Abstract: Aspects of the disclosure relate to adjusting a virtual camera's orientation when a vehicle is making a turn. One or more computing devices may receive the vehicle's original heading prior to making the turn and the vehicle's current heading. Based on the vehicle's original heading and the vehicle's current heading, the one or more computing devices may determine an angle of a turn the vehicle is performing and The one or more computing devices may determine a camera rotation angle and adjust the virtual camera's orientation relative to the vehicle to an updated orientation by rotating the virtual camera by the camera rotation angle and generate a video corresponding to the virtual camera's updated orientation. The video may be displayed on the display by the one or more computing devices.

Abstract: Methods and devices for detecting traffic signals and their associated states are disclosed. In one embodiment, an example method includes a scanning a target area using one or more sensors of a vehicle to obtain target area information. The vehicle may be configured to operate in an autonomous mode, and the target area may be a type of area where traffic signals are typically located. The method may also include detecting a traffic signal in the target area information, determining a location of the traffic signal, and determining a state of the traffic signal. Also, a confidence in the traffic signal may be determined. For example, the location of the traffic signal may be compared to known locations of traffic signals. Based on the state of the traffic signal and the confidence in the traffic signal, the vehicle may be controlled in the autonomous mode.

Abstract: In one example, a method is provided that includes receiving lidar data obtained by a lidar device. The lidar data includes a plurality of data points indicative of locations of reflections from an environment of the vehicle. The method includes receiving images of portions of the environment captured by a camera at different times. The method also includes determining locations in the images that correspond to a data point of the plurality of data points. Additionally, the method includes determining feature descriptors for the locations of the images and comparing the feature descriptors to determine that sensor data associated with at least one of the lidar device, the camera, or a pose sensor is accurate or inaccurate.

Abstract: One example system comprises a LIDAR sensor that rotates about an axis to scan an environment of the LIDAR sensor. The system also comprises one or more cameras that detect external light originating from one or more external light sources. The one or more cameras together provide a plurality of rows of sensing elements. The rows of sensing elements are aligned with the axis of rotation of the LIDAR sensor. The system also comprises a controller that operates the one or more cameras to obtain a sequence of image pixel rows. A first image pixel row in the sequence is indicative of external light detected by a first row of sensing elements during a first exposure time period. A second image pixel row in the sequence is indicative of external light detected by a second row of sensing elements during a second exposure time period.

Abstract: Aspects of the disclosure relate to evaluating quality of a predetermined pullover location for an autonomous vehicle. For instance, a plurality of inputs for the predetermined pullover location may be received. The plurality of inputs may each include a value representative of a characteristic of the predetermined pullover location. The plurality of inputs may be combined to determine a pullover quality value for the predetermined pullover location. The pullover quality value may be provided to a vehicle in order to enable the vehicle to select a pullover location for the vehicle.

Abstract: Example embodiments relate to user interface techniques for recommending remote assistance actions. A remote computing device may display a representation of the forward path for an autonomous vehicle based on sensor data received from the vehicle. The computing device may augment the representation of the forward path to further depict one or more proposed trajectories available for the autonomous vehicle to perform. Each proposed trajectory conveys one or more maneuvers positioned relative to road boundaries in the forward path. The computing device may receive a selection of a proposed trajectory from the one or more proposed trajectories available for the autonomous vehicle to perform and provide navigation instructions to the vehicle based on the proposed trajectory.

Abstract: The present disclosure relates to systems and devices having a rotatable mirror assembly. An example system includes a housing and a rotatable mirror assembly. The rotatable mirror assembly includes a plurality of reflective surfaces, a shaft defining a rotational axis, and a mirror body coupling the plurality of reflective surfaces to the shaft. The mirror body includes a plurality of flexible support members. The rotatable mirror assembly also includes a coupling bracket configured to removably couple the rotatable mirror assembly to the housing. The system also includes a transmitter configured to emit emission light into an environment of the system after interacting with at least one reflective surface of the plurality of reflective surfaces. The system additionally includes a receiver configured to detect return light from the environment after interacting with the at least one reflective surface of the plurality of reflective surfaces.

Abstract: One example method involves rotating a housing of a light detection and ranging (LIDAR) device about a first axis. The housing includes a first optical window and a second optical window. The method also involves transmitting a first plurality of light pulses through the first optical window to obtain a first scan of a field-of-view (FOV) of the LIDAR device. The method also involves transmitting a second plurality of light pulses through the second optical window to obtain a second scan of the FOV. The method also involves identifying, based on the first scan and the second scan, a portion of the FOV that is at least partially occluded by an occlusion.

Abstract: Described examples relate to an apparatus comprising a first sensor configured to scan an area of interest during a first time period and a second sensor configured to capture a plurality of images of a field of view. The apparatus may include at least one controller configured to receive the plurality of images captured by the second sensor, compare the timestamp information associated with at least one image of the plurality of images to at least one time period of the first time period, and select a base image from the plurality of images based on the comparison.

Abstract: Methods, computer systems, and apparatus, including computer programs encoded on computer storage media, for performing object detection. The system obtains a respective range image corresponding to each point cloud in a set of point clouds captured by one or more sensors. The system processes each range image using a segmentation neural network to generate range image features and a segmentation output. The system generates a feature representation of the set of point clouds from only the feature representations of the foreground points. The system processes the feature representation of the set of point clouds using a prediction neural network to generate a prediction characterizing the set of point clouds.

