECTORY PREDICTIONS BASED ON GEOMETRIC ANCHORING

    公开(公告)号:US20240246574A1

    公开(公告)日:2024-07-25

    申请号:US18158183

    申请日:2023-01-23

    IPC分类号: B60W60/00

    摘要: Systems and techniques are provided for multimodal trajectory predictions of an object near an autonomous vehicle (AV) based on geometric anchoring. An example process can include identifying one or more predicted trajectories of an object within a proximity of an AV; assigning modes to the one or more predicted trajectories, wherein a respective mode is assigned to each predicted trajectory of the one or more predicted trajectories based on a geometry of the predicted trajectory of the object; determining probabilities of the modes based on one or more parameters, wherein a first mode of the modes is associated with a first probability and a second mode of the modes is associated with a second probability; and updating a planned behavior of the AV based on a combination of the first mode associated with the first probability and the second mode associated with the second probability.

    MAJOR-MINOR INTERSECTION PREDICTION USING TRAFFIC SIGN FEATURES

    公开(公告)号:US20240246573A1

    公开(公告)日:2024-07-25

    申请号:US18158162

    申请日:2023-01-23

    IPC分类号: B60W60/00 B60W30/18

    摘要: The disclosed technology provides solutions for determining, by an autonomous vehicle, whether to yield to a target vehicle at a major-minor intersection based on an estimated trajectory for the target vehicle and road sign data. A method comprising: navigating an autonomous vehicle (AV) along a first roadway, wherein the first roadway intersects with a second roadway; receiving, by the AV, road data indicative of at least one road sign on the first roadway; updating a prediction model based on the received road data; implementing the prediction model to determine an estimated trajectory for a target vehicle on the second roadway; and updating a planned trajectory of the AV based on the estimated trajectory for the target vehicle. Systems and machine-readable media are also provided.

    Hierarchical mode anchoring
    4.
    发明授权

    公开(公告)号:US12091044B2

    公开(公告)日:2024-09-17

    申请号:US17557838

    申请日:2021-12-21

    摘要: The subject disclosure relates to techniques for increasing a quality of predicted trajectories output from a trained prediction algorithm. A process of the disclosed technology can include receiving by a trained prediction algorithm, information about objects in an environment as perceived by sensors of an autonomous vehicle, receiving by the trained prediction algorithm, information about a location of the autonomous vehicle in the environment, generating by the trained prediction algorithm, a predicted trajectory for an object among the objects in the environment, wherein the predicted trajectory being anchored by a path through a tree of paths including at least two modes, wherein each mode creates a node in the tree, wherein a mode is a semantic classification of a portion of the predicted trajectory, and outputting by the trained prediction algorithm, the predicted trajectory for the object.

    HIERARCHICAL MODE ANCHORING
    5.
    发明公开

    公开(公告)号:US20230192128A1

    公开(公告)日:2023-06-22

    申请号:US17557838

    申请日:2021-12-21

    IPC分类号: B60W60/00 G06K9/62

    摘要: The subject disclosure relates to techniques for increasing a quality of predicted trajectories output from a trained prediction algorithm. A process of the disclosed technology can include receiving by a trained prediction algorithm, information about objects in an environment as perceived by sensors of an autonomous vehicle, receiving by the trained prediction algorithm, information about a location of the autonomous vehicle in the environment, generating by the trained prediction algorithm, a predicted trajectory for an object among the objects in the environment, wherein the predicted trajectory being anchored by a path through a tree of paths including at least two modes, wherein each mode creates a node in the tree, wherein a mode is a semantic classification of a portion of the predicted trajectory, and outputting by the trained prediction algorithm, the predicted trajectory for the object.