Abstract: Aspects of the disclosure relate to detecting vehicle movement. For example, one or more computing devices may receive first image data representative of a vehicle's wheel and second image data representative of the wheel captured subsequent to the capture of the first image data. The one or more computing devices may determine a first location of a first portion of the wheel based on the first image data, and a second location of the first portion of the wheel based on the second image data. The one or more computing devices may calculate a value based on the angular distance between the first location and the second location of the first portion, and based on the value, determine whether the vehicle is in motion. Upon determining the vehicle is in motion the one or more computing devices may provide a signal that the vehicle is in motion.

Abstract: A system and method are arranged to provide recommendations to a user of a vehicle. In one aspect, the vehicle navigates in an autonomous mode and the sensors provide information that is based on the location of the vehicle and output from sensors directed to the environment surrounding the vehicle. In further aspects, both current and previous sensor data is used to make the recommendations, as well as data based on the sensors of other vehicles.

Abstract: The present disclosure relates to systems and methods that facilitate light detection and ranging operations. An example transmit block includes at least one substrate with a plurality of angled facets. The plurality of angled facets provides a corresponding plurality of elevation angles. A set of angle differences between adjacent elevation angles includes at least two different angle difference values. A plurality of light-emitter devices is configured to emit light into an environment along the plurality of elevation angles toward respective target locations so as to provide a desired resolution and/or a respective elevation angle. The present disclosure also relates to adjusting shot power and a shot schedule based on the desired resolution and/or a respective elevation angle.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating vehicle heading predictions from point cloud data using a neural network. One of the methods includes receiving a plurality of different projections of point cloud data, wherein the point cloud data represents different sensor measurements of electromagnetic radiation reflected off a vehicle. Each of the plurality of projections of point cloud data is provided as input to a neural network subsystem trained to receive projections of point cloud data for a vehicle and to generate one or more vehicle heading classifications as an output. At the output of the neural network subsystem, one or more vehicle heading predictions is received.

Abstract: The present disclosure provides methods and apparatuses that enable a radar system to transmit radar signals into lanes on a roadway in which a vehicle may turn. For example, when a car is making a protected right turn, that is a right turn when there is another vehicle traveling in the same direction in a lane adjacent to the lane of the turning vehicle, a traditional radar may have its view of the lane in which it is turning obscured by the vehicle in the lane adjacent to the lane of the turning vehicle. By using radar deflectors strategically located near the front of the vehicle, the radar signals may be deflected at angles to avoid being obstructed by the vehicle in the lane adjacent to the lane of the turning vehicle.

Abstract: The present disclosure is directed to an autonomous vehicle having a vehicle control system. The vehicle control system includes an image processing system. The image processing system receives an image that includes a plurality of image portions. The image processing system also calculates a score for each image portion. The score indicates a level of confidence that a given image portion represents an illuminated component of a traffic light. The image processing system further identifies one or more candidate portions from among the plurality of image portions. Additionally, the image processing system determines that a particular candidate portion represents an illuminated component of a traffic light using a classifier. Further, the image processing system provides instructions to control the autonomous vehicle based on the particular candidate portion representing an illuminated component of a traffic light.

Abstract: A computing device may be configured to receive sensor information indicative of respective characteristics of vehicles on a road of travel of a first vehicle. The computing device may be configured to identify, based on the respective characteristics, a second vehicle that exhibits an aggressive driving behavior manifested as an unsafe or unlawful driving action. Also, based on the respective characteristics, the computing device may be configured to determine a type of the second vehicle. The computing device may be configured to estimate a distance between the first vehicle and the second vehicle. The computing device may be configured to modify a control strategy of the first vehicle, based on the aggressive driving behavior of the second vehicle, the type of the second vehicle, and the distance between the first vehicle and the second vehicle; and control the first vehicle based on the modified control strategy.

Abstract: A vehicle configured to operate in an autonomous mode may operate a sensor to determine an environment of the vehicle. The sensor may be configured to obtain sensor data of a sensed portion of the environment. The sensed portion may be defined by a sensor parameter. Based on the environment of the vehicle, the vehicle may select at least one parameter value for the at least one sensor parameter such that the sensed portion of the environment corresponds to a region of interest. The vehicle may operate the sensor, using the selected at least one parameter value for the at least one sensor parameter, to obtain sensor data of the region of interest, and control the vehicle in the autonomous mode based on the sensor data of the region of interest.

Abstract: In some implementations, an image classification system of an autonomous or semi-autonomous vehicle is capable of improving multi-object classification by reducing repeated incorrect classification of objects that are considered rarely occurring objects. The system can include a common instance classifier that is trained to identify and recognize general objects (e.g., commonly occurring objects and rarely occurring objects) as belonging to specified object categories, and a rare instance classifier that is trained to compute one or more rarity scores representing likelihoods that an input image is correctly classified by the common instance classifier. The output of the rare instance classifier can be used to adjust the classification output of the common instance classifier such that the likelihood of input images being incorrectly classified is reduced.

