Abstract: Techniques are provided for extended object tracking using RADAR return points only. The techniques include receiving the return points from at least one RADAR sensor of the vehicle. One or more clusters based on or from the return points are generated. An estimated position and velocity of each of the one or more clusters is computed. A Recursive Least Squares (RLS) based algorithm is proposed to estimate the instantaneous velocity of a cluster in real-time that allows for accurate track-cluster association, removes the need to perform computationally expensive non-linear state updates, and allows for the estimation of the true velocity even in frames with a large amount of clutter. If it is determined that the one or more clusters are associated with an existing object track, the existing object track is updated using at least the respective positions of the one or more clusters associated with the existing object track.

Abstract: An object-detection system for an automated vehicle includes an object-detector, a receiver, and a controller. The object-detector detects detectable-objects proximate to a host-vehicle. The receiver receives an indication of an object-presence from other-transmitters proximate to the host-vehicle. The controller is in communication with the object-detector and the receiver. The controller is configured to operate the host-vehicle to avoid interference with a hidden-object when the hidden-object is not detected by the object-detector and the object-presence is indicated by at least two instances of the other-transmitters.

Abstract: Methods, apparatus, and systems for passenger authentication and entry for autonomous vehicles are disclosed. A vehicle receives information over a first communication channel from a first mobile device. The information specifies a geographical location. The information causes the vehicle to arrive at the geographical location. The vehicle transmits a first message to a second mobile device indicating that the vehicle has arrived at the geographical location. The first message includes a graphical icon representing the vehicle. The graphical icon is for display on a user interface of the second mobile device. The vehicle receives a second message from the second mobile device over a second communication channel. The second message indicates that the graphical icon was dragged across the user interface into a graphical receptacle displayed on the user interface. Responsive to receiving the second message, at least one door of the vehicle is unlocked.

Abstract: Among other things, techniques are described for forecasting occupancy of a vehicle location. This includes receiving, by at least one processor, status information of a parking location the status information representing an availability of the parking location; predicting, by the at least one processor, a future status of the parking location based on the received status information; determining, by the at least one processor, a destination based on the predicted future status of the parking location; and providing, by the at least one processor, the predicted future status to a controller of a vehicle for controlling the vehicle to drive to the destination.

Abstract: Techniques are provided for generating a driving trajectory for a vehicle while the vehicle is blocked (e.g., the sensor of the vehicle have detected an object results in the vehicle being unable to move) by an object (e.g., another vehicle, bicycle, or a pedestrian) and executing the driving trajectory when the vehicle becomes unblocked. In addition, techniques are provided for updating a portion of a driving trajectory of a vehicle based on a determination that an object will cross a segment of the current driving trajectory at a later point, without recalculating the whole trajectory.

Abstract: Among other things, techniques are described for monitoring and controlling autonomous vehicles (AVs). As an example, at least one processor determines that a first AV is in a field of view of a user wearing an augmented reality display device, determines first data regarding an operation of the first AV, and causes at least a portion of the first data to be presented to the user using the augmented reality display device. For example, a graphical user interface is presented in the field of view of the user using the augmented reality display device, and at least the portion of the first data is included on the graphical user interface, such that at least the portion of the first data appears to be arranged spatially proximal to the first AV in the field of view of the user.

Abstract: Among other things, we describe techniques for automatic annotation of environmental features in a map during navigation of a vehicle. The techniques include receiving, by the vehicle located within an environment, a map of the environment. Sensors of the vehicle receive sensor data and semantic data. The sensor data includes a plurality of features of the environment. A geometric model of a feature of the plurality of features is generated. The feature is associated with a drivable area within the environment. A drivable segment is extracted from the drivable area. The drivable segment is segregated into a plurality of geometric blocks, wherein each geometric block corresponds to a characteristic of the drivable area and the geometric model of the feature includes the plurality of geometric blocks. The geometric model is annotated using the semantic data. The annotated geometric model is embedded within the map.

Abstract: Among other things, a vehicle drives autonomously on a trajectory through a road network to a goal location based on an automatic process for planning the trajectory without human intervention; and an automatic process alters the planning of the trajectory to reach a target location based on a request received from an occupant of the vehicle to engage in a speed-reducing maneuver.

Abstract: Enclosed are embodiments for scoring one or more trajectories of a vehicle through a given traffic scenario using a machine learning model that predicts reasonableness scores for the trajectories. In an embodiment, human annotators, referred to as a “reasonable crowd,” are presented with renderings of two or more vehicle trajectories traversing through the same or different traffic scenarios. The annotators are asked to indicate their preference for one trajectory over the other(s). Inputs collected from the human annotators are used to train the machine learning model to predict reasonableness scores for one or more trajectories for a given traffic scenario. These predicted trajectories can be used to rank trajectories generated by a route planner based on their scores, compare AV software stacks, or used by any other application that could benefit from a machine learning model that scores vehicle trajectories.

Abstract: This disclosure describes a cyber-security protocol for validating messages being exchanged between two devices of an autonomous vehicle. The protocol includes the independent generation of multiple encryption or session keys by both devices. The encryption keys are generated based on a random number provided by each device. In some embodiments, the random numbers can be accompanied by a shared secret key installed on both devices that can help prevent an unauthorized device from creating a shared set of encryption keys with one of the devices. Including a hash generated using one of the encryption keys and a message sequence counter value in each message can help prevent the injection of previously transmitted messages as a means of disturbing operation of the autonomous vehicle.

