Abstract: An autonomous vehicle control system includes one or more processors. The one or more processors are configured to cause a transmitter to transmit a transmit signal from a laser source. The one or more processors are configured to cause a receiver to receive a return signal reflected by an object. The one or more processors are configured to cause one or more optics to generate a first polarized signal of the return signal with a first polarization, and generate a second polarized signal of the return signal with a second polarization. The one or more processors are configured to calculate a value of reflectivity based on a signal-to-noise ratio (SNR) value of the first polarized signal and an SNR value of the second polarized signal. The one or more processors are configured to operate a vehicle based on the value of reflectivity.

Abstract: A concurrent simulation of multiple sensor modalities includes identifying the multiple sensor modalities in association with a simulation scenario, determining a timeline interleaving a publishing or operating frequency of each of the multiple sensor modalities relative to each other, determining a current time interval of a sliding window in the timeline, determining a simulation segment of the simulation scenario using the current time interval of the sliding window, rendering a scene based on the simulation segment, executing a simulation to concurrently simulate the multiple sensor modalities using ray tracing in the rendered scene, and generating simulated sensor data of the multiple sensor modalities based on executing the simulation.

Abstract: A LIDAR system includes a laser source, a first scanner, and a second scanner. The first scanner receives a first beam from the laser source and applies a first angle modulation to the first beam to output a second beam at a first angle. The second scanner receives the second beam and applies a second angle modulation to the second beam to output a third beam at a second angle.

Abstract: A structure of a silicon photonics device for LIDAR includes a first insulating structure and a second insulating structure disposed above one or more etched silicon structures overlying a substrate member. A metal layer is disposed above the first insulating structure without a prior deposition of a diffusion barrier and adhesion layer. A thin insulating structure is disposed above the second insulating structure. A first configuration of the metal layer, the first insulating structure and the one or more etched silicon structures forms a free-space coupler. A second configuration of the thin insulating structure above the second insulating structure forms an edge coupler.

Abstract: One or more instances of sensor data collected using an autonomous vehicle sensor suite can be labeled using corresponding instance(s) of sensor data collected using an additional sensor suite. The additional vehicle can include a second autonomous vehicle as well as a non-autonomous vehicle mounted with a removable hardware pod. In many implementations, an object in the sensor data captured using the autonomous vehicle can be labeled by mapping a label corresponding to the same object captured using the additional vehicle. In various implementations, labeled instances of sensor data can be utilized to train a machine learning model to generate one or more control signals for autonomous vehicle control.

Abstract: A method may include obtaining lidar data comprising a plurality of lidar returns from an environment of an autonomous vehicle. The lidar data may be processed with a machine learning model to generate, for the plurality of lidar returns, a plurality of first outputs that each identify a respective lidar return as belonging to an object or non-object and a plurality of second outputs that identify lidar returns belonging to objects as harmful or non-harmful to the autonomous vehicle. A subset of the lidar returns identified as belonging to objects that (i) do not correspond to any of a plurality of pre-classified objects and (ii) were identified as harmful to the autonomous vehicle may be determined. The autonomous vehicle may be controlled based at least in part on the subset of lidar returns.

Abstract: A teleoperations system may be used to modify elements in the mapping data used by an autonomous vehicle to cause the autonomous vehicle to control its trajectory based on the modified elements. In addition, in some instances, a teleoperations system may be used to generate virtual paths of travel for an autonomous vehicle based upon teleoperations system virtual path suggestion inputs.

Abstract: A LIDAR system for a vehicle is provided. The LIDAR system includes a lid defining an internal volume. The LIDAR system includes one or more circuit modules disposed within the internal volume. The LIDAR system includes a cold plate including a first side coupled to the lid to enclose the one or more circuit modules within the internal volume. The cold plate further includes a second side that is different from the first side and defines a fluid channel through which a liquid coolant flows. The LIDAR system includes a cover coupled to the cold plate to cover the fluid channel defined in the second side of the cold plate.

Abstract: Example aspects of the present disclosure relate to an example computer-implemented method for predicting the intent of actors within an environment. The example method includes obtaining state data associated with a plurality of actors within the environment and map data indicating a plurality of lanes of the environment. The method include determining a plurality of potential goals each actor based on the state data and the map data. The method includes processing the state data, the map data, and the plurality of potential goals with a machine-learned forecasting model to determine (i) a forecasted goal for a respective actor of the plurality of actors, (ii) a forecasted interaction between the respective actor and a different actor of the plurality of actors based on the forecasted goal, and (iii) a continuous trajectory for the respective actor based on the forecasted goal.

Abstract: A light detection and ranging (LIDAR) sensor system includes a window, a laser source configured to generate a beam, one or more scanning optics configured to output the beam through the window, and a sensor. The sensor includes at least one light emitter configured to output light associated with a parameter different than a corresponding parameter of the beam, a first optic coupled to the window, the first optic configured to receive the light from the light emitter and provide the light into the window to undergo total internal reflection in the window, a second optic coupled to the window, the second optic configured to receive the light provided into the window by the first optic, and a detector configured to receive the light from the second optic and output a signal indicative of a presence of an obscurant on the window.

