Map-Anchored Object Detection
    131.
    发明申请

    公开(公告)号:US20250029277A1

    公开(公告)日:2025-01-23

    申请号:US18354415

    申请日:2023-07-18

    Inventor: Louis Foucard

    Abstract: An example method includes (a) obtaining sensor data descriptive of an environment of an autonomous vehicle; (b) obtaining a plurality of travel way markers from map data descriptive of the environment; (c) determining, using a machine-learned object detection model and based on the sensor data, an association between one or more travel way markers of the plurality of travel way markers and an object in the environment; and (d) generating, using the machine-learned object detection model, an offset with respect to the one or more travel way markers of a spatial region of the environment associated with the object.

    Systems and methods for training probabilistic object motion prediction models using non-differentiable prior knowledge

    公开(公告)号:US12205004B2

    公开(公告)日:2025-01-21

    申请号:US18495434

    申请日:2023-10-26

    Abstract: The present disclosure provides systems and methods for training probabilistic object motion prediction models using non-differentiable representations of prior knowledge. As one example, object motion prediction models can be used by autonomous vehicles to probabilistically predict the future location(s) of observed objects (e.g., other vehicles, bicyclists, pedestrians, etc.). For example, such models can output a probability distribution that provides a distribution of probabilities for the future location(s) of each object at one or more future times. Aspects of the present disclosure enable these models to be trained using non-differentiable prior knowledge about motion of objects within the autonomous vehicle's environment such as, for example, prior knowledge about lane or road geometry or topology and/or traffic information such as current traffic control states (e.g., traffic light status).

    LIDAR system to adjust doppler effects

    公开(公告)号:US12196854B2

    公开(公告)日:2025-01-14

    申请号:US18099842

    申请日:2023-01-20

    Abstract: Doppler correction of phase-encoded LIDAR includes a code indicating a sequence of phases for a phase-encoded signal, and determining a first Fourier transform of the signal. A laser optical signal is used as a reference and modulated based on the code to produce a transmitted phase-encoded optical signal. A returned optical signal is received in response. The returned optical signal is mixed with the reference. The mixed optical signals are detected to produce an electrical signal. A cross spectrum is determined between in-phase and quadrature components of the electrical signal. A Doppler shift is based on a peak in the cross spectrum. A device is operated based on the Doppler shift. Sometimes a second Fourier transform of the electrical signal and the Doppler frequency shift produce a corrected Fourier transform and then a cross correlation. A range is determined based on a peak in the cross correlation.

    Dual mode map for autonomous vehicle

    公开(公告)号:US12195038B1

    公开(公告)日:2025-01-14

    申请号:US18396363

    申请日:2023-12-26

    Abstract: An autonomous vehicle control system and method may utilize a dual mode map including sparse map data for some portions of an environment that lacks some of the data maintained in dense map data for other portions of the environment. Sparse map data may be used, for instance, to address a recently-established construction area on a roadway that is incompatible with dense map data that was previously used to operate on the roadway, enabling operation of an autonomous vehicle in the construction area to proceed even in the absence of dense map data for the construction area, e.g., by dynamically augmenting the sparse map data to incorporate additional data sensed by a perception system of the autonomous vehicle.

    Systems and methods for compressing and storing sensor data collected by an autonomous vehicle

    公开(公告)号:US12191888B2

    公开(公告)日:2025-01-07

    申请号:US17139440

    申请日:2020-12-31

    Abstract: A vehicle computing system onboard an autonomous vehicle can include one or more processors and one or more non-transitory computer-readable media that store instructions that, when executed by the one or more processors, cause the computing system to perform operations. The operations can include obtaining sensor data from one or more sensors of the autonomous vehicle; applying lossy compression to the sensor data to generate compressed sensor data; storing data describing the compressed sensor data; decompressing the compressed sensor data to generate decompressed sensor data; and inputting data describing the decompressed sensor data into an autonomy system comprising one or more machine-learned models. The autonomy system can be configured to control operations of the autonomous vehicle based on the decompressed sensor data.

    Detection or Correction For Multipath Reflection

    公开(公告)号:US20240418823A1

    公开(公告)日:2024-12-19

    申请号:US18799820

    申请日:2024-08-09

    Abstract: Ranging and detection data is processed to identify or correct for multipath reflection. A sensor point that represents a location of an object, the location based on an incidence of an electromagnetic wave received at a sensor is obtained. The first sensor point is determined to be a product of multipath reflection. A first point of reflection on a surface of a surface model is determined. The location of the first sensor point is corrected based on the first point of reflection on the surface of the surface model.

    Optical Coupler for LIDAR Sensor
    138.
    发明申请

    公开(公告)号:US20240410989A1

    公开(公告)日:2024-12-12

    申请号:US18411106

    申请日:2024-01-12

    Abstract: A LIDAR device for a vehicle includes an integrated chip. The integrated chip includes a substrate layer, a cladding layer, a waveguide, a scattering array, and a reflector layer. The cladding layer is disposed on the substrate layer to form an interface with the substrate layer. The waveguide is disposed within the cladding layer and configured to route an infrared optical field. The scattering array is disposed within the cladding layer between the waveguide and the interface and perturbs the infrared optical field and scatters the infrared optical field into a first beam propagating toward a surface of the cladding layer and into a second beam propagating towards the interface. The reflector layer is disposed within the cladding layer between the waveguide and the surface of the cladding layer to reflect the first beam towards the interface.

    Light Detection and Ranging (LIDAR) Module for a LIDAR System

    公开(公告)号:US20240393466A1

    公开(公告)日:2024-11-28

    申请号:US18191621

    申请日:2023-03-28

    Abstract: A LIDAR system includes a substrate and an emitter coupled to the substrate and configured to emit a light beam along a first axis of the substrate. The LIDAR system includes an optic device coupled to the substrate and configured to split the light beam into a plurality of light beams. The LIDAR system includes an optical amplifier array coupled to the substrate and configured to amplify the plurality of light beams received from the optic device to generate a plurality of amplified light beams. The LIDAR system includes a transceiver coupled to the substrate and configured to redirect the plurality of amplified light beams from traveling along the first axis of the substrate to traveling along a second axis of the substrate that is different from the first axis.

    Systems and methods for simulating dynamic objects based on real world data

    公开(公告)号:US12141995B2

    公开(公告)日:2024-11-12

    申请号:US17388372

    申请日:2021-07-29

    Abstract: Systems and methods for generating simulation data based on real-world dynamic objects are provided. A method includes obtaining two- and three-dimensional data descriptive of a dynamic object in the real world. The two- and three-dimensional information can be provided as an input to a machine-learned model to receive object model parameters descriptive of a pose and shape modification with respect to a three-dimensional template object model. The parameters can represent a three-dimensional dynamic object model indicative of an object pose and an object shape for the dynamic object. The method can be repeated on sequential two- and three-dimensional information to generate a sequence of object model parameters over time. Portions of a sequence of parameters can be stored as simulation data descriptive of a simulated trajectory of a unique dynamic object. The parameters can be evaluated by an objective function to refine the parameters and train the machine-learned model.

Patent Agency Ranking