Abstract: Aspects of the disclosure relate generally to generating and providing route options for an autonomous vehicle. For example, a user may identify a destination, and in response the vehicle's computer may provide routing options to the user. The routing options may be based on typical navigating considerations such as the total travel time, travel distance, fuel economy, etc. Each routing option may include not only an estimated total time, but also information regarding whether and which portions of the route may be maneuvered under the control of the vehicle alone (fully autonomous), a combination of the vehicle and the driver (semiautonomous), or the driver alone. The time of the longest stretch of driving associated with the autonomous mode as well as map information indicating portions of the routes associated with the type of maneuvering control may also be provided.

Abstract: Aspects of the disclosure provide for generation of trajectories for a vehicle driving in an autonomous driving mode. In one instance, a default number of trajectories to be generated may be identified. A set of maneuvering options may be selected from a set of predetermined maneuvering options based on the number of trajectories. The set of maneuvering options may be filtered based on the default number of trajectories. A set of trajectories may be generated based on the filtered set of maneuvering option such that each trajectory of the set corresponds to a different maneuvering behavior. A cost for each trajectory of the set of trajectories may be determined, and one of the trajectories of the set of trajectories may be selected based on the determined costs. The vehicle may be maneuvered in the autonomous driving mode according to the selected one of the trajectories.

Abstract: One example method of testing an electrical device comprises transmitting a data pattern to a memory device of the electrical device by a controller of the electrical device to provide a written data pattern to the memory device, wherein the data pattern replicates a resonant frequency of at least a portion of the electrical device, reading the written data pattern from the memory device with the controller, and comparing the written data pattern to the data pattern.

Abstract: The present radio system transmits an electromagnetic signal to nearby devices requesting the device respond. The radio system also receives responses to the electromagnetic signal from the nearby devices. Based on the radio technology used, the signaling of the transmitted electromagnetic signal may be varied. For example, the transmitted electromagnetic signal may be a Bluetooth, 802.11, or other radio signal. A device that received the signal from the radio unit may transmit a response signal with the same radio technology. However, in some instances, the radio technology used for communication may operate on several radio (e.g., frequency) channels. Both the transmitter and receiver must operate on the same channel at the same time in order to communicate. Thus, it may be desirable to transmit the electromagnetic signal on more than one channel at the same time, in order to increase the chances that a nearby device responds.

Abstract: Aspects and implementations of the present disclosure relate to performance and safety improvements for autonomous trucking systems, including techniques of obtaining an identification of an unfavorable condition on a route of an autonomous vehicle (AV), causing the AV to exit the route, and performing one or more waiting loops until the unfavorable condition is resolved, the AV is rerouted, assistance arrives, and the like.

Abstract: The technology relates to planning trajectories for self-driving vehicles in order to transport passengers or cargo from a pickup location to a destination. Trajectory planning includes generating a speed plan for an upcoming portion of the trip in view of one or more constraints. The constraints may be due to proximity to an adjacent vehicle or other road user, and can include projected overlaps between the vehicle and other objects in the vehicle's nearby environment. Certain constraints may be binary or otherwise discontinuous in nature, in which a condition either exists at a given point in time or it does not. Noise in sensor data or prediction models may trigger such binary conditions, which in turn may cause the vehicle to alter the speed plan.

Abstract: Example embodiments relate to radar image video compression techniques using per-pixel Doppler measurements, which can involve initially receiving radar data from a radar unit to generate a radar representation that represents surfaces in the environment. Based on Doppler scores in the radar representation, a range rate can be determined for each pixel that indicates a radial direction motion for a surface represented by the pixel. The range rates and backscatter values can then be used to estimate a radar representation prediction for subsequent radar data received from the radar unit, which enables a generation of a compressed radar data file that represents the difference between the radar representation prediction and the actual representation determined for the subsequent radar data. The compressed radar data file can be stored in memory, transmitted to other devices, and decompressed and used to train models via machine learning.

Abstract: Aspects of the disclosure provide a method of generating and following planned trajectories for an autonomous vehicle. For instance, a baseline for a planned trajectory that the autonomous vehicle can use to follow a route to a destination may be determined. A stopping point corresponding to a traffic control that will cause the autonomous vehicle to come to a stop using the baseline may be determined. Sensor data identifying objects and their locations may be received. A plurality of constraints may be generated based on the sensor data. A planned trajectory may be generated using the baseline, the stopping point, and the plurality of constraints, wherein constraints beyond the stopping point are ignored.

Abstract: An example method includes receiving, from one or more sensors associated with an autonomous vehicle, sensor data associated with a target object in an environment of the vehicle during a first environmental condition, where at least one sensor of the sensor(s) is configurable to be associated with one of a plurality of operating field of view volumes. The method also includes based on the sensor data, determining at least one parameter associated with the target object. The method also includes determining a degradation in the parameter(s) between the sensor data and past sensor data, where the past sensor data is associated with the target object in the environment during a second environmental condition different from the first and, based on the degradation, adjusting the operating field of view volume of the at least one sensor to a different one of the operating field of view volumes.