Abstract: The technology relates to determining a future heading of an object. In order to do so, sensor data, including information identifying a bounding box representing an object in a vehicle's environment and locations of sensor data points corresponding to the object, may be received. Based on dimensions of the bounding box, an area corresponding to a wheel of the object may be identified. An orientation of the wheel may then be estimated based on the sensor data points having locations within the area. The estimation may then be used to determine a future heading of the object.

Abstract: A vehicle configured to operate in an autonomous mode can obtain sensor data from one or more sensors observing one or more aspects of an environment of the vehicle. At least one aspect of the environment of the vehicle that is not observed by the one or more sensors could be inferred based on the sensor data. The vehicle could be controlled in the autonomous mode based on the at least one inferred aspect of the environment of the vehicle.

Abstract: Aspects of the present disclosure relate to differentiating between active and inactive construction zones. In one example, this may include identifying a construction object associated with a construction zone. The identified construction object may be used to map the area of the construction zone. Detailed map information may then be used to classify the activity of the construction zone. The area of the construction zone and the classification may be added to the detailed map information. Subsequent to adding the construction zone and the classification to the detailed map information, the construction object (or another construction object) may be identified. The location of the construction object may be used to identify the construction zone and classification from the detailed map information. The classification of the classification may be used to operate a vehicle having an autonomous mode.

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: The technology relates to detecting and responding to emergency vehicles. This may include using a plurality of microphones to detect a siren noise corresponding to an emergency vehicle and to estimate a bearing of the emergency vehicle. This estimated bearing is compared to map information to identify a portion of roadway on which the emergency vehicle is traveling. In addition, information identifying a set of objects in the vehicle's environment as well as characteristics of those objects is received from a perception system is used to determine whether one of the set of objects corresponds to the emergency vehicle. How to respond to the emergency vehicle is determined based on the estimated bearing and identified road segments and the determination of whether one of the set of objects corresponds to the emergency vehicle. This determined response is then used to control the vehicle in an autonomous driving mode.

Abstract: The technology relates to a system for clearing sensors. The system may comprise a sensor. The sensor may comprise a sensor cover and a cover window arranged on the sensor cover. The system may further comprise a sensor motor for rotating the sensor cover and the cover window around an axis of rotation. The sensor cover and cover window may be rotated to clear the cover window of debris. The cover window may have a linear cross section in a plane that passes through and is perpendicular to the axis of rotation. The sensor motor may be configured to rotate the sensor cover and the cover window.

Abstract: Aspects of the disclosure relate to detecting vehicle collisions. In one example, one or more computing devices may receive acceleration data of a vehicle and the expected acceleration data of the vehicle over a period of time. The one or more computing devices may determine a change in the vehicle's acceleration over the period of time, where the change in the vehicle's acceleration over the period of time is the difference between the expected acceleration data and the acceleration data. The one or more computing devices may detect an occurrence when the change in the vehicle's acceleration is greater than a threshold value and assign the occurrence into a collision category. Based on the assigned collision category, the one or more computing devices may perform a responsive action.

Abstract: The present disclosure relates to optical systems and methods for their manufacture. An example system includes a first substrate that has at least one bridge structure. The bridge structure has a first surface with one or more light-emitter devices disposed on it. The system also includes a second substrate that has a mounting surface that defines a reference plane. The second substrate includes a structure and an optical spacer on the mounting surface. The first and second substrates are coupled together such that the first surface of the first substrate faces the second substrate at an angle with respect to the reference plane. The system also includes at least one spacer coupled to the mounting surface. The at least one spacer is in physical contact with the one or more light-emitter devices.

Abstract: Aspects of the present disclosure relate switching between autonomous and manual driving modes. In order to do so, the vehicle's computer may conduct a series of environmental, system, and driver checks to identify certain conditions. The computer may correct some of these conditions and also provide a driver with a checklist of tasks for completion. Once the tasks have been completed and the conditions are changed, the computer may allow the driver to switch from the manual to the autonomous driving mode. The computer may also make a determination, under certain conditions, that it would be detrimental to the driver's safety or comfort to make a switch from the autonomous driving mode to the manual driving mode.

Abstract: One example method involves obtaining a plurality of images using a camera located at a given position relative to a light detection and ranging (LIDAR) device. A first image of the plurality may be obtained while a first aperture is interposed between the camera and the LIDAR device. A second image of the plurality may be obtained while a second aperture is interposed between the camera and the LIDAR device. The method also involves determining one or more alignment offsets between a transmitter of the LIDAR device and a receiver of the LIDAR device based on the plurality of images.

Abstract: A system and method for avoiding or mitigating collisions by an autonomous vehicle is provided. Raw sensor data for the vehicle's environment is received, such as through a perception system of the vehicle. A trajectory is also received, the trajectory having a width corresponding to a distance that is at least as wide as the vehicle. The trajectory and raw sensor data are projected onto a grid, such as a static occupancy grid, having a plurality of cells. It is determined whether any cell that overlaps with the trajectory includes the raw sensor data. It is further determined, based on the raw sensor data, whether an obstacle is present along the trajectory. When no obstacles are determined to be present along the trajectory, the trajectory may be verified. However, where obstacles are determined to be present, the vehicle may take an action to avoid collision.