Abstract: Example embodiments relate to techniques for three dimensional (3D) object detection and localization. A computing system may cause a radar unit to transmit radar signals and receive radar reflections relative to an environment of a vehicle. Based on the radar reflections, the computing system may determine a heading and a range for a nearby object. The computing system may also receive an image depicting a portion of the environment that includes the object from a vehicle camera and remove peripheral areas of the image to generate an image patch that focuses upon the object based on the heading and the range for the object. The image patch and the heading and the range for the object can be provided as inputs into a neural network that provides output parameters corresponding to the object, which can be used to control the vehicle.

Abstract: Aspects of the disclosure relate to controlling a transition between a manual driving mode and an autonomous driving mode of a vehicle. For instance, one or more processors of one or more control computing devices may control the vehicle in the autonomous driving mode. While controlling the vehicle in the autonomous driving mode and decelerating at a given rate, the processors may receive at a user input of the vehicle input requesting a transition from the autonomous driving mode to the manual driving mode. In response to the input, the processors may transition the vehicle to the manual driving mode. After transitioning the vehicle to the manual driving mode, the processors may send deceleration signals to a deceleration actuator thereby causing the vehicle to continue to decelerate at the given rate.

Abstract: Example embodiments described herein involve techniques for removing transmit phase noise. A system may cause a printed circuit board (PCB) to supply a signal enabling the transmission line to couple the signal to the radar unit and the delay line to couple the signal to the quadrature coupler. The radar unit may use the signal to transmit a radar signal on a radio channel having a centered radio frequency while the quadrature coupler uses the signal to produce an output from the quadrature coupler. The system may estimate phase noise relative to the radio channel having the centered RF based on the output from the quadrature coupler, receive, from the radar unit, a radar reflection corresponding to the radar signal, and determine information representative of one or more objects in an environment based on the radar reflection and the phase noise.

Abstract: The technology relates to facilitating transportation services between a user and a vehicle having an autonomous driving mode. For instance, one or more server computing devices having one or more processors may information identifying the current location of the vehicle. The one or more server computing devices may determine that the user is likely to want to take a trip to a particular destination based on prior location history for the user. The one or more server computing devices may dispatch the vehicle to cause the vehicle to travel in the autonomous driving mode towards a location of the user. In addition, after dispatching, the one or more server computing devices sending a notification to a client computing device associated with the user indicating that the vehicle is currently available to take the passenger to the particular destination.

Abstract: Example embodiments relate to gradually adjusting a vehicle sensor perspective using remote assistance. A computing device may receive a request for assistance from a vehicle operating in an environment. The request indicates that the vehicle is stopped at a location with a sensor perspective of the environment that is at least partially occluded. Responsive to receiving the request for assistance, the computing device may display a graphical user interface (GUI) that represents a current state of the vehicle and includes a selectable option configured to enable the vehicle to gradually move forward a predefined distance. The computing device may detect a selection of the selectable option and transmit instructions that enable the vehicle to gradually move forward the predefined distance. The vehicle can then gradually move forward the predefined distance while also monitoring for one or more changes in the environment responsive to receiving the instructions.

Abstract: Models can be generated of a vehicle's view of its environment and used to maneuver the vehicle. This view need not include what objects or features the vehicle is actually seeing, but rather those areas that the vehicle is able to observe using its sensors if the sensors were completely un-occluded. For example, for each of a plurality of sensors of the object detection component, a computer may generate an individual 3D model of that sensor's field of view. Weather information is received and used to adjust one or more of the models. After this adjusting, the models may be aggregated into a comprehensive 3D model. The comprehensive model may be combined with detailed map information indicating the probability of detecting objects at different locations. The model of the vehicle's environment may be computed based on the combined comprehensive 3D model and detailed map information.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for predicting occupancies of agents. One of the methods includes obtaining scene data characterizing a current scene in an environment; and processing a neural network input comprising the scene data using a neural network to generate a neural network output, wherein: the neural network output comprises respective occupancy outputs corresponding to a plurality of agent types at one or more future time points; the occupancy output for each agent type at a first future time point comprises respective occupancy probabilities for a plurality of locations in the environment; and in the occupancy output for each agent type at the first future time point, the respective occupancy probability for each location characterizes a likelihood that an agent of the agent type will occupy the location at the first future time point.

Abstract: Systems and methods described herein relate to the manufacture of optical elements and optical systems. An example method includes providing a first substrate that has a plurality of light-emitter devices disposed on a first surface. The method includes providing a second substrate that has a mounting surface defining a reference plane. The method includes forming a structure and an optical spacer on the mounting surface of the second substrate. The method additionally includes coupling the first and second substrates together such that the first surface of the first substrate faces the mounting surface of the second substrate at an angle with respect to the reference plane.

