Abstract: A method of lane detection for a non-transitory computer readable storage medium storing one or more programs is disclosed. The one or more programs include instructions, which when executed by a computing device, cause the computing device to perform the following steps comprising: generating a ground truth associated with lane markings expressed in god's view; receiving features from at least one of a hit-map image and a fitted lane marking, wherein the hit-map image includes a classification of pixels that hit a lane marking, and the fitted lane marking includes pixels optimized based on the hit-map image; and training a confidence module based on the features and the ground truth, the confidence module configured to determine on-line whether a fitted lane marking is reasonable, using parameters that express a lane marking in an arc.

Abstract: A telescopic sensor system for an autonomous vehicle enables sensors located on movable telescopic apparatuses to obtain sensor data when an object obstructs an area scanned by fixed sensors. An example method of controlling a movable telescopic apparatus on an autonomous vehicle includes obtaining, from a first sensor located on the autonomous vehicle, a first sensor data of a first area relative to a location of the autonomous vehicle, performing, from the first sensor data, a first determination that a view of the first area is obstructed, causing, in response to the first determination, a second sensor coupled to the movable telescopic apparatus to extend to a pre-determined position, and obtaining, from the second sensor, a second sensor data of a second area relative to the location of the autonomous vehicle, where the second area includes at least some of the first area.

Abstract: Devices, systems, and methods for a vehicular safety system in autonomous vehicles are described. An example method for safely controlling a vehicle includes selecting, based on a first control command from a first vehicle control unit, an operating mode of the vehicle, and transmitting, based on the selecting, the operating mode to an autonomous driving system, wherein the first control command is generated based on input from a first plurality of sensors, and wherein the operating mode corresponds to one of (a) a default operating mode, (b) a minimal risk condition mode of a first type that configures the vehicle to pull over to a nearest pre-designated safety location, (c) a minimal risk condition mode of a second type that configures the vehicle to immediately stop in a current lane, or (d) a minimal risk condition mode of a third type that configures the vehicle to come to a gentle stop.

Abstract: A vehicle position and velocity estimation based on camera and LIDAR data are disclosed.

Abstract: A system and method for implementing a neural network based vehicle dynamics model are disclosed. A particular embodiment includes: training a machine learning system with a training dataset corresponding to a desired autonomous vehicle simulation environment; receiving vehicle control command data and vehicle status data, the vehicle control command data not including vehicle component types or characteristics of a specific vehicle; by use of the trained machine learning system, the vehicle control command data, and vehicle status data, generating simulated vehicle dynamics data including predicted vehicle acceleration data; providing the simulated vehicle dynamics data to an autonomous vehicle simulation system implementing the autonomous vehicle simulation environment; and using data produced by the autonomous vehicle simulation system to modify the vehicle status data for a subsequent iteration.

Abstract: Disclosed are devices, systems and methods for using a rotating camera for vehicular operation. One example of a method for improving driving includes determining, by a processor in the vehicle, that a trigger has activated, orienting, based on the determining, a single rotating camera towards a direction of interest, and activating a recording functionality of the single rotating camera, where the vehicle comprises the single rotating camera and one or more fixed cameras, and where the single rotating camera provides a redundant functionality for, and consumes less power than, the one or more fixed cameras.

Abstract: Disclosed are devices, systems and methods for capturing an image. In one aspect an electronic camera apparatus includes an image sensor with a plurality of pixel regions. The apparatus further includes an exposure controller. The exposure controller determines, for each of the plurality of pixel regions, a corresponding exposure duration and a corresponding exposure start time. Each pixel region begins to integrate incident light starting at the corresponding exposure start time and continues to integrate light for the corresponding exposure duration. In some example embodiments, at least two of the corresponding exposure durations or at least two of the corresponding exposure start times are different in the image.

Abstract: Techniques are described for determining visibility conditions of an environment in which an autonomous vehicle is operated and performing driving related operations based on the visibility conditions. An example method of adjusting driving related operations of a vehicle includes determining, by a computer located in an autonomous vehicle, a visibility related condition of an environment in which the autonomous vehicle is operating, adjusting, based at least on the visibility related condition, a set of one or more values of one or more variables associated with a driving related operation of the autonomous vehicle, and causing the autonomous vehicle to be driven to a destination by causing the driving related operation of one or more devices located in the autonomous vehicle based on at least the set of one or more values.

Abstract: A system and method for vehicle wheel detection is disclosed. A particular embodiment can be configured to: receive training image data from a training image data collection system; obtain ground truth data corresponding to the training image data; perform a training phase to train one or more classifiers for processing images of the training image data to detect vehicle wheel objects in the images of the training image data; receive operational image data from an image data collection system associated with an autonomous vehicle; and perform an operational phase including applying the trained one or more classifiers to extract vehicle wheel objects from the operational image data and produce vehicle wheel object data.