Abstract: A light detection and ranging (LIDAR) system for a vehicle includes a transmitter, a receiver, one or more scanning optics, and a circulator. The transmitter is configured to output a transmit beam. The receiver includes a first receive grating coupler and a second receive grating coupler. The circulator is configured to receive the transmit beam and provide the transmit beam to the one or more scanning optics, receive a return beam from reflection of the transmit beam by an object, split the return beam into at least a first component and a second component, and direct the first component to the first receive grating coupler and the second component to the second receive grating coupler.

Abstract: Example methods for multistage autonomous vehicle motion planning include obtaining sensor data descriptive of an environment of the autonomous vehicle; identifying one or more objects in the environment based on the sensor data; generating a plurality of candidate strategies, wherein each candidate strategy of the plurality of candidate strategies comprises a set of discrete decisions respecting the one or more objects, wherein generating the plurality of candidate strategies includes: determining that at least two strategies satisfy an equivalence criterion, such that the plurality of candidate strategies include at least one candidate strategy corresponding to an equivalence class representative of a plurality of different strategies that are based on different discrete decisions; determining candidate trajectories respectively for the plurality of candidate strategies; and initiating control of the autonomous vehicle based on a selected candidate trajectory.

Abstract: Techniques are disclosed for training one or more cost functions of an autonomous vehicle (“AV”) control system based on difference between data generated using the AV control system and manual driving data. In many implementations, manual driving data captures action(s) of a vehicle controlled by a manual driver. Additionally or alternatively, multiple AV control systems can be evaluated by comparing deviations for each AV control system, where the deviations are determined using the same set of manual driving data.

Abstract: Sensor data collected from an autonomous vehicle can be labeled using sensor data collected from an additional vehicle. Labeled sensor data can generate targeted testing instances for a trained machine learning model, where the trained machine learning model is used in generating control signals for an autonomous vehicle. In many implementations, targeted training instances can generate an accuracy value for the trained neural network model. Additionally or alternatively, the sensor suite on the additional vehicle can include a removable hardware pod which can be mounted on a variety of vehicles.

Abstract: A vehicle radar system utilizes multiple radar sensors having overlapping fields of view to effectively synthesize a distributed radar antenna array aperture from the outputs of the multiple radar sensors and effectively enhance one or more of angular resolution, detection range and signal to noise ratio beyond that supported by any of the radar sensors individually.

Abstract: A modular LIDAR system comprising a seed laser configured to output a beam, a modular modulator coupled to receive the beam output the seed laser and modulate the beam to create a modulated beam, a modular amplifier coupled to receive the modulated beam from the modular modulator and generate an amplified beam, and a modular transceiver chip coupled to the modular modulator and the modular amplifier, the transceiver chip configured to emit the beam perpendicularly from a first surface of the transceiver chip through an optical window; and receive a reflected beam from a target through the optical window.

Abstract: Logged data from an autonomous vehicle is processed to generate augmented data. The augmented data describes an actor in an environment of the autonomous vehicle, the actor having an associated actor type and an actor motion behavior characteristic. The augmented data may be varied to create different sets of augmented data. The sets of augmented data can be used to create one or more simulation scenarios that in turn are used to produce machine learning models to control the operation of autonomous vehicles.

Abstract: A method and system for synthetic data generation and analysis includes generating a synthetic dataset. A set of parameters is determined and scenarios are generated from the parameters that represent three-dimensional scenes. Synthetic images are rendered for the scenarios. A synthetic dataset may be formed to have a controlled variation in attributes of synthetic images over a synthetic dataset. The synthetic dataset may be used for training or evaluating a machine learning model.

Abstract: A light detection and ranging (lidar) system may include a transceiver, a first device including a laser source configured to generate a beam, and one or more optical components, a second device including one or more analog-to-digital converters (ADCs), and a processor configured to alternately turn on the first device and turn on the transceiver. The first device may be configured to generate, based on the beam, an optical signal associated with a local oscillator (LO) signal. The transceiver may be configured to transmit the optical signal to an environment, in response to transmitting the optical signal, receive a returned optical signal that is reflected from an object in the environment, and pair the returned optical signal with the LO signal to generate an electrical signal. The second device may be configured to generate, based on the electrical signal, a digital signal.

Abstract: Systems and methods for using transmission sensor(s) in localization of an autonomous vehicle (“AV”) are described herein. Some implementations receive instance(s) of transmission sensor data generated by transmission sensor(s) of the AV, generate pose instance(s) of a pose of the AV based on the instance(s) of the transmission sensor data, and cause the AV to be controlled based on the pose instance(s). In some of those implementations, the pose instance(s) can be generated based on steering data the AV that indicates steering angle(s) of the autonomous, and that temporally correspond to the instance(s) of the transmission sensor data. In various implementations, the pose instance(s) may only be generated or utilized based on the instance(s) of the transmission sensor data (and optionally the steering data) in response to detecting an adverse event at the AV.