Abstract: Aspects of the disclosure relate to pre-computing routes for autonomous vehicles using map shards. For example, a shard from a plurality of shards of a map may be selected. Each shard including a plurality of nodes and edges connecting pairs of nodes of the plurality of nodes, and each node of the plurality represents a location. A plurality of port nodes for the shard are identified. Each port node has an edge that enters into the selected shard or exists the selected shard. For each port node of the plurality having an edge that enters into the selected shard, optimal routes to each other port node of the plurality having an edge that exits the selected shard may be determined. The optimal routes for the selected shard may be sent to the autonomous vehicles in order to enable the autonomous vehicles to use the optimal routes to determine routes.

Abstract: One example system for detecting an emergency siren comprises a microphone device, a band pass filter operably coupled to the microphone device and configured to filter sound data from the microphone device to produce filtered sound data, a phase lock loop circuit configured to phase filter the filtered sound data to produce phase filtered data, and a processor configured to determine a presence of a siren signal from the phase filtered data.

Abstract: A sensor module comprising a housing defining an internal cavity, the housing including an aperture, at least one microphone positioned in the internal cavity spaced from the aperture, a first barrier proximate the aperture, and a second barrier positioned between the at least one microphone and the first barrier.

Abstract: Aspects of the disclosure relate to detecting an emergency vehicle. For instance, a plurality of images may be taken from a perspective of an autonomous vehicle. One or more gates representing a region of interest at a respective distance from the vehicle may be generated for the images. A plurality of lights may be detected within the one or more gates. A first candidate emergency vehicle may be identified from a detected plurality of lights in one or more gates of one of the images, and a second candidate emergency vehicle may be identified from a detected plurality of lights in one or more gates of another of the images. The first and second candidate emergency vehicles are determined to be the same emergency vehicle and to be active. An operational system of the autonomous vehicle is controlled based on the determination that the given emergency vehicle is active.

Abstract: A method for preparing an optically transparent, superomniphobic glass composition is described. In one aspect, the present disclosure provides a method for preparing a glass composition, including heating a borosilicate glass comprising 45-85 wt. % silicon oxide and 10-40 wt. % boron oxide to form a phase-separated glass comprising an interpenetrating network of silicon oxide domains and boron oxide domains. The method includes removing at least a portion of the boron oxide domains from the phase-separated glass and depositing a hydrophobic silane to provide a porous glass having a hydrophobic silane layer disposed on a portion of the surface thereof, a total pore volume of 15-50 vol. %, and an average pore diameter of 20-300 nm. The method includes, within at least a portion of the volume of the porous glass, forming an aerogel precursor, and converting at least a portion of the aerogel precursor to an aerogel.

Abstract: A light detection and ranging (LIDAR) device scans through a scanning zone while emitting light pulses and receives reflected signals corresponding to the light pulses. The LIDAR device scans the emitted light pulses through the scanning zone by reflecting the light pulses from an array of oscillating mirrors. The mirrors are operated by a set of electromagnets arranged to apply torque on the mirrors, and an orientation feedback system senses the orientations of the mirrors. Driving parameters for each mirror are determined based on information from the orientation feedback system. The driving parameters can be used to drive the mirrors in phase at an operating frequency despite variations in moments of inertia and resonant frequencies among the mirrors.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating object detection predictions from a neural network. In some implementations, an input characterizing a first region of an environment is obtained. The input includes a projected laser image generated from a three-dimensional laser sensor reading of the first region, a camera image patch generated from a camera image of the first region, and a feature vector of features characterizing the first region. The input is processed using a high precision object detection neural network to generate a respective object score for each object category in a first set of one or more object categories. Each object score represents a respective likelihood that an object belonging to the object category is located in the first region of the environment.

Abstract: Aspects of the disclosure relate to parking behaviors and maneuvering a vehicle in an autonomous driving mode accordingly. For instance, a pullover location for the vehicle to stop and wait for a passenger may be identified. The vehicle may be maneuvered in the autonomous driving mode in order to pull over by pulling forward into the pullover location. Whether to maneuver the vehicle in reverse in the pullover location before or after the passenger enters the vehicle may be determined based on context for the pull over with respect to the passenger. The vehicle may be maneuvered in the autonomous driving mode in reverse based on the determination of whether to maneuver the vehicle in reverse in the pullover location before or after the passenger enters the vehicle.

Abstract: Aspects of the present disclosure relate generally to limiting the use of an autonomous or semi-autonomous vehicle by particular occupants based on permission data. More specifically, permission data may include destinations, routes, and/or other information that is predefined or set by a third party. The vehicle may then access the permission data in order to transport the particular occupant to the predefined destination, for example, without deviation from the predefined route. The vehicle may drop the particular occupant off at the destination and may wait until the passenger is ready to move to another predefined destination. The permission data may be used to limit the ability of the particular occupant to change the route of the vehicle completely or by some maximum deviation value. For example, the vehicle may be able to deviate from the route up to a particular distance from or along the route.