Abstract: Provided are methods for prediction error scenario mining for machine learning methods, which can include determining a prediction error indicative of a difference between a planned decision of an autonomous vehicle and an ideal decision of the autonomous vehicle. The prediction error is associated with an error-prone scenario for which a machine learning model of an autonomous vehicle is to make planned movements. The method includes searching a scenario database for the error-prone scenario based on the prediction error. The scenario database includes a plurality of datasets representative of data received from an autonomous vehicle sensor system in which the plurality of datasets is marked with at least one attribute of the set of attributes. The method further includes obtaining the error-prone scenario from the scenario database for inputting into the machine learning model for training the machine learning model. Systems and computer program products are also provided.

Abstract: Among other things, techniques are described for expressive vehicle systems. These techniques may include obtaining, with at least one processor, data associated with an environment, the environment comprising a vehicle and at least one object; determining an expressive maneuver including a deceleration of the vehicle such that the vehicle stops at least a first distance away from the at least one object and the vehicle reaches a peak deceleration when the vehicle is a second distance away from the at least one object; generating data associated with control of the vehicle based on the deceleration associated with the expressive maneuver; and transmitting the data associated with the control of the vehicle to cause the vehicle to decelerate based on the deceleration associated with the expressive maneuver.

Abstract: Among other things, techniques are described for identifying, in a light detection and ranging (LiDAR) scan line, a first LiDAR data point and a plurality of LiDAR data points within a vicinity of the first LiDAR data point. The techniques may further include identifying, based on a comparison of the first LiDAR data point to at least one LiDAR data point of the plurality of LiDAR return points, a coefficient of the first LiDAR data point, wherein the coefficient is related to image smoothness. The techniques may further include identifying, based on a comparison of the coefficient to a threshold, whether to include the first LiDAR data point in an updated LiDAR scan line, and then identifying, based on the updated LiDAR scan line, a location of the autonomous vehicle.

Abstract: Provided are methods for activating haptic feedback sensors to identify a state of vehicle, which can include receiving sensor data associated with the vehicle, determining a vehicle alert for a passenger of the vehicle based at least in part on the sensor data associated with the vehicle, and activating at least one haptic feedback sensor based at least in part on the vehicle alert for the passenger. Systems and computer program products are also provided.

Abstract: An object-detection system for an automated vehicle includes an object-detector, a receiver, and a controller. The object-detector detects detectable-objects proximate to a host-vehicle. The receiver receives an indication of an object-presence from other-transmitters proximate to the host-vehicle. The controller is in communication with the object-detector and the receiver. The controller is configured to operate the host-vehicle to avoid interference with a hidden-object when the hidden-object is not detected by the object-detector and the object-presence is indicated by at least two instances of the other-transmitters.

Abstract: Among other things, techniques are described for monitoring and controlling autonomous vehicles (AVs). As an example, at least one processor determines that a first AV is in a field of view of a user wearing an augmented reality display device, determines first data regarding an operation of the first AV, and causes at least a portion of the first data to be presented to the user using the augmented reality display device. For example, a graphical user interface is presented in the field of view of the user using the augmented reality display device, and at least the portion of the first data is included on the graphical user interface, such that at least the portion of the first data appears to be arranged spatially proximal to the first AV in the field of view of the user.

Abstract: Among other things, techniques are described for conditional motion predictions. The techniques include generating, by a planning circuit, a set of candidate trajectories for a vehicle based on possible macro actions by the vehicle, predicting, by the planning circuit and for at least some trajectories in the set of candidate trajectories, a response by a target vehicle to the respective trajectory and a probability of the response by the target vehicle, selecting, by the planning circuit, a trajectory from the set of candidate trajectories based at least in part on the response by the target vehicle to the trajectory, the probability of the response by the target vehicle, and characteristics of the trajectory, and operating, by a control circuit, the vehicle based on the selected trajectory.

Abstract: Among other things, techniques are described for implementing occlusion representations over road features. In general, one innovative aspect of the subject matter described in this specification can be embodied in methods that include: obtaining area information of at least one area of interest for a vehicle, identifying occlusion data associated with the at least one area of interest, determining occlusion information associated with the at least one area of interest based on the area information and the occlusion data, providing the occlusion information for planning a route for the vehicle, and operating the vehicle in accordance with the planned route. The occlusion data includes data associated with the occlusion information, and the occlusion information has a smaller data size than the occlusion data.

Abstract: Among other things, one or more travel signals are identified by analyzing one or more images and data from sensors, classifying candidate travel signals into zero, one or more true and relevant travel signals, and estimating a signal state of the classified travel signals.

Abstract: Provided are methods for efficient and optimal feature extraction, which can include identifying observations that satisfy a predetermined configuration of metrics. Some methods described also include extracting feature values of a respective metric from the identified observations and determining parameters for feature extraction based on the feature values extracted from the identified observations and a respective label of the identified observations. Systems and computer program products are also provided.