Abstract: A system and method for providing multiple agents for decision making, trajectory planning, and control for autonomous vehicles are disclosed. A particular embodiment includes: partitioning a multiple agent autonomous vehicle control module for an autonomous vehicle into a plurality of subsystem agents, the plurality of subsystem agents including a deep computing vehicle control subsystem and a fast response vehicle control subsystem; receiving a task request from a vehicle subsystem; dispatching the task request to the deep computing vehicle control subsystem or the fast response vehicle control subsystem based on the content of the task request or a context of the autonomous vehicle; causing execution of the deep computing vehicle control subsystem or the fast response vehicle control subsystem by use of a data processor to produce a vehicle control output; and providing the vehicle control output to a vehicle control subsystem of the autonomous vehicle.

Abstract: Disclosed are devices, systems and methods for curb detection using light detection and ranging (LiDAR) sensors. One example of a method for curb detection includes acquiring, based on a scan of an area around a vehicle, a point-cloud frame that includes a description of an intensity of a reflection of the area around the vehicle, selecting a subset of points from the point-cloud frame, seeding each of a plurality of clusters based on the selected points, determining a criterion for each of the selected points, increasing a size of each of the plurality of clusters by including additional points from the point-cloud frame that meet the criterion for the corresponding cluster, identifying a largest cluster of the plurality of clusters upon completion of the increasing step, and detecting a curb based on a boundary of the largest cluster.

Abstract: Disclosed is a camera mounting apparatus for a vehicle that attaches the camera to the vehicle, isolates the camera from vibration at the mounting points, and accommodates thermal expansion and contraction of the vehicle and the mounting structure. The apparatus includes a camera mounting beam coupled to the camera, and one or more elastomeric vibration isolators coupled to an attachment device and coupled to the camera mounting beam. The apparatus also includes a spherical bearing coupled to the attachment device, wherein the spherical bearing is configured to accommodate misalignment of the attachment device, and one or more spring washers to generate a force holding the one or more elastomeric vibration isolators and spherical bearing in positions.

Abstract: Image processing techniques are described to receive bounding box information that describes a bounding box located around a detected object in an image, determine one or more positions of one or more reference points on the bounding box, determine, for each reference point, 3D world coordinates of a point of intersection of the reference point and the road surface, and assign the 3D world coordinates of the one or more reference points to a location of the detected object.

Abstract: A lead autonomous vehicle (AV) includes a sensor configured to observe a field of view in front of the lead AV. Following AVs are on the same road behind the lead AV. A processor of the lead AV is configured to detect a road closure. The processor overrides driving instructions of the lead AV, such that the lead AV is stopped at first location coordinates. The processor sends a first message to an operation server, indicating that the road closure is detected. The operation server update the first portion of the map data, reflecting the road closure. The operation server determines whether re-routing is possible for each AV. If re-routing is possible is possible for that AV, the operation server sends re-routing instructions to the AV. If re-routing is not possible is possible for that AV, the operation server sends pulling over instructions to the AV.

Abstract: A system includes an autonomous vehicle (AV) comprising a sensor, a control subsystem, and an operation server. The control subsystem receives sensor data comprising location coordinates of the AV from the sensor. The operation server detects an unexpected event from the sensor data, comprising at least one of an accident, an inspection, and a report request. The operation server receives a message from a user comprising a request to access particular information regarding the AV and location data. The operation server associates the AV with the user if the location coordinates of the AV match location data of the user. The operation server generates a ticket for the unexpected ticket. The operation server communicates the particular information to the user. The operation server provides instructions to be forwarded to the user based on a request from the user. The operation server closes the ticket if the unexpected event is addressed.

Abstract: Techniques are described for determining weight distribution of a vehicle. A method of performing autonomous driving operation includes receiving two sets of values from two sets of sensors, where a first set of sensors measure weights or pressures applied on axles of a vehicle, and where a second set of sensors measure pressures in tires of the vehicle. The method performs an error detection and removal operation to remove or filter out any erroneous values from the two sets of values to obtain two sets of filtered values. The method determines one or more values that describe a weight or pressure applied on the axle to obtain the weight distribution of the vehicle based on the first set of filtered values or the second set of filtered values. Based on the obtained weight distribution of the vehicle, the method can determine a driving operation of the vehicle.

Abstract: A system and method for real world autonomous vehicle trajectory simulation may include: receiving training data from a data collection system; obtaining ground truth data corresponding to the training data; performing a training phase to train a plurality of trajectory prediction models; and performing a simulation or operational phase to generate a vicinal scenario for each simulated vehicle in an iteration of a simulation. Vicinal scenarios may correspond to different locations, traffic patterns, or environmental conditions being simulated. Vehicle intention data corresponding to a data representation of various types of simulated vehicle or driver intentions.

Abstract: An autonomous vehicle (AV) includes at least one vehicle sensor that is configured, when the AV is stopped, to observe at least a first portion of a zone around the stopped AV. A portable device is operated by a user near the stopped AV. A control subsystem receives a communication from the portable device that includes information regarding whether a second portion of the zone around the AV is free of obstructions. The control subsystem receives AV sensor data from the at least one vehicle sensor. The control subsystem determines whether the first portion of the zone free of obstructions and whether the second portion of the zone is free of obstructions. If that both the first and second portions of the zone around the stopped AV are free of obstructions, the stopped AV is allowed to begin moving.